Datasets:

ianiket23

/

podcast_llama_chat_format-5k

like 0

<s>[INST] So, if you could, philosophy is a little bit within this empirical question at where do you think would enter the by which mechanism would enter the possibility that we can't predict the outcome. So there would be something that would be akin to a free will. [/INST] Well, you could say the the sort of obvious possibility which was, you know, recognized by uh Addington and many others about as soon as quantum mechanics was discovered in the 19 twenties, was that uh if um, you know, let's say, a sodium ion channel, you know, in the in the in the brain, right? You know, it's it's behavior is chaotic, right? It it's sort of it's governed by these hodge hodge, we huck skin equations in neuroscience, right. Which are differential equations that have a stochastic component right now? Where does you know? And this ultimately governs, let's say whether a neuron will fire or not fire the basic </s><s>[INST] chemical process or electrical process by which signals are sent in the [/INST] brain? Exactly, Exactly. And and uh, you know, and so you could ask, well, where does the randomness in the process, you know, that uh that that neuroscientists or what neuroscientists would would would treat is randomness. Where does it come from? You know, ultimately it's thermal noise, right? Where it is thermal noise come from, But ultimately, you know, there are some quantum mechanical events at the molecular level that are getting sort of chaotically amplified by, you know, a sort of butterfly effect. Uh and so uh you know, even if you knew the complete quantum state of someone's brain at best, you could predict the probabilities that they would do one thing or do another thing. Right? I think that part is actually relatively uncontroversial, right? The controversial question is whether any of it matters for this sort of philosophical questions that we care about, because you could say if all it's doing is just injecting some randomness into an otherwise completely mechanistic process. Well then who cares? Right And more concretely, if you could build a machine that, you know, could just calculate the even just up the the probabilities of all the possible things that you would do. Right? And you know, um um You know, of all the things that said you had a 10% chance of doing, you did exactly 1/10 of them, you know? And and and and and and so on. That </s><s>[INST] somehow also takes away the feeling of freedom. [/INST] Exactly. I mean, I mean, to me it seems essentially just as bad as if the machine deterministic lee predicted you? It seems, you know, hardly different from that. So, so that so then, um uh but a more a more subtle question is, could you even learn enough about someone's brain to do that? Okay, because, you know, another central fact about quantum mechanics is that uh making a measurement on a quantum state is an inherently destructive operation. Okay, so, uh you know, if I want to measure the position of a particle, right? It was well before I measured, it had a superposition over many different positions as soon as I measure I localize it. Right? So now I know the position, but I've also fundamentally changed the state. And so so you could say, well, maybe in in trying to build a model of someone's brain that was accurate enough to actually, you know, make, let's say even even well calibrated probabilistic predictions of their future behavior. Maybe you would have to make measurements that were just so accurate that you would just fundamentally alter their brain. Okay. Or or or or maybe not, maybe you only, you know, you it would suffice to just make some nano robots that just measured some sort of a much larger scale. You know, macroscopic behavior, like, you know, is, you know, what is this neuron doing? What is that neuron doing? Maybe that would be enough. See, but now, you know, I I what I what I claim is that we're now asking a question, you know, in which, you know, it is it is it is possible to envision what progress on it would look like. </s><s>[INST] Yeah. But just as you said that question, maybe slightly detached from the philosophical question in a sense, if consciousness somehow has a role to the experience of free will, because ultimately, when we're talking about free will were also talking about not just the predictability of our actions, but somehow the experience of that predictability. [/INST] Yeah. Well, I mean, a lot of philosophical questions ultimately, like feedback to the hard problem of consciousness, you know, and as much as you can try to sort of talk around it or not, right? And, you know, And then and there is a reason why people try to talk around it, which is that, you know, a democritus talked about the hard problem of consciousness, You know, in 400 BC in terms that would be totally recognizable to us today. Right? And it's really not clear if there's been progress since or what progress could possibly consist of. Is </s><s>[INST] there a cute prime type of sub question that could help us get it consciousness. It's something about [/INST] consciousness. Well, I mean, well, I mean, there is the whole question of, you know, of a I write of, you know, can you build a human level or super human level ai and uh you know, can it can it work in a completely different substrate from the brain. I mean, you know, of course that was alan turing's point. And you know, and and even if that was done, it's, you know, maybe people would still argue about the hard problem of consciousness, right? And yet, you know, my my my claim is a little different. My claim is that in a world where, you know, there were, you know human level ai is where we had been even overtaken by such a eyes. The entire discussion of the hard problem of consciousness would have a different character, right? It would take place in different terms in such a world. Even if we hadn't answered the question and and my claim about free Will would be similar, right? If there if this prediction machine that I was talking about could actually be built. Well now the entire discussion of the you know, a free will is sort of transformed by that. Um You know, even if in some sense the metaphysical question hasn't been answered. </s>

<s>[INST] So then forgive me for romanticizing it, but it does seem pretty magical that You know that I kind of have an intuitive understanding of what happens when you have like 5, 10, maybe even 100 people together, the dynamics of that. But the fact that these large society of people for the most part acting in a self interested way and maybe electing representatives for themselves, that it all kind of seems to work. It's pretty magical. The fact that [/INST] there's uh you know that right now there's uh uh a wide assortment of fresh fruit and vegetables uh at the uh in the local markets up and down the street. Uh you know who's who's planning that? And the answer is nobody that's the that's the invisible hand at work and that's great. And um And that's a lesson that's that Adam Smith figured out more than 200 years ago and um it's it continues to apply and the the but you know, even Adam smith has a section of the book about why it's important to regulate banks. So the invisible hand has its limits. </s><s>[INST] Yeah. And that example is actually a powerful one in terms of the supermarket and fruit. That was my experience coming from Russia from the soviet union is when I first entered the supermarket and just seeing the assortment of fruit bananas. So I don't think I've seen bananas before, first of all, but just the selection of fresh fruit was just mind blowing and it's beyond words. And the fact that like like you said, I don't know what made that [/INST] happen. Well that there is some magic to the market, but the uh as as uh showing my age. But you know, the old movie quotes sometimes the magic works and sometimes it doesn't and you have to have some idea of when it doesn't. </s><s>[INST] So how do you get regulation right? How what can government at its best to government, strangely enough in this country today uh seems to get a bad rap. Like everyone seems to everybody is against the government. [/INST] Yeah, Well, a lot of money has been spent on making people hate the government. But the reality is government does some things pretty well. I mean we uh uh government does health insurance pretty well so much. So you've given our anti government bias there. It really is true that there are people out there saying don't let the government get its hands on Medicare. Um So government people actually love the government health insurance program far more than they love private health insurance. Um uh Basic education. Uh It turns out that uh your your your local public high school is the right place to have students trained and and private for certainly for profit education is a is a by and large a nightmare of of rip offs and and and Grift and people uh uh not getting what they thought they were paying for. Um It's a judgment case and it's funny, there are things, I mean, </s><s>[INST] everybody talks, [/INST] there's talks about the D. M. V. As being, you know, how do you want the economy? Actually, my experience is that the D. M. V. Have always been positive. Maybe I'm just going to the right D. M. V. S. But in fact that a lot of government works pretty well. So it you have to to some extent you can do these things on a priori grounds. You can talk about the logic of why health care is not going to be handled well by the market, but partly it's just experienced. We tried, or at least some countries have tried nationalizing their steel industries that didn't go well, but we've tried privatizing education and that didn't go well. So you you find out what works. </s><s>[INST] What about this new world of tech, How do you see, what do you think works for tech? Is that more regulation or less regulation? [/INST] There are some things that need more regulation. And we're finding out that uh the world of social media is is one in which competitive competitive forces aren't working very well and trusting the companies to regulate themselves isn't working very well. Um but I'm on the whole a tech skeptic, not in the sense that I think the tech doesn't work and it doesn't do stuff, but the idea that we're living through greater technological change than ever before is really an illusion. Uh we've ever since the beginning of the Industrial Revolution, we've had a series of epochal shifts in the nature of work and and the kinds of jobs that are available and it's not at all clear that what we're what's happening now is any bigger or faster or harder to cope with than past shocks. </s><s>[INST] It is a popular notion in today's sort of public discourse that automation is going to have a huge impact on job market now. Yeah, there is something something transformational happening now. He talked about that maybe elaborating a little bit more. Do you not see the software revolution is happening now with with machine learning, availability of data that kind of automation being able to sort of process clean, find patterns and data and you know, you don't see that disrupting anyone sector to a point where there's a huge loss of jobs, there [/INST] may be some things, I mean actually, you know, translators, there's really reduced demand for translators because we've machine translation ain't perfect, but it ain't bad um there are some kinds of things that are changed but uh uh it's not an overall productivity growth has actually been slow in recent years. Now it's been much slower than than in some past periods. So the idea that automation is taking away all the jobs, the counterpart would be that we would be able to produce stuff with many fewer workers than before and that's not happening. There are a few isolated sectors, there are some kinds of uh of jobs that are going away but that keeps on happening. I mean they uh new york city used to have thousands and thousands of longshoremen taking stuff off off ships and putting them on ships. Uh There are almost all gone now. Now you have the giant cranes taking containers on and off ships in Elizabeth, New Jersey. Um That's not robots, it doesn't doesn't sound high tech, but it actually pretty much destroyed an occupation. Um Well you know that was it wasn't fun for the longshoremen to say the least. And but it's not uh we coped, we moved on and that that sort of thing happens all the time. You mean farmers, we we used to be a nation which was mostly farmers. There are now very few farmers left the end. The reason is not that uh we've stopped eating, it's that farming has become so efficient that we don't need a lot of farmers and we coped with that too. So the idea that there's something uh qualitatively different about what's happening now so far isn't true. </s>

<s>[INST] well. And if we can't ask them. So you mean map like is there which [/INST] as you know, I have to would have to sacrifice them and take tissue and then bring it in the lab do genome sequence. </s><s>[INST] So you can do a thorough sequencing to determine with viruses. Is there a sufficiently good categorization of viruses where [/INST] you be? That's a very good question. So whenever you do sequence, right you get some environmental sample and you extract nucleic acid in your sequence is what you do is you run it past the database with the gold standard is the gen bank database which is maintained here in the US. And you see if you get any hits and then you can say ah look this this sequence is similar to this virus and you can classify all the viruses you see. The problem is 90% of your sequence is dark matter. It doesn't hit with anything. It's probably a lot of it is unknown viruses and that's going to be hard to figure out because someone's going to have to go after it and sorted through. So yes you can find a lot of viruses and the numbers you get are astounding. You can find thousands of new viruses just by looking in various life forms. But there are many more that we don't pick up because they're not in the database. </s><s>[INST] Maybe this is a good time to take a quick tangent. What do you think about alpha fold to? I don't know if you've been paying attention to that with them. Deepmind solving the protein folding problem and then also releasing First of all open source in the code which is for me as a software person I love. And then second of all also making like 300,000 predictions or something like that for different protein structures and releasing that data. UU So on the side of because you make you're saying there's dark matter, is there something? Um Well, first, what are your general thoughts level of excitement about their uh their work? And second, how can that be applied to viruses? Do you think we'll be able to explore the dark matter of virology using machine learning? [/INST] Because in all this dark sequence you can translated and make a protein, you can see what a protein looks like. It has what we call an open reading frame, right? Starting to stop and right now it's just a bunch of amino acids. But if we could fold it may be the fold would record would be like something. We already know some protein fold which gives you a lot of clues because there are only so many protein folds in biology. And that dark matter is probably one of them. So I think that's very exciting because for years I followed structural biologists for years. And you know in the beginning we couldn't even solve structures of viruses are too big. We could do small molecules like myoglobin. Like that was the first one done took years to do that. And then as computational power increase, then they could start to do viruses but it took a long time X ray crystallography dependent on getting crystals of the virus. Right? And now we can do cryo electron microscopy, which is much faster. Uh you could solve the spike of SARS-COV-2 was solved in two months by by Jason mcclelland here in Austin actually at the beginning of the pandemic. But you're limited. You can't do huge proteins. You can only do moderately sized ones. So or actually you can do viruses but you can't do small proteins So that speed it up. But it's still too fast to solve yet a new protein. You want to solve its structure. So if we could predict it and I know from talking to structural biologist has been their holy grail from day one, they want to be able to take a sequence of a protein, put it in a computer and have the structure put out without having to do all the experiments. So that's why this is very exciting that you can predict it. I mean it's not finished obviously and there's more to do. But I think it will be a day where you could take any amino acid sequence and predict what it's gonna look like </s><s>[INST] me. But like aren't structural biologist gonna get greedy. So once you have that, don't you want to go more complicated than don't you want to go? Because that's that's just the first step right to go from, you know, asset to the structure and there's like multiple protein interactions. Like how do you get to the virus. [/INST] Well, so that's what the ultimate goal of getting a structure is that. Then you can do experiments and figure out what the structure means. Right. So many in the old days, structural biology was a career in itself, you worked with people who had a system and just solve proteins for them. And then you moved on to another one. You didn't really do any experiments. The other people got to do all the interesting experiment. Now young structural biologists or multifaceted, they saw the structure and then they say what happens if we change this amino acid? Oh look it it blocks binding to the receptor, this must be the receptor binding interface. So that's the exciting stuff. Absolutely is doing the </s>

<s>[INST] Can you speak to that? So what do you think about Bitcoin? What do you think it is a response to? What are the growing problems of the fiat system? What is this moment in human history that is full of challenges that Bitcoin and cryptocurrencies trying to overcome. [/INST] I don't think Bitcoin was devised by Satoshi whoever he was for fear of a breakdown of the fiat currencies. If it was, it was a very farsighted enterprise because certainly in 2008 when the first Bitcoin paper appeared, it wasn't very likely that a wave of inflation was coming. If anything, there was more reason to fear deflation at that point. I think it would be more accurate to say that with the advent of the internet there was a need for a means of payment. Native to the internet. Type in your credit card number into random website. It's not the way to pay for things on the internet and I'd rather think of Bitcoin as the first iteration the first attempt to solve the problem of how do we pay for things in what we must learn to call the metaverse. But let's just call it the internet for old time's sake. And ever since that initial uh innovation, the realization that you could use computing power and cryptography to create Peer to peer payments without 3rd party verification. A revolution has been gathering momentum that poses a very profound threat to the existing legacy system of banks and currencies. Most money in the world today is made by banks, north central banks. Banks that's what most money is. Its entries in bank accounts and what Bitcoin represents is an alternative mode of payment that really ought to render banks obsolete. I think this financial revolution has got past the point at which it can be killed. It was vulnerable in the early years but it now has sufficient adoption and has generated sufficient additional layers. The theory and was in many ways the more important innovation because you can build a whole system of payments and ultimately smart contracts on top of ether. I think we've now reached the point that it's pretty hard to imagine it all being killed and it's just survived an amazing thing which was the chinese shutting down mining and shutting down everything and still here we are. Uh, in fact crypto is thriving. What we don't know is how much damage ill judged regulatory interventions are going to do to this financial revolution left to its own devices. I think decentralized finance provides the native monitoring financial system for the internet. And the more time we spend in the metaverse, the more use we will make of it, the next things that will happen I think will be that tokens in game spaces like roadblocks will become fungible as my nine year old spends a lot more time playing on computer games than I ever did. I can see the entertainment is becoming a game driven phenomenon and in the game space you need skins for your avatar, the economics of the internet. It's evolving very fast and in parallel you can see this payments revolution happening. I think that all that all goes naturally very well and generates an enormous amount of wealth and the process, the problem is, there are people in Washington with an overwhelming urge to intervene and disrupt this evolutionary process partly, I think, out of a muddled sense that there must be a lot of nefarious things going on. If we don't step in many more will go on this. I think greatly exaggerates how much criminal activity is in fact going on in the space, but there's also the vested interests at work. It was odd to me, maybe not old, perhaps it wasn't surprising that the bank for international settlements earlier this year published a report, one chapter of which said, this must all go, let's all stop. So we've got to be shut down and it's got to be replaced by a central bank digital currency and martin Wolf and the Financial Times read this and said, I agree with this. And one suddenly realized that the banks are clever. They had they achieved the intellectual counterattack uh, with almost no fingerprints on the weapon. I think central bank digital currency is a terrible idea. I can't imagine why we would want to copy a chinese model that essentially takes all transactions and puts them directly under the surveillance of a central government institution. But that suddenly is a serious counterproposal. So on the one side, we have a relatively decentralized, technologically innovative Internet native system of payments that has the possibility to evolve to produce a full set of, of smart contracts, reducing enormously the transaction costs that we currently encounter in the financial world because it gets rid of all those middle men who take their cut every time you take out a mortgage or whatever it is. That's one alternative. On the other side, we have highly centralized system in which transactions will by default be under the surveillance of the central bank. Seems like an easy choice to me, but hey, I have this thing about personal liberty. So that's where we are. I don't think that the regulators can can kill uh, Web three, I think we're supposed to call Web three because crypto is now an obsolescent term. They can't kill it but they can definitely make it difficult and through a lot of sand into the machine. And I think the worst of all they can spoil the evolutionary story by by creating a central bank digital currency that I don't think we really need or we certainly don't need it in the chinese form. </s>

<s>[INST] Yeah, I mean the meeting to be honest, it's one it's off, it's sometimes misused I think like with anything it's like people adult memes that used to be brilliant are still brilliant and they're just not good at using them. So they become mean [/INST] uh </s><s>[INST] but when you do with Grace it can tear down an argument and at the same time have like love and respect underneath it. I mean it's a beautiful dance, they have to be good at like you know, people just can suck a communication and even uh like even a powerful weapon, like a meme in the wrong hands just uh fires in a way that doesn't get anything done. But this is like a war of humor and memes, it's kind of fascinating. Exactly like you formulated that there's asymmetry of power so you have to have Guerilla warfare in this internet game, especially when there's no leader like you said in a distributed [/INST] culture say here that is um is really important I think is from a society standpoint, we've become very soft and very kind of coddling and not not in a way that's like, I think people take this argument too far sometimes, but what I mean by that is, um it's almost like if you're the person who hold somebody accountable, you become the bad person, right? If you're the person who um says, hey, you know, that's wrong, you're the bad person, right? And so in a world where I think and um this kind of influence serie all positive, if you have any negative feedback or constructive criticism, like you're the bad person, uh it's the ultimate echo chamber, right? And so I think that what the Bitcoin world does in some crazy, crazy way to look at it is Bitcoin is ultimately about truth, not about narrative, not about feelings or emotions, it's math. You look at a Blockchain and you can prove something or you can't. And so naturally people were attracted to that have a very similar approach in life, right? They say, hey, you made exclaim prove it. And as you can imagine, you know, a great example is like the financial media meets Bitcoin ear's and it's a bloodbath in kind of the the arena of ideas because what do they do? The financial media is used to the soft, you know, opinion pieces, etcetera and big corners show up and they're like uh here's data point A B and C. Here's example, 12 and three and you're wrong. Um and then all the yell and scream about is like I'm wrong, I'm wrong, I'm wrong, like you can't say I'm wrong and like no like disprove what I just said and so you're getting this like very very weird. </s><s>[INST] It's fascinating, it's a fascinating benefit but I do want to say is I've been watching this, it's kind of interesting. I think that the pursuit of truth through like tearing down bad ideas can be done with grace and to do it with grace requires a lot of skill. Like what people don't realize about this agreement. They think that this agreement is easy like they they see the the lies or the inaccuracies in the statement and and they just think they can say wrong yes, you can say that, but if you want to be effective, it requires great skill. Like you look at, I don't know, a beautiful verbal shit poster which is Christopher Hitchens right? It requires a lot of skill through your words to tear down an argument to criticize and to take a step towards truth. Uh what I am disheartened by internet culture, like the negative side is people don't put a lot of effort in there tear downs like into your ship posting into your memes, you should put effort and see it as a skill that you want to, if you want to be a part of this culture, you want to ah Get good at it, like any skill to 10,000 hours get improved deliberate practice self criticism, all of those things just because you're anonymous doesn't mean you won't get deep joy and actually have an impact on the world if you get good at ship posting. [/INST] But I think this is really, really important, right? Because you're right in that uh it's all about intention verse action if your intention is to uh tell somebody that they are wrong in an effort to get them to see the truth. Yes, that's very different than if your intention is to tell someone they're wrong and hurt their feelings, right? And so when you can unpack intention in action, you really quickly can tell what somebody ultimately is trying to accomplish. I also think that one of the craziest things that I've seen play out is uh memes when I use that term, I'm not just talking about like a static photo, right? When I'm talking about these elaborate kind of edited videos and kind of all this stuff. Um when done right, it is the most articulate way to deliver a blunt message and it's done in such a way that is humorous and entertaining yet really hammers the point home. And so it's a skill set that many people don't have. I don't make those. I'm assuming you don't make them either, right? I see them, I share the ones that I like, right? Uh But it does take practice and you can tell, look, there's people who are fantastic meme lords and there are people who absolutely suck at it. And it's like anything, it's just how good are you at communicating. And I've heard the idea a bunch of times, so I don't know who to kind of credit for it, but whether it's emojis, it's gifts, it's memes, whatever this is, the extension and evolution of just how hieroglyphics like we have been doing this for literally centuries. It's just that now we're doing it on the internet and you can press a button and go to millions and millions of people immediately. </s>

<s>[INST] That is starts to get to what I call a network state. There should be like a standing applause line here. This is brilliant. You're an incredible person. This was an incredible conversation we covered, how to fix our government, looking at the future of governments, moving into network state. We covered how to fix medicine, FDA, longevity. That was just like a stellar description. Really, I'll have to listen to that multiple times to really think and thank you for that, especially in this time where the lessons learned from the pandemic are unclear to at least me. And there's a lot of thinking that needs to be done there. And then just a discussion about how to fix social media and how to fix money. This was brilliant. So you're an incredibly successful person. So if you taught a course at Stanford for startups, that's a whole nother discussion that we can have. But let me just ask you, there's a lot of people that look up to you. So if there's somebody who's young in high school, early college, trying to figure out what the heck to do with their life, what to do with their career, what advice could you give them? How they can have a career they can be proud of [/INST] or how they can have a life they can be proud of. At least what I would do. And then you can take it or leave it or what have you. But so. </s><s>[INST] Yeah, to maybe to your younger self, [/INST] advice to your younger self. You know, my friend Neville is, this is a lot of what he puts out is the very practical brass tacks, next steps. And I tend towards the macro. Of course, we both do sort of, both kind of thing, right? But let's talk brass tacks, next steps. Because I actually am practical, or at least practical enough, </s><s>[INST] to get things done, I think. It's just like you said, [/INST] you're breaking up the new book into three. Yes, it's motivation, theory and practice. Motivation, theory and practice. That's right. And each of those- So let's talk practice. So let's talk practice. Okay. Especially in visual scale. Right, so first, what skill do you learn as a young kid, right? So let me just give what the ideal full stack thing is. And then you have to say, okay, I'm good quantitatively, I'm good verbally, I'm good this, I'm good that, right? So the ideal is you are full stack engineer and full stack influencer, or full stack engineer or full stack creator, okay? So that's both right brain and left brain, all right. So what does that mean with engineering? That means you master computer science and statistics. Okay. And of course, it's also good to know physics and continuous math. And so that's actually quite valuable to know. And you might need to use a lot of that continuous math with AI nowadays, right? So a lot of that is actually helpful, right? Great descent and whatnot. But computer science and stats are to this century what physics was to the last, why? Because, for example, what percentage of your time </s><s>[INST] do you spend looking at a screen of some kind? [/INST] Just a large percentage of the time. A large percentage of the time, right? Probably more than, you know, for many people, it's more than 50% of their waking hours. If you include laptop, you include cell phone, tablet, you know, your watch, you know, maybe a monitor of some kind, right? All those together is probably, it's a lot, okay? Which means, and then that's gonna only increase with AR glasses, okay? Which means most of the rest of your life </s><s>[INST] will be spent in a sense in the matrix. [/INST] A large percentage, okay? In a constructed digital world, which is more interesting in some sense than the offline world, because we look at it more, it changes faster, right? And where the physics are set by programmers, okay? And what that means is, you know, physics itself is obviously very important for the natural world. Computer science and stats are for the artificial world, right? And why is that? Because every domain has algorithms and data structures, whether it's aviation, okay? You go to American Airlines, right? They're gonna have, you know, planes and seats and tickets and so on. So it's data structures, and you're gonna have algorithms and functions that connect them. You're gonna have tables that those data are into. If it's Walmart, you're gonna have SKUs and you're gonna have shelves. So you have data structures and you have algorithms to connect them. So every single area, you have algorithms and data structures, which is computer science and stats. And so you're going to collect the data and analyze it, right? And so that means if you have that base of CS and stats, where you're really strong and you understand the theory as well as the practice and you need both, okay? Because you need to understand, you know, obviously, basic stuff like big O notation and whatnot. And you need to understand your probability distributions. Okay. Um, you know, a good exercise, by the way is to go from the Bernoulli trials, right? To everything else, because you can go through newly trials to the Binomial distribution, to the Gaussian, you can also go from, you know, Bernoulli trials to the geometric distribution, and so on, you can drive Everything from this right? </s><s>[INST] Computer science includes not just big go, but software engineering. [/INST] Well, computer sciences theory, software engineering is practice. Right. Yeah, you could argue probability and status theory. and then data science is practice. You include all of that together? In a package, that's theory in practice, right? Look, it's okay to use libraries once you know what's going on under the hood. That's fine, but you need to be able to write out </s>

<s>[INST] Okay, we've in 2020 and still living through a pandemic setting, the political and all those kinds of things aside, I've always found virus is fascinating as, as living as dynamical systems. I was going to say living systems, but I've always kind of thought of them as living, you know, but that's a whole nother kind of discussion. Maybe it would be great to put that on the table one. Do you find viruses beautiful slash terrifying and to do you think they're living things or there's some aspect to them for our discussion of life that makes them living? [/INST] I mean living in a pandemic saying viruses are beautiful, it's probably a hard thing, but I do find them beautiful to a degree. I think even even even in the sense of mediating a global pandemic, there's something deeply intriguing there because, you know, these are tiny, tiny little things, right? And yet they can um uh you know, essentially um like cause a seizure like, you know, handicap an entire civilization at a global scale. So just that intersection between, you know, our perceived invincibility and our susceptibility to things and and also the interaction across scales of those things is just really amazing feature of our world. </s><s>[INST] Most technology, whether it's viruses or ai that can scare scale in an exponential way, like kind of run, you know, like as opposed to like one thing, makes another thing makes another thing. It's one thing makes two things and those two things make four things. And like that kind of process, it also seems to be fundamental to life. And uh it's terrifying because in a matter of in a very short time scale it can, if it's good at being life, whatever that is, it can quickly overtake the other competing forms of life. And that's scary both for ai and for viruses. And it seems like understanding these processes that are underlying viruses and I don't mean like on the virology and biology side, but on some kind of more computational physics perspective as we've been talking about, seems to be really important to figure out how humans can survive right along with these kinds of this kind of life and perhaps becoming a multi planetary species is is a part of that. Like there's no maybe like we'll figure out from a physics perspective is like there's no way any living system can be stable for prolonged period of time and survive unless it expands exponentially throughout. Like we have to multiply otherwise, anything that doesn't multiply exponentially will die eventually, Maybe that's a fundamental [/INST] law. Maybe, I don't know, I, you know, I always get really bothered by these Darwinian narratives that are like, you know, like the fittest replicator winds and things and I don't, I just don't feel like that's exactly what's going on. I think like the copying of information is sort of ancillary to this other process of creativity, right? So like, the drive is actually the drive is creativity, but if you want to keep the creativity that's existed in the past, it has to be copied in the future. So replication, like if you, so that for me is so I had this set of arguments with Michael Lachmann and lee Cronin about the, like, life being about persistence, they thought it was about persistence and like survival of the fittest kind of thing, and I'm like, no, it's about existence, it's like because when you're talking about that, it's easy to say that in retrospect, you can post select on the things that survived and then say why they survived. But um but you can't do that going forward, </s><s>[INST] that's really profound that survival is just a nice little side effect feature of maximizing creativity, but it doesn't need to be there. [/INST] Yeah, </s><s>[INST] that's really beautiful. [/INST] Yeah, that's really </s><s>[INST] optimistic theories. Well, I don't know if that's the ultimate, that can be terrifying to people because because uh you know, a system that maximizes creativity may very quickly get rid of humans for some reason if it comes up with some other creative I mean, yeah, forms of existence. Yeah, but this is the Ai thing is like the moment you have an Ai system that can that can flourish in the space of ideas or in some other space much more effectively than humans and it's sufficiently integrated into the physical space to be able to modify the environment. [/INST] I think we'll just be like the core genetic architecture or something will be like the DNA for AI right? It's like we haven't lost the past informational architecture is on this planet. They're still there. </s><s>[INST] Yeah. Also the the Ai will use our brains in some part to like like ride, like accelerate the exchange of ideas. That's the new link dream is that well the humans will be still around because you're saying architecture. [/INST] Yeah, but I don't I don't even think they necessarily need to tap in our brains just collectively we do interesting things. What if they were just using the patterns in our communication or something. </s><s>[INST] Oh, without controlling it, just observing. [/INST] Well, I don't know in what sense do you control the chemistry happening in your body. </s><s>[INST] Uh huh. Yeah, [/INST] I mean I I obviously, I don't know, I'm just. I I just like the way I look at like people look at ai and then they look at this thing that's bigger than us and it's coming in the future and it's smarter than us. And I think though that looking at the past history of life on the planet and what information has been doing for the last four billion years is probably very informative to asking questions about what's coming next. Um And I don't one is a planetary scale transitions are really important for new phases. So the global internet and sort of global integration of our technology, I think it's an important thing. So that's again, life is a planetary scale phenomena but we're an integrated component of that phenomenon. I don't really see that the technology is going to replace us in that way. It's just gonna keep scaffolding and building. And I also don't have an idea that we're gonna build a I in a box. I think ai is going to emerge a G. I. To be is a planetary scale phenomenon that's going to emerge </s>

<s>[INST] restoration well [/INST] yeah how </s><s>[INST] can you do that so what's the what's the [/INST] what's </s><s>[INST] interesting about the technology [/INST] so it is very interesting because then that completely avoids the potential issues you may have with the blinds when you are diving the risk of bubbles are trapped in your lungs because of different pressures and different gases </s><s>[INST] also there's there's a complexity to the flow of oxygen under water [/INST] when you are breathing regular air when you are scuba diving here you know that you have to do some different when you're coming back when you are diving deep and you have conventional air then you need to stop so that you can equalize the gases in your lungs if you come back to fast then you can have air bubbles stock and then you can raise the blinds which you can use to be paralyzed you can reserve number of nasty issues and we wanted absolutely to avoid that so diving with pure oxygen avoids this completely N. it has a a another benefits and I'll see to that is that well the greater risk when you arrive at altitude is altitude sickness what is altitude sickness is just you not having enough oxygen in your blood so this is you know this was a good benefit it was a good tradeoff we were lucky enough to be also trained by the ministry so we came up with not the civilian retriever which is the big thing on the on the back that you carry on your back we actually we are given navy seals commando </s><s>[INST] rivers you are the navy seals for this [/INST] the director of I needed three operations yes we were traveling we were translate like astronauts for three months I spend more time I had to joke that if somebody wanted to reach me they better put a a phone line at the bottom of the swimming pool because this is where I was so we train and we train in the manual about the safety was about that's sick so it was a real operation that was three years into a into a because when your free diving in the years prior there is no risk you don't have any other gas in your long that what you're breathing the only Reese's come short of air and then you're in trouble which happened to me one time seriously so </s><s>[INST] about that time I got to ask you about freediving before we return to the rebreathers [/INST] well all these lakes I had all to do that they are cold they are the minimum temperature that you can have on bodies of water a clear bodies of water which is four degree C. it's very very cold and so you </s><s>[INST] think ice bath [/INST] you someone yeah so you cannot just dive with a wet suit so the idea was to take a dry suit and I learned how to free dive with a dry suit which is really the worst thing you can do </s><s>[INST] what's the what's the what's the dress itself a dry suit [/INST] a wet suit is usually what you use it in the ocean when it's not too cold you can use also dry suit but the white suit basically is going to keep you warm because the water is getting into the suit end of the call the curb your skin is getting to body temperature and so for a while you can in dive like that a need in the ocean here that's fine that that that that that's fine the dry suit is the opposite it's completely closed which means that you don't have any contact with the water outside and you keep your warmth through a your body temperature and and even clothing that you can put into it so this is dry issue they are used by the other goal really deep in very cold weather and need to stay a long time on the water </s><s>[INST] so what's the bad part [/INST] the best part is is that when you have those a dry suit you have a lot of air that can be trapped in it usually we do what we call burping dispute it's not very pleasant expression but you get in the water and as soon as you get in the water you can see the air pockets all over the place right so you burn dispute you open the valve and the air comes out once you have done that then you look with your lead belt in in in you know when you're ready to go down and so what happened that day is that I actually did birth dispute but didn't realize that </s><s>[INST] it [/INST] completely and so I went down N. immediately I felt the air pocket going to my legs so basically air was trapped in the suit and went on my legs as I was driving like that </s><s>[INST] and so I didn't pay too much attention to rescue diving down [/INST] yeah I was driving down in so didn't pay too much attention but that as always you will be seeing a chest the awkward position but then I wanted to turn and go up while no can do I was just like a bully and I was like that so I had a first time issue okay I tried a second time and the third time and by the fourth time I kind of realized I was in trouble in the fifth time I say okay now you better give it your best try otherwise it's going to be big trouble </s><s>[INST] error this is free diving [/INST] I was free diving and then </s><s>[INST] you can [/INST] and I cannot </s><s>[INST] feeling I mean is there panic or not [/INST] there is no pending because you can't you cannot afford to be panicking in fact you are always thinking because there is training this is the best part about training your training allows you that space to keep you cool N. and compose which you need to be in that kind of situation and so finally after the fifth time I was able to rectify the position and get myself up but when I get up my lungs where MT I'd been in the water for quite some time and I'll cook I knew what was going to happen so I decided to just be the plank hello not move and don't do anything just open my mouth and try to suck oxygen but obviously oxygen at six thousand meters twenty thousand feet there's not that much it's about a little it's a forty percent of what you breeze at sea level so although it was noon at the time the sky you stayed pretty dark and starry </s>

<s>[INST] Yeah, I've seen some interesting work on simulation, most simulation cosmology by the way. It's just awesome. But you know, just like simulation of the early formation of our solar system to understand how the like the or cloud and just, I don't know the whole of it, how earth came to be. Like how jupiter, just [/INST] just protects </s><s>[INST] us, protects us. And then there's like weird like uh moons and volcanoes and like modeling all of that. The formation of all of that uh is fascinating because that naturally is the question of how does life emerge and these kinds of rocks? How does Iraq become a rabbit? Uh but speaking of models, uh there's an equation called the drake equation we were talking about life have to ask when you at the highest level. First, when you look out there, How many alien civilizations do you think are out there? Well, 01 or many. [/INST] So if you say civilization, I would bring my number way down. It could be zero. If you talk about life, I think it could be many as we were saying before. I think the move from life to consciousness, the kinds of beings that would build what we would recognize as a civilization that may be extraordinarily rare. I hope it's not, you know, as a kid. I love star trek. I just love the idea that we would be part of some universal community where look experience on planet Earth suggests it doesn't always go so well when groups who are separated, try to come together and and live in some larger collective. But again, as an optimist, how amazing would it be to converse with an alien civilization and and learn what they've figured out about physics and cosmology and compare notes and and and learn from each other in in some some wonderful way. I love that idea. But if you ask me the likelihood of it, I would err on saying it may be so improbable that the conditions conspire to allow life to move to this place of consciousness that it might be rare. </s><s>[INST] It might be oversimplifying things, but just observing the power of the evolutionary process, I tend to believe. And like you read different theories of how we went, uh how homo sapiens evolved. It seems like the evolutionary process natural at least to homo sapiens are or creatures like that or much better than that. So to me, the there's several scary scenarios. So okay, the positive scenario is life itself is really difficult. So that origin of life is difficult. That's exciting for many reasons because we might be able to prove that wrong easily in the near term by finding life elsewhere. The scary thing to me is if uh, life is easy and there's plenty of conscious intelligent civilizations out there and we have not obviously made contact, which means with intelligence and consciousness comes responsibility and uh ultimately destruction. So with power comes great responsibility and then we end up destroying ourselves. That's the, the scariest, the positive, I guess version is that maybe we're being watched, sort of like there's a transition to where you don't want to ruin the primitive villages out there and so there's a protective layer around us they're watching. So where where do you in these possible explanations for the fermi paradox? Why haven't we contacted aliens? Do you, do you land [/INST] on? I think the most straightforward explanation is that there aren't any. Now there are many other explanations too. So you can't be dogmatic about things that are just sort of gut feel. But you know, one of my favorite twilight zone episodes and I never saw this one where this alien civilization finally comes to planet Earth and gives us this book that they really want us to to have and to hold and it's in this foreign language, you don't understand the cryptographers. They desperately try to decipher it as humans are going to visit this other alien plant and they're all standing back postcards how wonderful it is and so forth. And they finally decide for the title, it's to serve Man and everyone is so thrilled they're here to service it all makes sense. And then just as one of the final cryptographers is going onto the alien ship, his his helper runs and says, I've deciphered the rest of the book to serve man. It's a cookbook. So you know, so yeah. Is that is that a possibility? Sure. You know, and and so could they be watching us and just sort of waiting for us to get to a mature enough level? I don't know, it strikes me well, you know, I think it be better to have this conversation after the James webb telescope. I mean I do think that um if we look at the atmospheres of many planets, there's now an estimate now that there's on the order of one planet per star on average. So we've long known that, you know, the galaxy, hundreds of billions of stars, numbers of Galaxies, hundreds of billions of Galaxies that we're talking about hundreds of billions of hundreds of billions of planets. You know. And if we start to survey some of these planets and one after the other after the other, we just sort of find no evidence for any of the biological markers. It could be of course, maybe life takes a radically different form. It would be hard to know that. But I think, you know, that would at least give us some insight on the life question. But I just don't see how we get insight on the civilization or consciousness question without, you know, the direct connection. And and it strikes me that if consciousness is ubiquitous, let's say life is, I'm willing to grant that if consciousness is also ubiquitous, then I don't understand why they haven't been here or why there hasn't been separate section because presumably they should be much further ahead of us. How unlike would it be that were like of all consciousness in the universe, were the most advanced. That'd be such a special place for human beings. That it's hard for me to grant That is a likely possibility. Rather. I think we're kind of running the mill and there are many who are far more advanced than us and I don't think that they would expend the energy to hide themselves. So I don't think they care enough. </s>

<s>[INST] For 180 bucks a year, you get an all-access pass to watch courses from, to list some of my favorites. Chris Hadfield on space exploration, Neil deGrasse Tyson on scientific thinking and communication, Will Wright, creator of the greatest city building game ever, SimCity and Sims on game design, Carlos Santana on guitar, Gary Kasparov, the greatest chess player ever on chess, Daniel Negrono on poker, and many more. Chris Hadfield explaining how rockets work and the experience of being launched into space alone is worth the money. Once again, sign up at Masterclass.com slash Lex to get a discount and to support this podcast. And now here's my conversation with Ben Kurtzel. [/INST] What books, authors, ideas had a lot of impact on you in your life in the early days? You know, what got me into AI and science fiction and such in the first place wasn't a book, but the original Star Trek TV show, which my dad watched with me like in its first run, it would have been like 1968, 69 or something. And that was incredible because every show that visited a different alien civilization with different culture and weird mechanisms. But that got me into science fiction, and there wasn't that much science fiction to watch on TV at that stage. So that got me into reading the whole literature of science fiction, you know, from the beginning of the previous century until that time. And I mean, there was so many science fiction writers who were inspirational to me. I'd say if I had to pick two, it would have been Stanislaw Lem, the Polish writer. Yeah. Solaris. And then he had a bunch of more obscure writings on superhuman AIs that were engineered. Solaris was sort of a superhuman, naturally occurring intelligence. And Philip K. Dick, who ultimately my fandom for Philip K. Dick is one of the things that brought me together with David Hansen, my collaborator on Robotics Project. So Stanislaw Lem was very much an intellectual, right? So he had a very broad view of intelligence going beyond the human and into what I would call open-ended super intelligence. The Solaris super intelligent ocean was intelligent, in some ways more generally intelligent than people, but in a complex and confusing way so that human beings could never quite connect to it. But it was still probably very, very smart. And then the Golem Four supercomputer in one of Lem's books, this was engineered by people, but eventually it became very intelligent in a different direction than humans and decided that humans were kind of trivial and not that interesting. So it put some impenetrable shield around itself, shut itself off from humanity, and then issued some philosophical screed about the pathetic and hopeless nature of humanity and all human thought, and then disappeared. Now Philip K. Dick, he was a bit different. He was human focused, right? His main thing was human compassion and the human heart and soul are going to be the constant that will keep us going through whatever aliens we discover or telepathy machines or super AIs or whatever it might be. So he didn't believe in reality, like the reality that we see may be a simulation or a dream or something else we can't even comprehend, but he believed in love and compassion as something persistent through the various simulated realities. So those two science fiction writers had a huge impact on me. Then a little older than that, I got into Dostoevsky and Friedrich Nietzsche and Rambod </s><s>[INST] and a bunch of more literary typewriting, Polish writers. Yeah. Can we talk about some of those things? So on the Solaris side, Stanislaus Lem, this kind of idea of there being intelligences out there that are different than our own, do you think there are intelligences maybe all around us that we're not able to even detect? So this kind of idea of maybe you can comment also on Steven Wolfram thinking that there's computations all around us and we're just not smart enough to kind of detect their intelligence or appreciate their intelligence. [/INST] Yeah, so my friend Hugo de Geras, who I've been talking to about these things for many decades since the early 90s, he had an idea he called SIPI, the search for intraparticulate intelligence. So the concept there was as AIs get smarter and smarter and smarter, assuming the laws of physics as we know them now are still what these super intelligences perceived hold and are bound by, as they get smarter and smarter, they're going to shrink themselves littler and littler, because special relativity makes it so they can communicate between two spatially distant points. So they're going to get smaller and smaller. But then ultimately, what does that mean? The minds of the super, super, super intelligences, they're going to be packed into the interaction of elementary particles or quarks or the partons inside quarks or whatever it is. So what we perceive as random fluctuations on the quantum or sub-quantum level may actually be the thoughts of the micro, micro, micro miniaturized super intelligences because there's no way we can tell random from structured but with an algorithmic information more complex in our brains. Right? We can't tell the difference. So what we think is random could be the thought processes of some really tiny super minds. And if so, there's not a damn thing we can do about it except try to upgrade our intelligences </s><s>[INST] and expand our minds so that we can perceive more of what's around us. But if those random fluctuations, like even if we go to like quantum mechanics, if that's actually super intelligent systems, aren't we then part of the soup of super intelligence? So aren't we just like a finger of the entirety of the body of the super intelligence system? [/INST] It could be, I mean, a finger is a strange metaphor. I mean, a finger is dumb is what I mean. But a finger is also useful and it's controlled with intent by the brain, whereas we may be much less than that. Right? I mean, yeah, we may be just some random epiphenomenon that they don't care about too much. Like think about the shape of the crowd emanating from a sports stadium or something, right? There's some emergent shape to the crowd. It's there. You could take a picture of it. It's kind of cool. It's irrelevant to the main point of the sports event or where the people are going or what's on the minds of the people making that shape in the crowd, right? So we may just be some semi-arbitrary higher level pattern popping out of a lower level hyper intelligent self-organization. And I mean, so be it, right? I mean, that's one thing that... Still a fun ride. Yeah. I mean, the older I've gotten, the more respect I've achieved for our fundamental ignorance. I mean, mine and everybody else's. I mean, I look at my two dogs, two beautiful little toy poodles and, you know, they watch me sitting at the computer typing. They just think I'm sitting there wiggling my fingers to exercise them maybe or guarding the monitor on the desk that they have no idea that I'm communicating with other people halfway around the world, let alone, you know, creating complex algorithms running in RAM on some computer server in St. Petersburg or something, right? Although they're right there, they're right there in the room with me. So what things are there right around us that were just too stupid or close-minded to comprehend? </s><s>[INST] Probably quite a lot. Your very poodle could also be communicating across multiple dimensions with other beings and you're too unintelligent to understand the kind of communication mechanism they're [/INST] going through. There have been various TV shows and science fiction novels, positing cats, dolphins, mice and whatnot are actually super intelligences here to observe that. I would guess, as one of the other quantum physics founders said, those theories are not crazy enough to be true. </s><s>[INST] The reality is probably crazier than that, the reality is probably crazier than that. We put, so on the human side with Philip K. Dick and in general, where do you fall on this idea that love and just the basic spirit of human nature persists throughout these multiple realities? Are you on the side, like the thing that inspires you about artificial intelligence, is it the human side of somehow persisting through all of the different systems we engineer? Or does AI inspire you to create something that's greater than human, that's beyond human, [/INST] that's almost non-human? I would say my motivation to create AGI comes from both of those directions actually. So when I first became passionate about AGI when I was, it would have been two or three years old after watching robots on Star Trek, I mean then it was really a combination of intellectual curiosity, like can a machine really think, how would you do that? And yeah, just ambition to create something much better than all the clearly limited and fundamentally defective humans I saw around me. Then as I got older and got more enmeshed in the human world and got married, had children, and my parents began to age, I started to realize, well, not only will AGI let you go far beyond the limitations of the human, but it could also like stop us from dying and suffering and feeling pain and tormenting ourselves mentally, so you can see AGI has amazing capability to do good for humans, as humans, alongside with its capability to go far, far beyond the human level. So I mean both aspects are there, which makes it even more exciting and important. </s><s>[INST] So you mentioned Dostoyevsky Nietzsche, where did you pick up from those guys? [/INST] That would probably go beyond the scope of a brief interview certainly. Both of those are amazing thinkers who one will necessarily have a complex relationship with, right? I mean Dostoyevsky on the minus side, he's kind of a religious fanatic, and he sort of helped squash the Russian nihilist movement, which was very interesting, because what nihilism meant originally in that period of the mid-late 1800s in Russia was not taking anything fully a hundred percent for grand. It was really more like what we'd call Bayesianism now, where you don't want to adopt anything as a dogmatic certitude and always leave your mind open, and how Dostoyevsky parodied nihilism was a bit different, right? He parodied people who believe absolutely nothing, so they must assign an equal probability weight to every proposition, which doesn't really work. So on the one hand, I didn't really agree with Dostoyevsky on his sort of religious point of view. On the other hand, if you look at his understanding of human nature and sort of the human mind and heart and soul, it's really unparalleled. He had an amazing view of how human beings construct a world for themselves based on their own understanding and their own mental predisposition. I think if you look in the Brothers Karamazov in particular, the Russian literary theorist Mikhail Bakhtin wrote about this as a polyphonic mode of fiction, which means it's not third person, but it's not first person from any one person, really. There are many different characters in the novel, and each of them is sort of telling part of the story from their own point of view. So the reality of the whole story is an intersection, like, synergetically of the many different characters' worldviews, and that really, it's a beautiful metaphor and even a reflection I think of how all of us socially create our reality. Each of us sees the world in a certain way. Each of us, in a sense, is making the world as we see it based on our own minds and understanding, but it's polyphony, like in music where multiple instruments are coming together to create the sound. The ultimate reality that's created comes out of each of our subjective understandings intersecting with each other. That was one of the many beautiful things in Dostoyevsky. </s><s>[INST] So maybe a little bit to mention, you have a connection to Russia and the Soviet culture. I mean, I'm not sure exactly what the nature of the connection is, but at least the spirit [/INST] of your thinking is in there. Oh, yeah. Oh, yeah. Well, my ancestry is three quarters Eastern European Jewish, so I mean, three of my great grandparents emigrated to New York from Lithuania and sort of border regions of Poland, which were in and out of Poland around the time of World War I, and they were socialists and communists as well as Jews, mostly Menshevik, not Bolshevik, and they sort of, they fled at just the right time to the US for their own personal reasons. And then almost all or maybe all of my extended family that remained in Eastern Europe was killed either by Hitler's or Stalin's minions at some point. So the branch of the family then emigrated to the US was pretty much the only one. </s><s>[INST] So how much of the spirit of the people is in your blood still? When you look in the mirror, do you see, what do you see? [/INST] I see a bag of meat that I want to transcend by uploading into some sort of superior reality. But very, I mean, yeah, very clearly, I mean, I'm not religious in a traditional sense, but clearly the Eastern European Jewish tradition was what I was raised in. I mean, there was my grandfather, Leo Zwell, was a physical chemist to work with Linus Pauling and a bunch of the other early greats in quantum mechanics. I mean, he was into x-ray diffraction, he was on the material science side, experimentalist rather than a theorist. His sister was also a physicist and my father's father, Victor Gertzel, was a PhD in psychology who had the unenviable job of giving psychotherapy to the Japanese in internment camps in the US in World War II, like to counsel them why they shouldn't kill themselves, even though they had all their stuff taken away and been imprisoned for no good reason. So I mean, yeah, there's a lot of Eastern European Jewish tradition in my background. One of my great uncles was, I guess, conductor of San Francisco orchestra. So there was a lot of Mickey Salk and a bunch of music, music, and there also, and clearly this culture was all about learning and understanding the world and also not quite taking yourself too seriously while you do it, right? There's a lot of Yiddish humor in there. So I do appreciate that culture, although the whole idea that like the Jews are the </s><s>[INST] chosen people of God never resonated with me too much. The graph of the Gertzel family, I mean, just the people I've encountered, just doing some research and just knowing your work through the decades, it's kind of fascinating. Just the number of PhDs. [/INST] Yeah, yeah. It's kind of fascinating. My dad is a sociology professor who recently retired from Rutgers University, but that clearly that gave me a head start in life. My grandfather gave me all this quantum mechanics books and I was like seven or eight years old and I remember going through them and it was all the old quantum mechanics, like Rutherford Adams and stuff. So I got to the part of wave functions, which I didn't understand, although I was a very bright kid and I realized he didn't quite understand it either, but at least like he pointed me to some professor he knew at UPenn nearby who understood these things, right? So that's an unusual opportunity for a kid to have, right? My dad, he was programming Fortran when I was 10 or 11 years old on like HP 3000 mainframes at Rutgers University, so I got to do linear regression in Fortran on punch cards when I was in middle school, right? Because he was doing, I guess, analysis of demographic and sociology data. So yes, certainly that gave me a head start and a push towards science beyond what would have been the case with many, many different situations. </s><s>[INST] When did you first fall in love with AI? Is it the programming side of Fortran? Is it maybe the sociology psychology that you picked up from your dad? [/INST] I fell in love with AI when I was probably three years old when I saw a robot on Star Trek. I was turning around in a circle going, error, error, error, error, because Spock and Kirk had tricked into a mechanical breakdown by presenting with a logical paradox. And I was just like, well, this makes no sense. This AI is very, very smart. It's been traveling all around the universe, but these people could trick it with a simple logical paradox. Like why, if the human brain can get beyond that paradox, why can't this AI? So I felt the screenwriters of Star Trek had misunderstood the nature of intelligence. I complained to my dad about it and he wasn't going to say anything one way or the other. Before I was born, when my dad was at Antioch College in the middle of the US, he led a protest movement called SLAM, Student League Against Mortality. They were protesting against death, wandering across the campus. So he was into some futuristic things even back then, but whether AI could confront logical paradoxes or not, he didn't know. But when I, 10 years after that or something, I discovered Douglas Hofstadter's book, Gortel Escherbach, and that was sort of to the same point of AI and paradox and logic, right? Because he was over and over with Gortel's incompleteness theorem and can an AI really fully model itself reflexively or does that lead you into some paradox? Can the human mind truly model itself reflexively or does that lead you into some paradox? So I think that book, Gortel Escherbach, which I think I read when it first came out, I would have been 12 years old or something. I remember it was like 16 hour day, I read it cover to cover and then reread it. I reread it after that because there was a lot of weird things with little formal systems in there that were hard for me at the time. But that was the first book I read that gave me a feeling for AI as like a practical academic or engineering discipline that people were working in. Because before I read Gortel Escherbach, I was into AI from the point of view of a science fiction fan. And I had the idea, well, it may be a long time before we can achieve immortality and superhuman AGI. So I should figure out how to build a spacecraft traveling close to the speed of light, go far away, then come back to the earth in a million years when technology is more advanced and we can build these things. Reading Gortel Escherbach, well, it didn't all ring true to me, a lot of it did. But I could see like there are smart people right now at various universities around me who are actually trying to work on building what I would now call AGI, although Hofstadter didn't call it that. So really, it was when I read that book, which would have been probably middle school, that then I started to think, well, this is something that I could practically work on. </s><s>[INST] Yeah, it's supposed to flying away and waiting it out, you can actually be one of the people that actually... Yeah, exactly. [/INST] Yeah, exactly. And if you think about, I mean, I was interested in what we'd now call nanotechnology and in the human immortality and time travel, all the same cool things as every other like science fiction loving kid. But AI seemed like if Hofstadter was right, you just figure out the right program, sit there and type it. Like you don't need to spin stars into weird configurations or get government approval to cut people up and fill with their DNA or something, right? It's just programming. And then of course, that can achieve anything else. There's another book from back then, which was by Gerald Feinbaum, who was a physicist at Princeton. And that was the Prometheus Project. And this book was written in the late 1960s, though I encountered it in the mid 70s. But what this book said is in the next few decades, humanity is going to create superhuman thinking machines, molecular nanotechnology, and human immortality. And then the challenge we'll have is what to do with it. Do we use it to expand human consciousness in a positive direction or do we use it just to further vapid consumerism? And what he proposed was that the UN should do a survey on this and the UN should send people out to every little village in remotest Africa or South America and explain to everyone what technology was going to bring the next few decades and the choice that we had about how to use it and let everyone on the whole planet vote about whether we should develop super AI, nanotechnology, and immortality for expanded consciousness or for rampant consumerism. And needless to say, that didn't quite happen. And I think this guy died in the mid 80s, so he didn't even see his ideas start to become more mainstream. But it's interesting, many of the themes I'm engaged with now from AGI and immortality, and to trying to democratize technology, as I've been pushing forward with singularity in my work in the blockchain world. Many of these themes were there in, you know, Fine Bomb's book in the late 60s even. And of course, Valentin Turchin, a Russian writer, and a great Russian physicist who I got to know when we both lived in New York in the late 90s and early aughts, I mean, he had a book in the late 60s in Russia, which was The Phenomenon of Science, which laid out all these same things as well. And Val died in, I don't remember, 2004 or 2005 or something of Parkinson'sism. So yeah, it's easy for people to lose track now of the fact that the futurist and singularitarian advanced technology ideas that are now almost mainstream, and they're on TV all the time. I mean, these are not that new, right? They're sort of new in the history of the human species, but I mean, these were all around in fairly mature form in the middle of the last century were written about quite articulately by fairly mainstream people who were professors at top universities. It's just until the enabling technologies got to a certain point, then you couldn't make it real. And even in the 70s, I was sort of seeing that and living through it, right? From Star Trek to Douglas Hofstadter, things were getting very, very practical from the late 60s to the late 70s. And the first computer I bought, you could only program with hexadecimal machine code and you had to solder it together. And then like a few years later, there's punch cards, and a few years later, you could get like Atari 400 and Commodore VIC-20, and you could type on the keyboard and program in higher level languages alongside the assembly language. So these ideas have been building up a while, and I guess my generation got to feel them build up, which is different than people coming into the field now for whom these things have just been part of the ambiance of culture for their whole career or even their whole </s><s>[INST] life. Well, it's fascinating to think about there being all of these ideas kind of swimming almost with the noise all around the world, all the different generations, and then some kind of nonlinear thing happens where they percolate up and capture the imagination of the mainstream. [/INST] And that seems to be what's happening with AI now. I mean, Nietzsche, who you mentioned, had the idea of the Superman, right? But he didn't understand enough about technology to think you could physically engineer a Superman by piecing together molecules in a certain way. He was a bit vague about how the Superman would appear, but he was quite deep at thinking about what the state of consciousness and the mode of cognition of a Superman would be. He was a very astute analyst of how the human mind constructs the illusion of a self, how it constructs the illusion of free will, how it constructs values like good and evil out of its own desire to maintain and advance its own organism. He understood a lot about how human minds work. Then he understood a lot about how post-human minds would work. I mean, this Superman was supposed to be a mind that would basically have complete root access to its own brain and consciousness and be able to architect its own value system and inspect and fine-tune all of its own biases. So that's a lot of powerful thinking there, which then fed in and sort of seeded all of postmodern continental philosophy and all sorts of things have been very valuable in development of culture and indirectly even of technology. But of course, without the technology there, it was all some quite abstract thinking. So now we're at a time in history when a lot of these ideas can be made real, which is amazing, </s><s>[INST] amazing and scary. It's kind of interesting to think, what do you think Nietzsche would if he was born a century later or transported through time? What do you think he would say [/INST] about AI? I mean, those are quite different. If he's born a century later or transported through </s><s>[INST] time? Well, he'd be on like TikTok and Instagram and he would never write the great works he's [/INST] written. So let's transport him through time. Maybe also Sprach Zarathustra would be a music </s><s>[INST] video, right? I mean, who knows? Yeah, but if he was transported through time, do you think... That'd be interesting actually to go back. You just made me realize that it's possible to go back and read Nietzsche with an eye of, is there some thinking about artificial beings? I'm sure he had inklings. I mean, with Frankenstein before him, I'm sure he had inklings of artificial beings somewhere in the text. It'd be interesting to see, to try to read his work, to see if he hadn't... If Superman was actually an AGI system, like if he had inklings of that kind of thinking. [/INST] He didn't. He didn't. He didn't. He didn't. No, I would say not. I mean, he had a lot of inklings of modern cognitive science, which are very interesting. If you look in like the third part of the collection that's been titled The Will to Power, I mean, in book three there, there's very deep analysis of thinking processes, but he wasn't so much of a physical </s><s>[INST] tinkerer type guy, right? He was very abstract. Do you think... What do you think about The Will [/INST] to Power? Do you think human... What do you think drives humans? Is it... Oh, an unholy mix of things. I don't think there's one pure, simple, and elegant objective function driving humans </s><s>[INST] by any means. What do you think, if we look at... I know it's hard to look at humans in an aggregate, but do you think, overall, humans are good? Or do we have both good and evil within us that, depending on the circumstances, depending on the whatever, can perfectly to the top, [/INST] now perfectly to the top? Good and evil are very ambiguous, complicated, and in some ways, silly concepts. But if we... We could dig into your question from a couple directions. So, I think if you look in evolution, humanity is shaped both by individual selection and what biologists would call group selection, like tribe level selection, right? So, individual selection has driven us in a selfish DNA sort of way, so that each of us does to a certain approximation what will help us propagate our DNA to future generations. I mean, that's why I've got to four kids so far, and probably that's not the last one. On the other hand... I like the ambition. Tribal group selection means humans, in a way, will do what will advocate for the persistence of the DNA of their whole tribe or their social group. And in biology, you have both of these, right? And you can see, say, in ant colony or beehive, there's a lot of group selection in the evolution of those social animals. On the other hand, say, a big cat or some very solitary animal, it's a lot more biased toward individual selection. Humans are an interesting balance, and I think this reflects itself in what we would view as selfishness versus altruism to some extent. So, we just have both of those objective functions contributing to the makeup of our brains, and then as Nietzsche analyzed in his own way, and others have analyzed in different ways, I mean, we abstract this as, well, we have both good and evil within us, right? Because a lot of what we view as evil is really just selfishness. A lot of what we view as good is altruism, which means doing what's good for the tribe. And on that level, we have both of those just baked into us, and that's how it is. Of course, there are psychopaths and sociopaths and people who, you know, get gratified by the suffering of others, and that's a different thing. Yeah, those are exceptions. But I think at core, we're not purely selfish, we're not purely altruistic, we are a mix, and that's the nature of it. And we also have a complex constellation of values that are just very specific to our evolutionary history, like we, you know, we love waterways and mountains, and the ideal place to put a house is in a mountain overlooking the water, right? And, you know, we care a lot about our kids, and we care a little less about our cousins, and even less about our fifth cousins. I mean, there are many particularities to human values, which, whether they're good or evil, depends on your perspective. I spent a lot of time in Ethiopia, in Addis Ababa, where we have one of our AI development offices for my Singularity Net project. And when I walked through the streets in Addis, you know, there's so, there's people lying by the side of the road, like, just living there by the side of the road, dying probably of curable diseases without enough food or medicine. And when I walk by them, you know, I feel terrible, I give them money. When I come back home to the developed world. </s><s>[INST] Yeah, those are exceptions. [/INST] They're not on my mind that much. I do donate some, but I mean, I also spend some of the limited money I have enjoying myself in frivolous ways rather than donating it to those people who are right now, like, starving, dying, and suffering on the roadside. So, does that make me evil? I mean, it makes me somewhat selfish and somewhat altruistic. And we each balance that in our own way, right? So, whether that will be true of all possible AGIs is a subtler question. </s><s>[INST] That's how humans are. So, you have a sense, you kind of mentioned that there's a selfish, I'm not going to bring up the whole Ayn Rand idea of selfishness being the core virtue. That's a whole interesting kind of tangent that I think will just distract their souls on it. [/INST] Well, I have to make one amusing comment. Sure. A comment that has amused me anyway. So, yeah, I have extraordinary negative respect for Ayn Rand. Negative? What's the negative respect? But when I worked with a company called Genescent, which was evolving flies to have extraordinary long lives in Southern California. So, we had flies that were evolved by artificial selection to have five times the lifespan of normal fruit flies. But the population of super long-lived flies was physically sitting in a spare room at an Ayn Rand elementary school in Southern California. So, that was just like, well, if I saw this in a movie, I wouldn't believe it. </s><s>[INST] What's the negative respect for you? So, that was just like... Well, yeah, the universe has a sense of humor in that kind of way. That fits in. Humor fits in somehow into this whole absurd existence. But you mentioned the balance between selfishness and altruism as kind of being innate. Do you think it's possible that's kind of an emergent phenomena, those peculiarities of our value system, how much of it is innate? How much of it is something we collectively, kind of like a Dostoevsky novel, bring to life together as a civilization? [/INST] I mean, the answer to nature versus nurture is usually both. And of course, it's nature versus nurture versus self-organization, as you mentioned. So, clearly, there are evolutionary roots to individual and group selection leading to a mix of selfishness and altruism. On the other hand, different cultures manifest that in different ways. Well, we all have basically the same biology. And if you look at sort of pre-civilized cultures, you have tribes like the Yanamamo in Venezuela, which their culture is focused on killing other tribes. And you have other Stone Age tribes that are mostly peaceful and have big taboos against violence. So you can certainly have a big difference in how culture manifests these innate biological characteristics. But still, you know, there's probably limits that are given by our biology. I used to argue this with my great-grandparents, who were Marxists, actually, because they believed in the withering away of the state. They believed that, you know, as you move from capitalism to socialism to communism, people would just become more social-minded so that a state would be unnecessary and everyone would give everyone else what they needed. Now, setting aside that that's not what the various Marxist experiments on the planet </s><s>[INST] seem to be heading toward in practice. Just as a theoretical point, I was very dubious that, [/INST] of course, its human nature could go there. Like at that time when my great-grandparents are alive, I was just like, you know, I'm a cynical teenager. I think humans are just jerks. The state is not going to wither away. If you don't have some structure keeping people from screwing each other over, they're going to do it. So now I actually don't quite see things that way. I mean, I think my feeling now, subjectively, is the culture aspect is more significant than I thought it was when I was a teenager. And I think you could have a human society that was dialed dramatically further toward, you know, self-awareness, other awareness, compassion, and sharing than our current society. And, of course, greater material abundance helps. But to some extent, material abundance is a subjective perception also because many Stone Age cultures perceive themselves as living in great material abundance. They had all the food and water they wanted. They lived in a beautiful place. They had sex lives. They had children. I mean, they had abundance without any factories. So I think humanity probably would be capable of fundamentally more positive and joy-filled mode of social existence than what we have now. Clearly, Marx didn't quite have the right idea about how to get there. I mean, he missed a number of key aspects of human society and its evolution. And if we look at where we are in society now, how to get there is a quite different question because there are very powerful forces </s><s>[INST] pushing people in different directions than a positive, joyous, compassionate existence, right? So if we were trying to, you know, Elon Musk is dreams of colonizing Mars at the moment, so we maybe will have a chance to start a new civilization with a new governmental system. And certainly there's quite a bit of chaos. We're sitting now, I don't know what the date is, but this is June. There's quite a bit of chaos and all different forms going on in the United States and all over the world. So there's a hunger for new types of governments, new types of leadership, new types of systems. And so what are the forces at play and how do we move forward? [/INST] Yeah, I mean, colonizing Mars, first of all, it's a super cool thing to do. We should be doing it. So you love the idea. Yeah, I mean, it's more important than making chocolatier chocolates and sexier lingerie and spend many of the things that we spend a lot more resources on as a species, right? So I mean, we certainly should do it. I think the possible futures in which a Mars colony makes a critical difference for humanity are very few. I mean, I think, I mean, assuming we make a Mars colony, people go live there in a couple of decades. I mean, their supplies are going to come from Earth, the money to make the colony came from Earth and whatever powers are supplying the goods there from Earth are going to in effect be in control of that Mars colony. Of course, there are outliers situations where, you know, Earth gets nuked into oblivion and somehow Mars has been made self-sustaining by that point. And then Mars is what allows humanity to persist. But I think </s><s>[INST] that those are very, very, very unlikely. Do you don't think it could be a first step on a long [/INST] journey? Of course it's a first step on a long journey, which is awesome. I'm guessing the colonization of the rest of the physical universe will probably be done by AGIs that are better designed to live in space than by the meat machines that we are. But I mean, who knows, we may crowd preserve ourselves in some superior way to what we know now and like shoot ourselves out to Alpha Centauri and beyond. I mean, that's all cool. It's very interesting. And it's much more valuable than most things that humanity is spending its resources on. On the other hand, with AGI, we can get to a singularity before the Mars colony becomes sustaining for sure, </s><s>[INST] possibly before it's even operational. So your tuition is that that's the problem if we really invest resources and we can get to faster than a legitimate full self-sustaining colonization of [/INST] Mars. Yeah. And it's very clear that we will to me because there's so much economic value in getting from their AI toward AGI, whereas the Mars colony, there's less economic value until you get quite far out into the future. So I think that's very interesting. I just think it's somewhat off to the side. I mean, just as I think say, you know, art and music are very, very interesting and I want to see resources go into amazing art and music being created. And I'd rather see that than a lot of the garbage that society spends their money on. On the other hand, I don't think Mars colonization or inventing amazing new genres of music is not one of the things that is most likely to make a critical difference in the evolution of human or non-human life in this part of the universe over the next decade. Do you think AGI is really true? AGI is by far the most important thing that's on the horizon. And then technologies that have direct ability to enable AGI or to accelerate AGI are also very important. For example, say quantum computing, I don't think that's critical to achieve AGI, but certainly you could see how the right quantum computing architecture could massively accelerate AGI, similar other types of nanotechnology. Right now, now the quest to cure aging and end disease while not in the big picture as important as AGI, of course, it's important to all of us as individual humans. And if someone made a super longevity pill and distributed it tomorrow, I mean, that would be huge and a much larger impact </s><s>[INST] than a Mars colony is going to have for quite some time. But perhaps not as much as an AGI [/INST] system. No, because if you can make a benevolent AGI, then all the other problems are solved. Once it's as generally intelligent as humans, it can rapidly become massively more generally intelligent than humans. And then that AGI should be able to solve science and engineering problems much better than human beings, as long as it is in fact motivated to do so. That's why I said a </s><s>[INST] benevolent AGI. There could be other kinds. Maybe it's good to step back a little bit. I mean, we've been using the term AGI. People often cite you as the creator, or at least the popularizer of the term AGI, artificial general intelligence. Can you tell the origin story of the term? [/INST] For sure. For sure. So yeah, I would say I launched the term AGI upon the world for what it's worth without ever fully being in love with the term. What happened is I was editing a book, and this process started around 2001 or two. I think the book came out 2005, funnily. I was editing a book which I provisionally was titling real AI. And I mean, the goal was to gather together fairly serious academic-ish papers on the topic of making thinking machines that could really think in the sense like people can, or even more broadly than people can, right? So then I was reaching out to other folks that I'd encountered here or there who were interested in that, which included some other folks who I knew from the transhumist and singularitarian world, like Peter Vos, who has a company AGI Incorporated still in California, and included Shane Legge, who had worked for me at my company WebMind in New York in the late 90s, who by now has become rich and famous. He was one of the co-founders of Google DeepMind. But at that time, Shane was, I think he may have just started doing his PhD with Marcus Hutter, who at that time hadn't yet published his book, Universal AI, which sort of gives a mathematical foundation for artificial general intelligence. So I reached out to Shane and Marcus and Peter Vos and Pei Wang, who was another former employee of mine, who had been Douglas Hofstadter's PhD student, who had his own approach to AGI. And a bunch of some Russian folks reached out to these guys and they contributed papers for the book. But that was my provisional title, but I never loved it because in the end, you know, I was doing some, what we would now call narrow AI as well, like applying machine learning to genomics data or chat data for sentiment analysis. And I mean, that work is real. And then in a sense, it's really AI, it's just a different kind of AI. Ray Kurzweil wrote about narrow AI versus strong AI. But that seemed weird to me because first of all, narrow and strong are not entities. But secondly, strong AI was used in the cognitive science literature to mean the hypothesis that digital computer AIs could have true consciousness like human beings. So there was already a meaning to strong AI, which was complexly different, but related. So we were tossing around on an email list what title it should be. And so we talked about narrow AI, broad AI, wide AI, narrow AI, general AI. And I think it was either Shane Legge or Peter Vos on the private email discussion we had. He said, well, why don't we go with AGI, artificial general intelligence? And Pei Wang wanted to do GAI, general artificial intelligence, because in Chinese, it goes in that order. But we figured, gay wouldn't work in US culture at that time. So we went with AGI, we used it for the title of that book. And part of Peter and Shane's reasoning was you have the G factor in psychology, which is IQ, general intelligence. So you have a meaning of GI, general intelligence, in psychology. So then you're looking like artificial GI. So then Oh, that makes a lot of sense. Yeah, we use that for the title of the book. And so I think </s><s>[INST] Oh, that makes a lot of sense. [/INST] Yeah, we use that for I may be both Shane and Peter think they invented the term. But then later after the book was published, this guy Mark Gubrid came up to me and he's like, well, I I published an essay with the term AGI in like 1997 or something. And so I'm just waiting for some Russian to come out and say they published that in 1953. But I mean, that term is not dramatically innovative or anything. It's one of these obvious, in hindsight things, which is also for sure. It's annoying in a way because, you know, Josh Habak, who you interviewed is a close friend of mine, he likes the term synthetic intelligence, which I like much better, but it hasn't, it hasn't actually caught on, right? Because, I mean, artificial is a bit off to me because art of artifies is like a tool or something, but, but not all AGIs are going to be tools. I mean, they may be now, but we're aiming toward making them agents rather than, than tools. And in a way, I don't like the distinction between artificial and natural, because I mean, we're we're part of nature also, and machines are part of, are part of nature. I mean, you can look at evolved versus engineered, but that's a different, that's a different distinction. Then it should be engineered general intelligence, right? And then general, well, if you look at Marcus Hutter's book, Universally, what he argues there is, you know, within the domain of computation theory, which is limited, but interesting. So if you assume computable environments, the computable reward functions, then he articulates what would be a truly general intelligence, a system called AIXI, which is quite beautiful. IXC. IXC, and that's, that's the middle name of my latest child, actually, is it? What's the first name? First name is Quarksey, Q-R-X-I, which my wife came out with, but that, that's an acronym for Quantum Organized Rational Expanding Intelligence. IXC. His middle name is, his middle name is X-X-Xiphanes, actually, which is a mean, means that </s><s>[INST] the former principal underlying IXC. But in any case, what's the first name? IXC. In any case, [/INST] you're giving Elon Musk a new child to run for. Well, I did it first. He copied me with this, this new freakish name. But now if I have another baby, I'm going to have to outdo him. It's becoming an arms race of weird geeky baby names. We'll see what the babies think about it, right? Yeah. But I mean, my oldest son, Zarathustra, loves his name, and my daughter, Charizard, loves her name. So, so far, basically, if you give your kids weird names, They live up to it. Well, you're obliged to make the kids weird enough that they like the names, right? It directs their upbringing in a certain way. But yeah, anyway, I mean, I mean, what Marcus showed in that book is that a truly general intelligence, theoretically, is possible, but would take infinite computing power. So then the artificial is a little off, the general is not really achievable within physics, as we know it. And I mean, physics, </s><s>[INST] as we know it may be limited, but that's what we have to work with now. Intelligence is a new freakish name. But now, it's becoming, what you're obliged to make. Infinitely general, you mean, like, yeah, information processing perspective, yeah. [/INST] Yeah, yeah, intelligence is not very well defined either. I mean, what does it mean? I mean, in AI now, it's fashionable to look at it as maximizing and expected reward over the future. But that's, that sort of definition is pathological in various ways. And my friend, David Weinbaum, aka Weaver, he had a beautiful PhD thesis on open-ended intelligence, trying to conceive intelligence in a... Without a reward. Yeah, he's just looking at it differently. He's looking at complex self-organizing systems and looking at an intelligence system as being one that, you know, revises and grows and improves itself in conjunction with its environment without necessarily there being one objective function it's trying to maximize. Although, over certain intervals of time, it may act as if it's optimizing a certain objective function. It's very much hilarious from Stan Slalom's novels, right? So yeah, the point is, artificial, general, and intelligence... Don't work. They're all bad. On the other hand, everyone knows what AI is. And AGI seems immediately comprehensible to people with a technical background. So I think that the term has served as sociological function. Now it's out there everywhere, which... It's stuck. Which baffles me. </s><s>[INST] It's like KFC without the depth of work. Yeah, he's... They're all bad. On the other... Yeah, it's stuck. Which baffles me. It's like KFC. I mean, that's it. We're stuck with AGI probably for a very long time, until AGI systems take over and rename themselves. [/INST] Yeah, yeah. And then we'll be biological. We'll start with GPUs too, which mostly have nothing to do with graphics anymore, right? </s><s>[INST] Biological. We'll start with... I wonder what the AGI system will call us humans. That was maybe... Grandpa. GPs. Grandpa processing unit, yeah. Biological grandpa processing units. Yeah. Okay, so maybe also just a comment on AGI representing before even the term existed, representing a kind of community. You've talked about this in the past, sort of AI is coming in waves, but there's always been this community of people who dream about creating general human level super intelligent systems. Can you maybe give your sense of the history of this community, as it exists today, as it [/INST] existed before this deep learning revolution, all throughout the winters and the summers of AI? Sure. First, I would say, as a side point, the winters and summers of AI are greatly exaggerated by Americans. And if you look at the publication record of the artificial intelligence community, since, say, the 1950s, you would find a pretty steady growth and advance of ideas and papers. And what's thought of as an AI winter or summer was sort of how much money is the US military pumping into AI, which was meaningful. On the other hand, there was AI going on in Germany, UK, and in Japan, and in Russia, all over the place, while US military got more and less enthused </s><s>[INST] about AI. That happened to be, just for people who don't know, the US military happened to be the main source of funding for AI research. So another way to phrase that is it's up and down of funding [/INST] for artificial intelligence research. And I would say the correlation between funding and intellectual advance was not 100%, right? Because in Russia, as an example, or in Germany, there was less dollar funding than in the US, but many foundational ideas were laid out, but it was more theory than implementation, right? And US really excelled at sort of breaking through from theoretical papers to working implementations, which did go up and down somewhat with US military funding. But still, I mean, you can look in the 1980s, Dietrich Döner in Germany had self-driving cars on the Autobahn, right? And I mean, it was a little early with regard to the car industry, so it didn't catch on, such as has happened now. But I mean, that whole advancement of self-driving car technology in Germany was pretty much independent of AI military summers and winters in the US. So there's been more going on in AI globally than not only most people on the planet realize, but then most new AI PhDs realize because they've come up within a certain sub field of AI and haven't had to look so much beyond that. But I would say when I got my PhD in 1989 in mathematics, I was interested in AI already. In Philadelphia, but... Yeah, I started at NYU, then I transferred to Philadelphia, to Temple University, good old North Philly. Yeah, the Pearl of the US. You never stopped at a red light then, because you were afraid if you stopped at a red light, some more carjackies. So you drive through every red light. Every day driving or bicycling to Temple from my house was a new adventure. But yeah, the reason I didn't do a PhD in AI was what people were doing in the academic AI field then was just astoundingly boring and seemed wrong headed to me. It was really like rule-based expert systems and production systems. And actually I loved mathematical logic. I had nothing against logic as the cognitive engine for an AI, but the idea that you could type in the knowledge that AI would need to think seemed just completely stupid and wrong headed to me. I mean, you can use logic if you want, but somehow the system has got to be learning, right? It should be learning from experience. And the AI field then was not interested in learning from experience. I mean, some researchers certainly were. I mean, I remember in mid 80s, I discovered a book by John Andreas, which was... It was about a reinforcement learning system called PRPUS, P-U-R dash P-U-S-S, which was an acronym that I can't even remember what it was for, PRPUS anyway. But I mean, that was a system that was supposed to be an AGI. And basically by some sort of fancy like Markov decision process learning, it was supposed to learn everything just from the bits coming into it and learn to maximize its reward and become intelligent, right? So that was there in academia back then, but it was like isolated, scattered, weird people. But all these isolated, scattered, weird people in that period, I mean, they laid the intellectual grounds for what happened later. So you look at John Andreas at University of Canterbury with his PRPUS reinforcement learning Markov system. He was the PhD supervisor for John Cleary in New Zealand. Now, John Cleary worked with me when I was at Waikato University in 1993 in New Zealand. And he worked with Ian Witten there and they launched Weka, which was the first open source machine learning toolkit, which was launched in, I guess, 93 or 94 when I was at Waikato University. </s><s>[INST] Yeah, but nice feeling. Yeah, you were afraid to strap through every red light. Yeah, written in Java, unfortunately. [/INST] Written in Java, which was a cool language back then. </s><s>[INST] I guess it's still, well, it's not cool anymore, but it's powerful. [/INST] Like most programmers now, I find Java unnecessarily bloated. But back then, it was like Java or C++, basically, and Java was easier for students. Amusingly, a lot of the work on Weka when we were in New Zealand was funded by, sorry, a New Zealand government grant to use machine learning to predict the menstrual cycles of cows. So, in the US, all the grant funding for AI was about how to kill people or spy on people. In New Zealand, it's all about cows or kiwi fruits. So, yeah, anyway, I mean, John Andreas had his probability theory-based reinforcement learning, proto-AGI. John Cleary was trying to do much more ambitious, probabilistic AGI systems. Now, John Cleary helped do Weka, which is the first open source machine learning tool. Gets to the predecessor for TensorFlow and Torch and all these things. Also, Shane Legge was at Waikato working with John Cleary and Ian Witten and this whole group. And then working with my own companies, my company, WebMind, an AI company I had in the late 90s with a team there at Waikato University, which is how Shane got his head full of AGI, which led him to go on and with Jamis Asabas found deep mind. So, what you can see through that lineage is, you know, in the 80s and 70s, John Andreas was trying to build probabilistic reinforcement learning AGI systems. The technology, the computers just weren't there to support his ideas were very similar to what people are doing now. But, you know, although he's long since passed away and didn't become that famous outside of Canterbury, I mean, the lineage of ideas passed on from him to his students to their students. You can go trace directly from there to me and to deep mind, right? So, there was a lot going on in AGI that did. </s><s>[INST] Object-oriented, so it's nice. [/INST] In New Zealand, ultimately, like the groundwork for what we have today, but there wasn't a community, right? And so, when I started trying to pull together an AGI community, it was in, I guess, the early aughts when I was living in Washington, D.C. and making a living doing AI consulting for various US government agencies. And I organized the first AGI workshop in 2006. And, I mean, it wasn't like it was literally in my basement or something. I mean, it was in the conference room within Marriott in Bethesda. It's not that edgy or underground, unfortunately, but still... How many people attended? About 60 or something. That's not bad. I mean, D.C. has a lot of AI going on. Probably until the last five or ten years, much more than Silicon Valley, although it's just quiet because of the nature of what happens in D.C. Their business isn't driven by PR. Mostly, when something starts to work really well, it's taken black and becomes even more quiet, right? But, yeah, the thing is that really had the feeling of a group of starry-eyed mavericks like huddled in a basement, like plotting how to overthrow the narrow AI establishment. And, you know, for the first time, in some cases, coming together with others who share their passion for AGI and the technical seriousness about working on it, right? I mean, that's still very, very different than what we have today. I mean, now it's a little bit different. We have AGI conference every year, and there's several hundred people rather than 50. Now it's more like this is the main gathering of people who want to achieve AGI and think that large-scale, nonlinear regression is not the golden path to AGI. So I mean, it's- </s><s>[INST] AK and neural networks. Yeah, yeah, yeah, yeah, yeah, yeah, yeah, yeah. [/INST] Well, certain architectures for learning using neural networks. So, yeah, the AGI conferences are sort of now the main concentration of people not obsessed with deep neural nets and deep reinforcement learning, but still interested in AGI. Not the only ones. I mean, there's other little conferences and groupings interested in human-level AI and cognitive architectures and so forth. But yeah, it's been a big shift. Like back then, you couldn't really, it'll be very, very edgy then to give a university department seminar that mentioned AGI or human level AI. It was more like you had to talk about something more short-term and immediately practical. Then in the bar after the seminar, you could bullshit about AGI in the same breath as time travel or the simulation hypothesis or something, right? Whereas now AGI is not only in the academic seminar room, Vladimir Putin knows what AGI is. And he's like, Roshan needs to become the leader in AGI, right? So national leaders and CEOs of large corporations. I mean, the CTO of Intel, Justin Ratner, this was years ago, a Singularity Summit conference, 2008 or something. He's like, we believe very curse while the Singularity will happen in 2045 and it will have Intel inside. I mean, so it's gone from being something which is the pursuit of like crazed mavericks, crackpots and science fiction fanatics to being a marketing term for large corporations and the national leaders, which is a, it's gone for a astounding transition. But yeah, in the course of this transition, I think a bunch of sub-communities have formed. And the community around the AGI conference series is certainly one of them. It hasn't grown as big as I might've liked it to. On the other hand, you know, sometimes a modest size community can be better for making intellectual progress. Also, like you get it with society for neuroscience conference, you have 35 or 40,000 neuroscientists. On the one hand, it's amazing. On the other hand, you're not going to talk to the leaders of the field there if you're an outsider. </s><s>[INST] Yeah. In the same sense, the AAAI, the artificial intelligence, the main kind of generic artificial intelligence conference is too big. It's too amorphous. Like it doesn't make, [/INST] Nips has become a company advertising outlet in the whole. </s><s>[INST] So yeah, so I mean, to comment on the role of AGI in the research community, I'd still, if you look at NeurIPS, if you look at CVPR, if you look at these iClear, you know, AGI is still seen as the outcast. I would still, I would say in these main machine learning, in these main artificial intelligence conferences amongst the researchers, I don't know if it's an accepted term yet. What I've seen bravely, you mentioned Shane Wagg, is DeepMind and then OpenAI are the two places that are, I would say unapologetically so far. I think it's actually changing, unfortunately, [/INST] but so far, they've been pushing the idea that the goal is to create an AGI. Well, they have billions of dollars behind them. So I mean, in the public mind, that certainly carries some um right. I mean, I mean... But they also have really strong researchers, right? They do, </s><s>[INST] they're great teams, I mean, DeepMind, in particular, but... But they also have really strong researchers, right they... They do... And they have... I mean DeepMind as Markus Hutter walking around. I mean, there's all these folks who basically their full-time [/INST] position involves dreaming about creating AGI. I mean, Google Brain has a lot of amazing AGI-oriented people also. And I mean, so I'd say from a public marketing view, DeepMind and OpenAI are the two large well-funded organizations that have put the term and concept AGI out there sort of as part of their public image. But I mean, there are other groups that are doing research that seems just as AGI-ish to me. I mean, including a bunch of groups in Google's main Mountain View office. So yeah, it's true. AGI is somewhat away from the mainstream now, but if you compare it to where it was 15 years ago, there's been an amazing mainstreaming. You could say the same thing about super longevity research, which is one of my my application areas that I'm excited about. I mean, I've been talking about this since the 90s, but working on this since 2001. And back then, really to say you're trying to create therapies to allow people to live hundreds of thousands of years, you were way, way, way, way out of the industry, academic mainstream. But now, you know, Google had project Calico, Craig Venter at Human Longevity Incorporated. And then once the suits come marching in, right? I mean, once there's big money in it, then people are forced to take it seriously, because that's the way modern society works. So it's still not as mainstream as cancer research, just as AGI is not as mainstream as automated driving or something. But the degree of mainstreaming that's happened in the last 10 to 15 years is astounding to those of us who've been at it for </s><s>[INST] a while. Yeah, but there's a marketing aspect to the term. But in terms of actual full force research that's going on under the header of AGI, it's currently, I would say, dominated, maybe you can disagree, dominated by neural networks research that the nonlinear regression, as you mentioned, like, what's your sense with with OpenCog, with your work in general? So I was logic based systems and expert systems for me, always seemed to capture a deep element of intelligence that needs to be there. Like you said, it needs to learn needs to be automated somehow. But that seems to be missing from a lot of a lot of research currently. So what's your sense? I guess one way to ask this question, what's your sense of what kind of things will an AGI [/INST] system need to have? Yeah, that's a very interesting topic that I've thought about for a long time. And I think there are many, many different approaches that can work for getting to human level AI. So I don't, I don't think there's like one golden algorithm or one golden design that can work. And I mean, flying machines is the the much worn analogy here, right? Like, I mean, you have airplanes, you have helicopters, you have balloons, you have stealth bombers, it don't look like regular airplanes, you've got all blimps. Birds too. Birds, yeah, and bugs, right? Yeah. And you, I mean, and there are certainly many kinds of flying machines, birds, yeah, and bugs, right? Yeah. And there's a catapult that you can just launch. There's bicycle powered like flying machines, right? Yeah. Yeah. So now these are all analyzable by a basic theory of aerodynamics, right? Now, so one issue with AGI is we don't yet have the analog of the theory of aerodynamics. And that's, that's what Marcus Hutter was trying to make with the AXE and his general theory of general intelligence. But that theory, in its most clearly articulated parts, really only works for either infinitely powerful machines or almost or insanely impractically powerful machines. So I mean, if, if you were going to take a theory based approach to AGI, what you would do is say, well, let's, let's take what's called, say, AXE-TL, which is, which is Hutter's AXE machine that can work on merely insanely much processing power rather than infinitely much. What does TL stand for? Time and length. Okay. So you're, basically how it constrains them up. Yeah, yeah, yeah. So how AXE works basically is each, each, each action that it wants to take before taking that action, it looks at all its history. Yeah. And then it looks at all possible programs that it could use to make a decision. Yeah. And it decides like, which decision program would have let it make the best decisions according to its reward function, over its history? And he uses that decision program to take, to make the next </s><s>[INST] decision. It's not afraid is constrained. Yeah, yeah, right. It's not afraid of infinite resources [/INST] and it's searching through the space of all possible computer programs in between each action and eats next action. Now, XCTL searches through all possible computer programs that have runtime less than T and length less than L, which is still an impracticably humongous space, right? So what you would like to do to make an AGI and what will probably be done 50 years from now to make an AGI is say, okay, well, we have some constraints. We have these processing power constraints and, you know, we have space and time constraints on the program. We have energy utilization constraints and we have this particular class of environments that we care about, which may be, say, you know, manipulating physical objects on the surface of the earth, communicating in human language, I mean, whatever our particular, not annihilating humanity, right? Whatever our particular requirements happen to be, if you formalize those requirements in some formal specification language, you should then be able to run an automated program specializer on AXCTL, specialize it to the computing resource constraints and the particular environment and goal, and then it will spit out like the specialized version of AXCTL to your resource restrictions in your environment, which will be your AGI, right? And that, I think, is how our super AGI will create new AGI systems, right? But that's a very Russian approach, by the way, like the whole field of program specialization came out of Russia. </s><s>[INST] It seems really inefficient. Can you backtrack? So what is program specialization? So it's [/INST] basically... Take sorting, for example. You can have a generic program for sorting lists, but what if all your lists you care about are length 10,000 or less? Got it. You can run an automated program specializer on your sorting algorithm, and then we'll come up with the algorithm that's optimal for sorting lists of length 1,000 or 10,000 or less, right? </s><s>[INST] Got it. It's kind of like, isn't that the kind of the process of evolution is a program specializer to the environment? So you're kind of evolving human beings or... Exactly. I mean, [/INST] your Russian heritage is showing that. So Alexander Vityaev and Peter Anokin and so on. I mean, there's a long history of thinking about evolution that way also, right? So well, my point is that what we're thinking of as a human-level general intelligence, you know, if you start from narrow AIs, like are being used in the commercial AI field now, then you're thinking, okay, how do we make it more and more general? On the other hand, if you start from AXE or Schmidhuber's girdle machine or these infinitely powerful, but practically infeasible AIs, then getting to a human-level AGI is a matter of specialization. It's like, how do you take these maximally general learning processes and how do you specialize them so that they can operate within the resource constraints that you have, but will achieve the particular things that you care about? Because we humans are not maximally general intelligence, right? If I ask you to run a maze in 750 dimensions, you'll probably be very slow, whereas at two dimensions, you're probably way better, right? So, I mean, we're special because our hippocampus has a two-dimensional map in it, right? And it does not have a 750-dimensional map in it. So, I mean, we're, you know, a peculiar mix of generality and specialization, right? </s><s>[INST] We'll probably start quite general at birth, not obviously still narrow, but like more general [/INST] than we are at age 20 and 30 and 40 and 50 and 60. I don't think that I think it's more complex than that because I'm in some sense, a young child is less biased and their brain has yet to sort of crystallize into appropriate structures for processing aspects of the physical and social world. On the other hand, the young child is very tied to their sensorium, whereas we can deal with abstract mathematics like 750 dimensions and the young child cannot because they haven't grown what Piaget called the formal capabilities. They haven't learned to abstract yet, right? And the ability to abstract gives you a different kind of generality than what the baby has. So, there's both more specialization and more generalization that </s><s>[INST] comes with the development process actually. I mean, I guess just the trajectories of the specialization are most controllable at the young age, I guess is one way to put it. [/INST] Do you have kids? No. Do you have kids? No. They're not as controllable as you think. So, you think it's interesting. I think, honestly, a human adult is much more generally intelligent than a human baby. Babies are very stupid. I mean, they're cute, which is why we put up with their repetitiveness and stupidity. And they have what the Zen guys would call a beginner's mind, which is a beautiful thing, but that doesn't necessarily correlate with a high </s><s>[INST] level of intelligence. So, you think it's interesting. I mean, they're on a plot of cuteness and stupidity there. There's a process that allows us to put up with their stupidity. [/INST] Well, yeah, yeah. So, by the time you're an ugly old man like me, you gotta get really, really smart to compensate. To compensate. Okay, cool. </s><s>[INST] But yeah, going back to your original question, so get really, really smart to compensate. [/INST] And to compensate. Okay, cool. The way I look at human level AGI is, yeah, how do you specialize unrealistically inefficient super human brute force learning processes to the specific goals that humans need to achieve and the specific resources that we have. And both of these, the goals and the resources and the environments, all this is important. And on the resources side, it's important that the hardware resources we're bringing to bear are very different than the human brain. So, the way I would want to implement AGI on a bunch of neurons in a vat that I could rewire arbitrarily is quite different than the way I would want to create AGI on, say, a modern server form of CPUs and GPUs, which in turn may be quite different than the way I would want to implement AGI on whatever quantum computer we'll have in 10 years, supposing someone makes a robust quantum Turing machine or something. So, I think there's been co-evolution of the patterns of organization in the human brain and the physiological particulars of the human brain over time. And when you look at neural networks, that is one powerful class of learning algorithms, but it's also a class of learning algorithms that evolved to exploit the particulars of the human brain as a computational substrate. If you're looking at the computational substrate of a modern server farm, you won't necessarily want the same algorithms that you want on the human brain. And from the right level of abstraction, and you could look at maybe the best algorithms on the brain and the best algorithms on a modern computer network as implementing the same abstract learning and representation processes, but finding that level of abstraction is its own AGI research project then, right? So that's about the hardware side and the software side, which follows from that. Then regarding what are the requirements, I wrote the paper years ago on what I called the embodied communication prior, which was quite similar in intent to Yoshua Bengio's recent paper on the consciousness prior, except I didn't wanna wrap up consciousness in it because to me, the quell your problem and subjective experience is a very interesting issue also, which we can chat about, but I would rather keep that philosophical debate distinct from the debate of what kind of biases do you wanna put in the general intelligence to give it human-like, </s><s>[INST] general intelligence. And I'm not sure Yoshua Bengio is really addressing that kind of consciousness. [/INST] He's just using the term at that kind of consciousness. He's just using the term. I love Yoshua to pieces, like he's by far my favorite of the lines of deep learning, he's such a good-hearted guy and a creative thinker. I am not sure he has plumbed to the depths the philosophy of consciousness. </s><s>[INST] No, he's a good human being. Yeah, for sure. No, he's using it as a sexy type. [/INST] Yeah, yeah, yeah, yeah, yeah, yeah, yeah. So what I called it was the embodied communication prior. And can you maybe explain it a little bit? Yeah, yeah, what I meant was, you know, what are we humans evolved for? You can say being human, but that's very abstract, right? I mean, our minds control individual bodies, which are autonomous agents moving around in a world that's composed largely of solid objects, right? And we've also evolved to communicate via language with other, you know, solid object agents that are going around doing things collectively with us in a world of solid objects. And these things are very obvious, but if you compare them to the scope of all possible intelligences, or even all possible intelligences that are physically realizable, that actually constrains things a lot. So if you start to look at, you know, how would you realize some specialized or constrained version of universal general intelligence in a system that has, you know, limited memory and limited speed of processing, but whose general intelligence will be biased toward controlling a solid object agent, which is mobile in a solid object world, for manipulating solid objects and communicating via language with other similar agents in that same world, right? Then starting from that, you're starting to get a requirements analysis for human level general intelligence. And then that leads you into cognitive science. And you can look at, say, what are the different types of memory that the human mind and brain has? And this has matured over the last decades. And I got into this a lot in, so after I gave my PhD in math, I was an academic for eight years. I was in departments of mathematics, computer science, and psychology. When I was in the psychology department at the University of Western Australia, I was focused on cognitive science of memory and perception. Actually, I was teaching neural nets and deep neural nets, and it was multi-layered perceptrons, right? Psychology? Yeah. Cognitive science, it was cross-disciplinary among engineering, math, psychology, philosophy, linguistics, computer science. But yeah, we were teaching psychology students to try to model the data from human cognition experiments using multi-layered perceptrons, which was the early version of a deep neural network. Very, very, yeah, recurrent backprop was very, very slow to train back then, right? </s><s>[INST] I'll tell you that. So this is the study of these constraints, [/INST] systems that are supposed to deal with physical objects. Systems that are supposed to deal with physical objects. So if you look at cognitive psychology, you can see there's multiple types of memory, which are to some extent represented by different subsystems in the human brain. So we have episodic memory, which shakes into account our life history and everything that's happened to us. We have declarative or semantic memory, which is like facts and beliefs abstracted from the particular situations that they occurred in. There's sensory memory, which to some extent is sense modality specific, and to some extent is unified across sense modalities. There's procedural memory, memory of how to do stuff, like how to swing the tennis racket, right? Which is, there's motor memory, but it's also a little more abstract than motor memory. It involves cerebellum and cortex working together. Then there's memory linkage with emotion which has to do with linkages of cortex and limbic system. There's specifics of spatial and temporal modeling connected with memory, which has to do with hippocampus and thalamus connecting to cortex. And the basal ganglia, which influences goals. So we have specific memory of what goals, sub-goals and sub-sub-goals we wanted to perceive in which context in the past. Human brain has substantially different subsystems for these different types of memory and substantially differently tuned learning, like differently tuned modes of long-term potentiation to do with the types of neurons and neurotransmitters in the different parts of the brain corresponding to these different types of knowledge. And these different types of memory and learning in the human brain, I mean, you can back these all into embodied communication for controlling agents in worlds of solid objects. Now, so if you look at building an AGI system, one way to do it, which starts more from cognitive science and neuroscience is to say, okay, what are the types of memory that are necessary for this kind of world? Yeah, yeah. Necessary for this sort of intelligence. What types of learning well with these different types of memory and then how do you connect all these things together? And of course the human brain did it incrementally through evolution because each of the sub-networks of the brain, I mean, it's not really the lobes of the brain. It's the sub-networks, each of which is widely distributed. which of each of the subnetworks of the brain co-evolved with the other subnetworks of the brain, both in terms of its patterns of organization and the particulars of the neurophysiology. So they all grew up communicating and adapting to each other. It's not like they were separate black boxes that were then glommed together, right? Whereas as engineers, we would tend to say, let's make the declarative memory box here and the procedural memory box here and the perception box here and wire them together. And when you can do that, it's interesting. I mean, that's how a car is built, right? But on the other hand, that's clearly not how biological systems are made. The parts co-evolved so as to adapt and work together. </s><s>[INST] So this- That's better necessary for this kind of world. By the way, how every human engineered system that flies, that was what we were using that analogy before is built as well. So do you find this at all appealing? Like there's been a lot of really exciting which I find strange that it's ignored work in cognitive architectures, for example, throughout the last few decades. [/INST] Do you find that more interesting? Yeah, I mean, I had a lot to do with that community. And Paul Rosenbloom, who was one of the, and John Laird who built the SOAR architecture are friends of mine. And I learned SOAR quite well in ACT-R and these different cognitive architectures. And how I was looking in the AI world about 10 years ago before this whole commercial deep learning explosion was, on the one hand, you had these cognitive architecture guys who were working closely with psychologists and cognitive scientists who had thought a lot about how the different parts of a human-like mind should work together. On the other hand, you had these learning theory guys who didn't care at all about the architecture but were just thinking about like, how do you recognize patterns and large amounts of data? And in some sense, what you needed to do was to get the learning that the learning theory guys were doing and put it together with the architecture, the cognitive architecture guys were doing. And then you would have what you needed. Now, you can't, unfortunately, when you look at the details, you can't just do that without totally rebuilding what what is happening on both the cognitive architectures and the learning side. So, I mean, they tried to do that in SOAR, but what they ultimately did like taking deep neural net or something for perception and you put it as one of the black boxes. </s><s>[INST] Yeah, it becomes one of the boxes. The learning mechanism becomes one of the boxes as opposed to fundamentally. [/INST] Yeah, that doesn't quite work. You could look at some of the stuff DeepMind has like the differential neural computer or something. That sort of has a neural net for deep learning perception, it has another neural net which which is like a memory matrix. It stores, say, the map of the London subway or something. So probably Demas Sosabas was thinking about this like part of cortex and part of hippocampus, because hippocampus has a spatial map. And when he was a neuroscientist, he was doing a bunch on cortex, hippocampus interconnection. So there, the DNC would be an example of folks from the deep neural net world trying to take a step in the cognitive architecture direction by having two neural modules that correspond roughly to two different parts of the human brain that deal with different kinds of memory and learning. But on the other hand, it's super, super, super crude from the cognitive architecture view. Just as what John Laird and SOAR did with neural nets was super, super crude from a learning point of view. Because the learning was like off to the side not affecting the core representations. I mean, you weren't learning the representation. You were learning the data that feeds into the TD. You were learning abstractions of perceptual data to feed into the representation that wasn't was not learned. So, yeah, this was clear to me a while ago, and one of my hopes with the AGI community was to sort of bring people from those two directions together. That didn't happen much in terms of... Not yet. And what I was gonna say is it didn't happen in terms of bringing like the lines of cognitive architecture together with the lines of deep learning. It did work in the sense that a bunch of younger researchers have had their heads filled with both of those ideas. This comes back to saying my dad, who was a university professor, often quoted to me, which was, science advances one funeral at a time. Which I'm trying to avoid, like I'm 53 years old, and I'm trying to invent amazing, weird-ass new things that nobody ever thought about, which we'll talk about in a few minutes. But there is that aspect, right? Like the people who've been at AI a long time and have made their career at developing one aspect, like a cognitive architecture or a deep learning approach, it can be hard once you're old and have made your career doing one thing. It can be hard to mentally shift gears. </s><s>[INST] I mean, I try quite hard to remain quite flexible minded. Have you been successful somewhat in changing, maybe have you changed your mind on some aspects of what it takes to build an AGI, [/INST] like tech technical things, tech technical things? The hard part is that the world doesn't want you to. The world or your own brain? No, the world, well, that one point is that your brain doesn't want to. The other part is that the world doesn't want you to. Like the people who have followed your ideas get mad at you if you change your mind. And, you know, the media wants to pigeonhole you as an avatar of a certain idea, but yeah, I've changed my mind on a bunch of things. I mean, when I started my career, I really thought quantum computing would be necessary for AGI. And I doubt it's necessary now, although I think it will be a super major enhancement. </s><s>[INST] But I mean, I'm also- The world or your own brain? [/INST] If you change. I'm now in the middle of embarking on the complete rethink and rewrite from scratch of our OpenCog AGI system, together with Alexei Podipov and his team in St. Petersburg, who's working with me in singularity net. So now we're trying to like go back to basics, take everything we learned from working with the current OpenCog system, take everything everybody else has learned from working with their proto AGI systems and design the best framework for the next stage. And I do think there's a lot to be learned from the recent successes with deep neural nets and deep reinforcement systems. I mean, people made these essentially trivial systems work much better than I thought they would. And there's a lot to be learned from that. And I wanna incorporate that knowledge appropriately in our OpenCog 2.0 system. On the other hand, I also think current deep neural net architectures as such will never get you anywhere near AGI. So I think you wanna avoid the pathology of throwing the baby out with the bathwater and like saying, well, these things are garbage because foolish journalists over blow them as being the path to AGI and a few researchers over blow them as well. Yeah. There's a lot of interesting stuff to be learned there </s><s>[INST] even though those are not the golden path. So maybe this is a good chance to step back. [/INST] You mentioned OpenCog 2.0, but- Go back to OpenCog 0.0, which exists now. Alpha. </s><s>[INST] Yeah. Yeah, maybe talk to the history of OpenCog and you're thinking about these ideas. OpenCog and you're thinking about these ideas. [/INST] I would say OpenCog 2.0 is a term we're throwing around sort of tongue-in-cheek because the existing OpenCog system that we're working on now is not remotely close to what we'd consider a 1.0, right? I mean, it's a very, very, very, very, very important and it's been around, what, 13 years or something but it's still an early stage research system, right? And actually we are going back to the beginning in terms of theory and implementation because we feel like that's the right thing to do but I'm sure what we end up with is gonna have a huge amount in common with the current system. I mean, we all still like the general approach. So first of all, what is OpenCog? Sure, OpenCog is an open source software project that I launched together with several others in 2008 and probably the first code that was written in 2001 or two or something that was developed as a proprietary code base within my AI company, Novamente LLC and we decided to open source it in 2008, cleaned up the code throughout some things, added some new things and- Well, language is it written in? It's C++, primarily there's a bunch of scheme as well </s><s>[INST] but most of it's C++. OpenCog. Well, language is it written in? And it's separate from something we'll also talk about is singularity net. [/INST] So it was born as a non-networked thing. Yeah, yeah, it was correct. Correct, well, there are many levels of networks involved here, right? </s><s>[INST] No connectivity to the internet or no. [/INST] No connectivity to the internet or no, at birth. Yeah, I mean, singularity net is a separate project and a separate body of code and you can use singularity net as part of the infrastructure for a distributed OpenCog system but there are different layers. Yeah. Got it. OpenCog, on the one hand as a software framework could be used to implement a variety of different AI architectures and algorithms. But in practice, there's been a group of developers which I've been leading together with Linus Wepstas, Neil Geisweiler and a few others which have been using the OpenCog platform and infrastructure to implement certain ideas about how to make an AGI. So there's been a little bit of ambiguity about OpenCog, the software platform versus OpenCog, the AGI design. Because in theory, you could use that software to do, you could use it to make a neural net. </s><s>[INST] You could use it to make a lot of different AGIs. What kind of stuff does the software platform provide? [/INST] Like in terms of utilities, those like what? Let me first tell about OpenCog as a software platform and then I'll tell you the specific AGI R&D we've been building on top of it. So the core component of OpenCog as a software platform is what we call the ADAMSPACE which is a weighted labeled hypergraph. ATOM, ADAMSPACE. ADAMSPACE, yeah, not ADAM like Adam and Eve, although that would be cool too. Yeah, so you have a hypergraph which is like, so a graph in this sense is a bunch of nodes with links between them. A hypergraph is like a graph, but links can go between more than two nodes. So you have a link between three nodes. And in fact, OpenCog's ADAMSPACE would properly be called a metagraph because you can have links pointing to links or you could have links pointing to whole subgraphs, right? So it's an extended hypergraph or a metagraph. Is metagraph a technical term? It is now a technical term. Interesting. But I don't think it was yet a technical term when we started calling this a generalized hypergraph. But in any case, it's a weighted labeled generalized hypergraph or weighted labeled metagraph. The weights and labels mean that the nodes and links can have numbers and symbols attached to them. So they can have types on them. They can have numbers on them that represent, say a truth value or an importance value </s><s>[INST] For a certain purpose, that would be cool too. And off course like with all things you can reduce that to a hypergraph and then a hypergraph [/INST] can be reduced to a graph and you can reduce a graph to an adjacency matrix. </s><s>[INST] So there is always multiple representations. [/INST] But there is a layer of representation that seems to work well here… Got it, right? Right right, so similarly, you could have a link to a whole graph because a whole graph could represent say a body of information, and I could say I reject this body of information then one way to do that is make that link go to that whole sub graph representing the body of information, right? I mean, there's many alternate representations, but that's anyway what we have in OpenCog, we have an atom space which is this weighted labeled generalized hypergraph. Knowledge Store, it lives in RAM. There's also a way to back it up to disk. There are ways to spread it among multiple different machines. Then there are various utilities for dealing with that. So there's a pattern matcher, which lets you specify a sort of abstract pattern and then search through a whole atom space, with labeled hypergraph, to see what sub-hypergraphs may match that pattern, for an example. So then there's something called the COG server in Open COG, which lets you run a bunch of different agents or processes in a scheduler. And each of these agents, basically, it reads stuff from the atom space, and it writes stuff to the atom space. So this is sort of the basic operational model. That's the software framework. And of course, there's a lot there, </s><s>[INST] just from a scalable software engineering standpoint. That's the software framework. So you could use this, I don't know if you've, have you looked into the Stephen Wolfram's physics project recently with the hypergraphs and stuff? [/INST] Could you theoretically use the software framework to play? Yeah, you certainly could, although Wolfram would rather die than use anything but Mathematica for his work. </s><s>[INST] Well, yeah, but there's a big community of people who would love integration. And like you said, the young minds love the idea of integrating, of connecting things. [/INST] Yeah, that's right. And I would add on that note, the idea of using hypergraph type models in physics is not very new. Like if you look at the Russians did it first. Well, I'm sure they did. And a guy named Ben Dribus, who's a mathematician, a professor in Louisiana or somewhere, had a beautiful book on quantum sets and hypergraphs and algebraic typology for discrete models of physics and then carried it much farther than Wolfram has. But he's not rich and famous, so it didn't get in the headlines. But yeah, Wolfram aside, yeah, certainly that's a good way to put it. The whole OpenCog framework, you could use it to model biological networks and simulate biology processes. You could use it to model physics on discrete graph models of physics. So you could use it to do, say, biologically realistic neural networks, for example. </s><s>[INST] And so that's a framework. What do agents and processes do? Do they grow the graph? [/INST] What kind of computations just to get a sense? In theory, they could do anything they want to do. They're just C++ processes. On the other hand, the computation framework is sort of designed for agents where most of their processing time is taken up with reads and writes to the atom space. And so that's a very different processing model than, say, the matrix multiplication-based model as underlies most deep learning systems. So you could create an agent that just factored numbers for a billion years. It would run within the OpenCog platform, but it would be pointless. I mean, the point of doing OpenCog is because you want to make agents that are cooperating via reading and writing into this weighted labeled hypergraph. And that has both cognitive architecture importance, because then this hypergraph is being used as a sort of shared memory among different cognitive processes. But it also has software and hardware implementation implications, because current GPU architectures are not so useful for OpenCog, whereas a graph chip would be incredibly useful. And I think graph core has those now, but they're not ideally suited for this. But I think in the next, let's say, three to five years, we're going to see new chips where a graph is put on the chip. And the back and forth between multiple processes acting SIMD and MIMD on that graph is going to be fast. And then that may do for OpenCog type architectures </s><s>[INST] what GPUs did for deep neural architecture. As a small tangent, can you comment on thoughts about neuromorphic computing? So like hardware implementations of all these different kind [/INST] of, are you excited by that possibility? I'm excited by graph processors, because I think they can massively speed up OpenCog, which is a class of architectures that I'm working on. I think if, you know, because I think they speed up OpenCog. In principle, neuromorphic computing should be amazing. I haven't yet been fully sold on any of the systems that are out there, like memristors should be amazing too, right? So a lot of these things have obvious potential, but I haven't yet put my hands on the system that seemed to manifest that. </s><s>[INST] Mark systems, yeah, mark systems should be amazing, [/INST] but the current systems have not been great. I mean, look, for example, if you wanted to make a biologically realistic hardware neural network, like making a circuit in hardware that emulated like the Hodgkin-Huxley equation or the Isakovich equation, like differential equations for biologically realistic neuron, and putting that in hardware on the chip, that would seem that it would make more feasible to make a large-scale truly biologically realistic neural network. So what's been done so far is not like that. So I guess, personally, as a researcher, I mean, I've done a bunch of work in computational neuroscience, where I did some work with IARPA in DC Intelligence Advanced Research Project Agency. We were looking at how do you make a biologically realistic simulation of seven different parts of the brain cooperating with each other, using realistic non-linear dynamical models of neurons, and how do you get that to simulate what's going on in the mind of a geoint intelligence analyst while they're trying to find terrorists on a map, right? So if you want to do something like that, having neurometric hardware that really let you simulate a realistic model of the neuron would be amazing. But that's sort of with my computational neuroscience hat on, with an AGI hat on, I'm just more interested in these hypergraph knowledge representation-based architectures, which would benefit more from various types of graph processors. Because the main processing bottleneck is reading writing to RAM. It's reading writing to the graph in RAM. The main processing bottleneck for this kind of proto AGI architecture is not multiplying matrices. And for that reason, GPUs, which are really good at multiplying matrices, don't apply as well. There are frameworks like Gunrock and others that try to boil down graph processing to matrix operations. And they're cool, but you're still putting a square peg into a round hole in a certain way. The same is true, I mean, current quantum machine learning, which is very cool. It's also all about how to get matrix and vector operations in quantum mechanics. And I see why that's natural to do. I mean, quantum mechanics is all unitary matrices and vectors. On the other hand, you could also try to make graph-centric quantum computers, which I think is where things will go. And then we can take the open cog implementation layer, implement it on collapsed state inside a quantum computer. But that may be the singularity squared. I'm not sure we need that to get to human level. </s><s>[INST] Singularity squared, that's singularity squared. That's already beyond the first singularity. But can we just? Yeah, let's go back to open cog. No, no. Yeah, and the hypergraph and open cog. [/INST] No, no, open cog. That's the software framework, right? So then the next thing is our cognitive architecture tells us particular algorithms to put there. </s><s>[INST] Got it. Can we backtrack on the kind of, is this graph designed, is it in general supposed to be sparse and the operations constantly grow and change the graph? Yeah, the graph is sparse. But is it constantly adding links and so on to graph? [/INST] Yeah, the graph is sparse. Links and so on? </s><s>[INST] It's a self-modifying hypergraph. So it's not the right and read operations you're referring to. This isn't just a fixed graph to which you change the way. No, no, no. [/INST] It's a disgusted growing graph. No, no, no. similar disgusted growing graph. Yeah, that's true. But it is different model than say current deep neural nets, and they have a fixed neural architecture, and you're updating the weights, although there have been cascade correlational neural that architectures that grow new nodes and links. But the most common neural architectures now have a fixed neural architecture. You're updating the weights. And in OpenCog, you can update the weights. And that certainly happens a lot. But adding new nodes, adding new links, Removing nodes and links is an equally critical part of the systems operations. </s><s>[INST] Got it, got it. So now when you start to add these cognitive algorithms on top of this open cog architecture, what does that look like? [/INST] So what? Yeah, so within this framework then, creating a cognitive architecture is basically two things. It's choosing what type system you wanna put on the nodes and links in the hypergraph, what types of nodes and links you want. And then it's choosing what collection of agents, what collection of AI algorithms or processes are gonna run to operate on this hypergraph. And of course those two decisions are closely connected to each other. So in terms of the type system, there are some links that are more neural net like, they're just like have weights to get updated by heavy and learning and activation spreads along them. They're other links that are more logic like and nodes that are more logic like. So you could have a variable node and you can have a node representing a universal or existential quantifier as in predicate logic or term logic. So you can have logic like nodes and links or you can have neural like nodes and links. You can also have procedure like nodes and links as in say a combinatorial logic or Lambda calculus representing programs. So you can have nodes and links representing many different types of semantics, which means you could make a horrible ugly mess, or you could make a system where these different types of knowledge all interpenetrate and synergize </s><s>[INST] with each other beautifully, right? [/INST] So the hypergraph can contain programs? Yeah, it can contain programs. Although in the current version, it is a very inefficient way to guide the execution of programs, which is one thing that we are aiming to resolve </s><s>[INST] with our rewrite of the system now. So what to you is the most beautiful aspect of OpenCog? Just to you personally, some aspect that captivates your imagination from beauty or power? [/INST] What fascinates me is finding a common representation that underlies abstract declarative knowledge and sensory knowledge and movement knowledge and procedural knowledge and episodic knowledge, finding the right level of representation where all these types of knowledge are stored in a sort of universal and interconvertible, yet practically manipulable way, right? So to me, that's the core. Because once you've done that, then the different learning algorithms can help each other out. Like what you want is if you have a logic engine that helps with declarative knowledge and you have a deep neural net that gathers perceptual knowledge and you have, say, an evolutionary learning system that learns procedures, you want these to not only interact on the level of sharing results and passing inputs and outputs to each other, you want the logic engine, when it gets stuck, to be able to share its intermediate state with the neural net and with the evolutionary learning algorithm so that they can help each other out of bottlenecks and help each other solve combinatorial explosions by intervening inside each other's cognitive processes. But that can only be done if the intermediate state of a logic engine, the evolutionary learning engine and a deep neural net are represented in the same form. And that's what we figured out how to do by putting the right type system </s><s>[INST] on top of this weighted labeled hypergraph. So is there, can you maybe elaborate on what are the different characteristics of a type system that can coexist amongst all these different kinds of knowledge that needs to be represented? And is, I mean, like, is it hierarchical? Just any kind of insights you can give on that kind of type system? [/INST] Yeah, yeah, yeah, yeah. So this gets very nitty gritty and mathematical, of course. But one key part is switching from predicate logic to term logic. </s><s>[INST] What is predicate logic? What is term logic? [/INST] So term logic was invented by Aristotle, or at least that's the oldest recollection we have of it. But term logic breaks down basic logic into basically simple links between nodes, like an inheritance link between node A and node B. Node A and node B. So in term logic, the basic deduction operation is A implies B, B implies C, therefore A implies C. Whereas in predicate logic, the basic operation is modus ponens. It's like A, A implies B, therefore B. So it's a slightly different way of breaking down logic. But by breaking down logic into term logic, you get a nice way of breaking logic down into nodes and links. So your concepts can become nodes, the logical relations become links. And so then inference is like, so if this link is A implies B, this link is B implies C, then deduction builds a link A implies C. And your probabilistic algorithm can assign a certain weight there. Now, you may also have a heavy and neural link from A to C, which is the degree to which A being the focus of attention should make B the focus of attention. So you could have then a neural link, and you could have a symbolic logical inheritance link in your term logic. And they have separate meaning, but they could be used to guide each other as well. If there is a large amount of neural weight on the link between A and B, that may direct your logic engine to think about, well, what is the relation? Are they similar? Is there an inheritance relation? Are they similar in some context? On the other hand, if there's a logical relation between A and B, that may direct your neural component to think, well, when I'm thinking about A, should I be directing some attention to B also? Because there's a logical relation. So in terms of logic, there's a lot of thought that went into how do you break down logic relations, including basic sort of propositional logic relations, as Aristotelian term logic deals with, and then quantifier logic relations also. How do you break those down elegantly into a hypergraph? Because I mean, you can boil logic expression into a graph in many different ways. Many of them are very ugly, right? We tried to find elegant ways of sort of hierarchically breaking down complex logic expression into nodes and links so that if you have, say, different nodes representing Ben, AI, Lex, interview, or whatever, the logic relations between those things are compact in the node and link representation. So that when you have a neural net acting on those same nodes and links, the neural net </s><s>[INST] and the logic engine can sort of interoperate with each other. And also interpretable by humans? Is that an important? [/INST] That's tough. In simple cases, it's interpretable by humans. But honestly, I would say logic systems give more potential for transparency and comprehensibility than neural net systems. But you still have to work at it because, I mean, if I show you a predicate logic proposition with like 500 nested universal and existential quantifiers and 217 variables, that's no more comprehensible than the weight matrix of a neural network, right? So I'd say the logic expressions that an AI learns from its experience are mostly totally opaque to human beings and maybe even harder to understand than a neural net. Because, I mean, when you have multiple nested quantifier bindings, it's a very high level of abstraction. There is a difference, though, in that within logic, it's a little more straightforward to pose the problem of normalize this and boil this down to a certain form. I mean, you can do that in neural nets, too. You can distill a neural net to a simpler form. But that's more often done to make a neural net that'll run on an embedded device or something. It's harder to distill a net to a comprehensible form than is to simplify a logic expression to a comprehensible form. But it doesn't come for free. Like, what's in the AI's mind is incomprehensible to a human unless you do some special work to make it comprehensible. So on the procedural side, there's some different and sort of interesting voodoo there. I mean, if you're familiar in computer science, there's something called the Curry-Howard correspondence, which is a one-to-one mapping between proofs and programs. So every program can be mapped into a proof. Every proof can be mapped into a program. You can model this using category theory and a bunch of nice math. But we want to make that practical, right? So that if you have an executable program that moves a robot's arm or figures out in what order to say things in a dialogue, that's a procedure represented in OpenCog's hypergraph. But if you want to reason on how to improve that procedure, you need to map that procedure into logic using Curry-Howard isomorphism, so that the logic engine can reason about how to improve that procedure and then map that back into the procedural representation that is efficient for execution. So again, that comes down to not just can you make your procedure into a bunch of nodes and links. Because I mean, that can be done trivially. a C++ compiler has nodes and links inside it, can you boil down your procedure into a bunch of nodes and links in a way that's like hierarchically decomposed and simple enough? You can reason about it. Yeah, yeah. Given the resource constraints at hand, you can map it back and forth to your term logic like fast enough and without having a bloated logic expression, right? </s><s>[INST] So there's just a lot of that, like, what's what you can reason about. [/INST] Yeah, yeah. There's a lot of nitty gritty particulars there. I mean, by the same token, if you if you ask a chip designer, like, how do you make the Intel i7 chip so good, there's a there's a long list of technical answers there, which will take take a while to go through, right? And this has been decades of work. I mean, the first AI system of this nature I tried to build was called WebMind in the mid 1990s. And we had a big graph, a big graph operating in RAM implemented with Java 1.1, which is a terrible, terrible implementation idea. And then each each node had its own processing. So like that there, the core loop loop through all nodes in the network and let each node enact what it what its little thing was doing. And we had logic and neural nets in there, but an evolutionary learning, but we hadn't done enough of the math to get them to operate together very cleanly. So it was really, it was quite a horrible mess. So as, as well as shifting an implementation, where the graph is its own object, and the agents are separately scheduled, we've also done a lot of work on how do you represent programs? How do you represent procedures? You know, how do you represent genotypes for evolution, in a way that the interoperability between the different types of learning associated with these different types of knowledge actually works. And that's been quite difficult. It's taken decades, and it's totally off to the side of what the commercial mainstream of the of the AI field is doing, which isn't thinking about representation at all, really. Although you could see, like in the DNC, they had to think a little bit about how do you make representation of a map in this memory matrix work together with the representation needed for say visual pattern recognition in the hierarchical neural network. So I would say we have taken that direction of taking the types of knowledge you need for different types of learning, like declarative, procedural, attentional. And how do you make these types of knowledge represent in a way that allows cross learning across these different types of memory? We've been prototyping and experimenting with this within OpenCog, and before that web mind since the mid, mid 1990s. Now, disappointingly, to all of us, this has not yet been cashed out in an AGI system, right? I mean, we've used this system within our consulting business. So we've built natural language processing and robot control and financial analysis. We built a bunch of sort of vertical market specific proprietary AI projects that use OpenCog on the back end, but we haven't. That's not the AGI goal, right? It's interesting, but it's not the AGI goal. So now what we're looking at with our rebuild of the system. 2.0. Yeah, we're also calling it true AGI. So we're not quite sure what the name is yet. We made a website for 2agi.io, but we haven't put anything on there yet. </s><s>[INST] We may come up with an even better name, but. [/INST] It's kind of like the real AI starting point for your AGI, but a starting point for your AGI. It's like true better because true has like, you can be true hearted, right? You can be true to your girlfriend. So true has true has a number and it also has logic in it, right? Because logic is a key point. I like it. Yeah. So, yeah, with the true AGI system, we're sticking with the same basic architecture, but we're trying to build on what we've learned. One thing we've learned is that, you know, we need type checking among dependent types to be much faster and among probabilistic dependent types to be much faster. So as it is now, you can have complex types on the nodes and links, but if you want to put like, if you want types to be first class citizens so that you can have the types can be variables and then you do type checking among complex, higher order types, you can do that in the system now. It's very slow. This is stuff like is done in cutting-edge program languages like Agda or something, these obscure research languages. On the other hand, we've been doing a lot, tying together deep neural nets with symbolic learning. So we did a project for Cisco, for example, which was on, this was street scene analysis, but they had deep neural models for a bunch of cameras watching street scenes, but they trained a different model for each camera because they couldn't get the transfer learning to work between camera A and camera B. So we took what came out of all the deep neural models for the different cameras. We fed it into an OpenCog symbolic representation. Then we did some pattern mining and some reasoning on what came out of all the different cameras within the symbolic graph. And that worked well for that application. I mean, Hugo Latapie from Cisco gave a talk, touching on that at last year's AGI conference. It was in Shenzhen. On the other hand, we learned from there, it was kind of clunky to get the deep neural models to work well with the symbolic system because we were using Torch and Torch keeps a sort of state computation graph, but you needed like real time access to that computation graph within our hypergraph. And then we certainly did it. Alexei Polipov who leads our St. Petersburg team wrote a great paper on cognitive modules in OpenCog explaining sort of how do you deal with the Torch compute graph inside OpenCog. But in the end, we realized like that just hadn't been one of our design thoughts when we built OpenCog, right? So between wanting really fast dependent type checking and wanting much more efficient inner operation between the computation graphs of deep neural net frameworks and OpenCog's hypergraph, and adding on top of that, wanting to more effectively run an OpenCog hypergraph distributed across RAM and 10,000 machines, which is, we're doing dozens of machines now, but it's just not, we didn't architect it with that sort of modern scalability in mind. So these performance requirements are what have driven us to want to re-architect the base. But the core AGI paradigm doesn't really change. Like the mathematics is the same. It's just, we can't scale to the level that we want in terms of distributed processing or speed of various kinds of processing with the current infrastructure that was built in the phase 2001 to 2008, which is hardly shocking, right? </s><s>[INST] Well, I mean, the three things you mentioned are really interesting. So what do you think about in terms of interoperability communicating with computational graph of neural networks? What do you think about the representations [/INST] that neural networks form? Um, they're bad, but there's many ways that you could deal with that. So I've been wrestling with this a lot in some work on unsupervised grammar induction. And I have a simple paper on that that I'll give at the next AGI conference, </s><s>[INST] the online portion of which is next week, actually. [/INST] What is grammar induction? So this isn't AGI either, but it's sort of on the verge between narrow AI and AGI or something. Unsupervised grammar induction is the problem. Throw your AI system, a huge body of text, and have it learn the grammar of the language that produced that text. So you're not giving it labeled examples. So you're not giving it like a thousand sentences where the parses were marked up by graduate students. So it's just got to infer the grammar from the text. It's like the Rosetta Stone, but worse, right? Because you only have the one language, right? And you have to figure out what is the grammar. So that's not really AGI because, I mean, the way a human learns language is not that, right? I mean, we learn from language that's used in context. So it's a social embodied thing. We see how a given sentence is grounded in observation. </s><s>[INST] There's an interactive element, I guess. [/INST] Yeah, yeah, yeah, yeah, yeah, yeah. On the other hand, so I'm more interested in that. I'm more interested in making an AGI system learn language from its social and embodied experience. On the other hand, that's also more of a pain to do. And that would lead us into Hanson Robotics and their robotics work I've known, which we'll talk about in a few minutes. But just as an intellectual exercise, as a learning exercise, trying to learn grammar from a corpus is very, very interesting, right? And that's been a field in AI for a long time. No one can do it very well. So we've been looking at transformer neural networks and tree transformers, which are amazing. These came out of Google Brain, actually. And actually on that team was Lucas Kaiser, who used to work for me and won the period 2005 through eight or something. So it's been fun to see my former sort of AGI employees disperse and do all these amazing things. Way too many sucked into Google, actually. Well, yeah, anyway. We'll talk about that, too. Lucas Kaiser and a bunch of these guys, they create transformer networks, that classic paper like attention is all you need and all these things following on from that. So we're looking at transformer networks and like these are able to, I mean, this is what underlies GPT-2 and GPT-3 and so on, which are very, very cool and have absolutely no cognitive understanding of any of the texts we're looking at. Like they're very intelligent idiots, right? </s><s>[INST] But anyway. We'll talk about that, too. So sorry to take, but it's a small, I'll bring this back, but do you think GPT-3 understands? [/INST] No, no, no, it understands nothing. It's a complete idiot. </s><s>[INST] But it's a brilliant idiot, but it's a brilliant idiot. [/INST] You don't think GPT-20 will understand what you're saying? No, no, no, no, no. </s><s>[INST] What size is this, what size is this not gonna buy you? [/INST] Understanding, you understanding. And any more than a faster car is gonna get you to Mars. It's a completely different kind of thing. I mean, these networks are very cool. And as an entrepreneur, I can see many highly valuable uses for them. And as an artist, I love them, right? So I mean, we're using our own neural model, which is along those lines to control the Philip K. Dick robot now. And it's amazing to like train a neural model on the robot Philip K. Dick and see it come up with like crazed, stoned philosopher pronouncements very much like what Philip K. Dick might've said, right? Like these models are super cool. And I'm working with Hanson Robotics now on using a similar but more sophisticated one for Sofia, which we haven't launched yet, but so I think it's cool. But no, these are- But it's not understanding. They're recognizing a large number of shallow patterns. They're not forming an abstract representation. And that's the point I was coming to when we're looking at grammar induction, we tried to mine patterns out of the structure, </s><s>[INST] the transformer network. [/INST] But it's not understanding. And you can, but the patterns aren't what you want. They're nasty. So I mean, if you do supervised learning, if you look at sentences where you know the correct parts of a sentence, you can learn a matrix that maps between the internal representation of the transformer and the parts of the sentence. And so then you can actually train something that will output the sentence parts from the transformer network's internal state. And we did this, I think, Christopher Manning, some others have not done this also. But I mean, what you get is that the representation is horribly ugly and is scattered all over the network and doesn't look like the rules of grammar that you know are the right rules of grammar. It's kind of ugly. So what we're actually doing is we're using a symbolic grammar learning algorithm, but we're using the transformer neural network as a sentence probability oracle. So like if you have a rule of grammar and you weren't sure if it's the correct rule of grammar or not, you can generate a bunch of sentences using that rule of grammar and a bunch of sentences violating that rule of grammar. And you can see the transformer model doesn't think the sense is obeying the rule of grammar or more probable than the sense is disobeying the rule of grammar. So in that way you can use a neural model as a sense probability oracle to guide symbolic grammar learning process. That seems to work better than trying to milk the grammar out of the neural network that doesn't have it in there. So I think the thing is, these neural nets are not getting a semantically meaningful representation internally by and large. So one line of research is to try to get them to do that. And Infogam was trying to do that. So like if you look back like two years ago, there was all these papers like Edward, this probabilistic programming neural net framework that Google had, which came out of Infogam. So the idea there was like, you could train an Infogam neural net model, which is a generative associative network to recognize and generate faces. And the model would automatically learn a variable for how long the nose is. And automatically learned a variable for how wide the eyes are or how big the lips are or something, right? So it automatically learn these variables, which have a semantic meaning. So that was a rare case where a neural net trained with a fairly standard GAN method was able to actually learn a semantic representation. So for many years, many of us tried to take that the next step and get a GAN type neural network that would have not just a list of semantic latent variables But we'd have, say, a Bayes net of semaglutin variables with dependencies between them. The whole programming framework, Edward, was made for that. I mean, no one got it to work, right? And it could be- Do you think it's possible? Yeah, do you think- I don't know. It might be that backpropagation just won't work for it because the gradients are too screwed up. Maybe you could get it to work using CMAES or some like floating point evolutionary algorithm. We tried, we didn't get it to work. Eventually, we just paused that rather than gave it up. We paused that and said, well, okay, let's try more innovative ways to learn, to learn what are the representations implicit in that network without trying to make it grow inside that network. </s><s>[INST] And I described how we're doing that in, you know, and that seems to be variables with dependencies between them. Do you think it's possible? [/INST] Yeah, do you think- I don't like, you can do similar things in vision, right? So what- Use it as an oracle. Yeah, yeah, yeah. So you can, that's one way is that you use a structure learning algorithm, which is symbolic. And then you use the deep neural net as an oracle to guide the structure learning algorithm. And the other way to do it is like Infogam was trying to do and try to tweak the neural network to have this symbolic representation inside it. I tend to think what the brain is doing is more like using the deep neural net type thing as an oracle. Like I think the visual cortex or the cerebellum are probably learning a non-semantically meaningful opaque tangled representation. And then when they interface with the more cognitive parts of the cortex, the cortex is sort of using those as an oracle and learning the abstract representation. So if you do sports, say, take for example serving in tennis, right? I mean, my tennis serve is okay, not great, but I learned it by trial and error, right? And I mean, I learned music by trial and error, too. I just sit down and play. But then if you're an athlete, which I'm not a good athlete, I mean, then you'll watch videos of yourself serving and your coach will help you think about what you're doing. And you'll then form a declarative representation. But your cerebellum maybe didn't have a declarative representation. Same way with music, like I will hear something in my head. I'll sit down and play the thing like I heard it. And then I will try to study what my fingers did to see like, what did you just play? How did you do that, right? Because if you're composing, you may want to see how you did it and then declaratively morph that in some way that your fingers wouldn't think of, right? But the physiological movement may come out of some opaque, like, cerebellar reinforcement learning thing, right? And so that's, I think, trying to milk the structure of a neural net by treating it as an oracle, maybe more like how your declarative mind post processes, what your visual or motor cortex. I mean, in vision, it's the same way. Like, you can recognize beautiful art. </s><s>[INST] Use it as an oracle. [/INST] Much better than you can say why you think that piece of art is beautiful. But if you're trained as an art critic, you do learn to say why. And some of it is bullshit, but some of it isn't, right? Some of it is learning to map sensory knowledge into declarative and linguistic knowledge. Yet without necessarily making the sensory system itself use a transparent and easily communicable representation. </s><s>[INST] Yeah, that's fascinating. To think of neural networks as like dumb question answers that you can just milk to build up a knowledge base. And there could be multiple networks, I suppose, from different... [/INST] Yeah, yeah, yeah. So I think if a group like DeepMind or OpenAI were to build AGI, and I think DeepMind is like a thousand times more likely from what I could tell. Because they've hired a lot of people with broad minds and many different approaches and angles on AGI, whereas OpenAI is also awesome. But I see them as more of like a pure deep reinforcement learning shop. Yeah, this time, I got you. So far. </s><s>[INST] Yeah, this time, I got you. So far. Yeah, there's a lot of... You're right. I mean, there's so much interdisciplinary work at DeepMind, like neural sciences at the core. [/INST] And you put that together with Google Brain, which granted they're not working that closely together now. But my oldest son Zarathustra is doing his PhD in machine learning applied to automated theorem proving in Prague under Joseph Erban. So the first paper, DeepMath, which applied deep neural nets to guide theorem proving was out of Google Brain. I mean, by now, the automated theorem proving community has gone way, way, way beyond anything Google was doing. But still, yeah, but anyway, if that community was going to make an AGI, probably one way they would do it was, you know, take 25 different neural modules architected in different ways, maybe resembling different parts of the brain. Like a basal ganglia model, cerebellum model, a thalamus model, a few hippocampus models, a number of different models representing parts of the cortex. Take all of these and then wire them together to co-train and learn them together like that. That would be an approach to creating an AGI. One could implement something like that efficiently on top of our true AGI, like OpenCog 2.0 system once it exists. Although, obviously, Google has their own highly efficient implementation architecture. So I think that's a decent way to build AGI. I was very interested in that in the mid-90s. But I mean, the knowledge about how the brain works sort of pissed me off. Like it wasn't there yet. Like, you know, in the hippocampus, you have these concept neurons, like the so-called grandmother neuron, which everyone laughed at, it's actually there. Like I have some Lex Friedman neurons that fired differentially when I see you and not when I see any other person, right? Yeah. So how do these Lex Friedman neurons, how do they coordinate with the distributed representation of Lex Friedman they have in my cortex, right? There's some back-and-forth between cortex and hippocampus that lets these discrete symbolic representations in hippocampus correlate, and cooperate with the distributed representations in cortex. This probably has to do with how the brain does its version of abstraction and quantifier logic, right? Like, you can have a single neuron and hippocampus that activates a whole distributed activation pattern in cortex, well, this may be how the brain does, like, symbolization and abstraction as in functional programming or something. But we can't measure it, like, we don't have enough electrodes stuck between the cortex and the hippocampus in any known experiment to measure it. So I got frustrated with that direction, not because it's impossible, because it's a valid research direction. You can try to understand more and more. And we are measuring more and more about what happens in the brain now than ever before. So it's quite interesting. On the other hand, I sort of got more of an engineering mindset about AGI. I'm like, well, okay, we don't know how the brain works that well. We don't know how birds fly that well yet either. We have no idea how a hummingbird flies in terms of the aerodynamics of it. On the other hand, we know basic principles of, like, flapping and pushing the air down. And we know the basic principles of how the different parts of the brain work. So let's take those basic principles and engineer something that embodies those basic principles, but is well-designed for the hardware that we have on hand right now. So do you think we can create AGI before we understand how the brain works? I think that's probably what will happen. And maybe the AGI will help us do better brain imaging that will then let us build artificial humans, which is very, very interesting to us, because we are humans, right? I mean, building artificial humans is super worthwhile. </s><s>[INST] I just think it's probably not the shortest path to AGI. So it's a fascinating idea that we would build AGI to help us understand ourselves. A lot of people ask me if young people interested in doing artificial intelligence, they look at sort of doing graduate level, even undergrads, but graduate level research, and they see what the artificial intelligence community stands now. It's not really AGI type research for the most part. So the natural question they ask is, what advice would you give? I mean, maybe I could ask, if people were interested in working on OpenCog or in some [/INST] kind of direct or indirect connection to OpenCog or AGI research, what would you recommend? Yeah. OpenCog, first of all, is open source project. There's a Google group discussion list, there's a GitHub repository. So if anyone's interested in landing a hand with that aspect of AGI, introduce yourself on the OpenCog email list, and there's a Slack as well. I mean, we're certainly interested to have inputs into our redesign process for a new version of OpenCog, but also we're doing a lot of very interesting research. I mean, we're working on data analysis for COVID clinical trials, we're working with Hanson Robotics, we're doing a lot of cool things with the current version of OpenCog now. So there's certainly opportunity to jump into OpenCog or various other open source AGI-oriented </s><s>[INST] project. [/INST] So would you say there's like master's and PhD theses in there? Plenty, yeah, plenty, of course. I mean, the challenge is to find a supervisor who wants to foster that sort of research, but it's way easier than it was when I got my PhD. </s><s>[INST] Sure. Okay, great, we talked about OpenCog, which is kind of one, the software framework, but also the actual attempt to build an AGI system. And then there is this exciting idea of SingularityNet. [/INST] So maybe, can you say first, what is SingularityNet? Sure, sure. SingularityNet is a platform for realizing a decentralized network of artificial intelligences. So Marvin Minsky, the AI pioneer, who I knew a little bit, he had the idea of a society of minds. Like you should achieve an AI, not by writing one algorithm or one program, but you should put a bunch of different AIs out there. And the different AIs will interact with each other, each playing their own role. And then the totality of the society of AIs would be the thing that displayed the human-level intelligence. And when he was alive, I had many debates with Marvin about this idea. And I think he really thought the mind was more like a society than I do. I think you could have a mind that was as disorganized as a human society, but I think a human-like mind has a bit more central control than that, actually. I mean, we have this thalamus and the medulla and limbic system. We have a sort of top-down control system that guides much of what we do, more so than a society does. So I think he stretched that metaphor a little too far, but I also think there's something interesting there. And so in the 90s, when I started my first sort of non-academic AI project, WebMind, which was an AI startup in New York, in the Silicon Alley area in the late 90s, what I was aiming to do there was make a distributed society of AIs, the different parts of which would live on different computers all around the world, and each one would do its own thinking about the data local to it. But they would all share information with each other and outsource work with each other and cooperate, and the intelligence would be in the whole collective. And I organized a conference together with Francis Heiligen at Free University of Brussels in 2001, which was the Global Brain Zero conference, and we're planning the next version, the Global Brain One conference at the Free University of Brussels for next year, 2021, so 20 years after. And then maybe we can have the next one 10 years after that, like exponentially faster until the singularity comes, right? The timing is right, yeah. Yeah, yeah, exactly. So yeah, the idea with the Global Brain was maybe the AI won't just be in a program, one guy's computer, but the AI will be, you know, in the internet as a whole, with the cooperation of different AI modules living in different places. </s><s>[INST] So one of the issues you face when architecting a system like that exponentially faster until the singularity comes, right? The timing is right, yeah. [/INST] Yeah, yeah, exactly, is, you know, how is the whole thing controlled? Do you have like a centralized control unit that pulls the puppet strings of all the different modules there, or do you have a fundamentally decentralized network where the society of AIs is controlled in some democratic and self-organized web, all the AIs in that society, right? And Francis and I had different view of many things, but we both wanted to make like a global society of AI minds with a decentralized organizational mode. Now, the main difference was he wanted the individual AIs to be all incredibly simple and all the intelligence to be on the collective level, whereas I thought that was cool, but I thought a more practical way to do it might be if some of the agents in the society of minds were fairly generally intelligent on their own. So like, you could have a bunch of OpenCogs out there and a bunch of simpler learning systems and then these are all cooperating and coordinating together. So like, in the brain, okay, the brain as a whole is the general intelligence, but some parts of the cortex, you could say, have a fair bit of general intelligence on their own, whereas say parts of the cerebellum or limbic system have very little general intelligence on their own, and they're contributing to general intelligence, you know, by way of </s><s>[INST] their connectivity to other modules. So you see instantiations of the same kind of, you know, maybe different versions of OpenCog, but also just the same version of OpenCog and maybe many instantiations of it [/INST] as part of it as part of it. That's what David and Hans and I want to do with many Sophia and other robots. Each one has its own individual mind living on a server, but there's also a collective intelligence infusing them and a part of the mind living on the edge in each robot, right? Yeah. So the thing is, at that time, as well as WebMind being implemented in Java 1.1 as like a massive distributed system, the blockchain wasn't there yet. So had them do this decentralized control, you know, we sort of knew it, we knew about distributed systems, we knew about encryption, so I mean, we had the key principles of what underlies blockchain now, but I mean, we didn't put it together in the way that it's been done now. So when Vitalik Buter and then colleagues came out with Ethereum blockchain, you know, many, many years later, like 2013 or something, then I was like, well, this is interesting. Like this is Solidity scripting language. It's kind of dorky in a way, and I don't see why you need to turn complete language for this purpose. But on the other hand, this is like the first time I could sit down and start to like script infrastructure for decentralized control of the AIs in a society of minds in a tractable way. Like you could hack the Bitcoin code base, but it's really annoying, whereas Solidity is Ethereum scripting language is just nicer and easier to use. I'm very annoyed with it by this point, but like Java, I mean, these languages are amazing when they first come out. So then I came up with the idea that turned into SingularityNet, okay, let's make a decentralized agent system where a bunch of different AIs, you know, wrapped up in, say, different Docker containers or LXC containers, different AIs can each of them have their own identity on the blockchain. And the coordination of this community of AIs has no central controller, no dictator, right? And there's no central repository of information. The coordination of the society of minds is done entirely by the decentralized network in a decentralized way by the algorithms, right, because, you know, the model of Bitcoin is in math we trust, right? And so that's what you need. You need the society of minds to trust only in math, not trust only in one centralized server. </s><s>[INST] So the AI systems themselves are outside of the blockchain, but then the communication [/INST] between them. That's the moment. Yeah. Yeah. I would have loved to put the AIs operations on chain in some sense, but in Ethereum, it's just too slow. </s><s>[INST] You can't do it. Somehow, it's the basic communication between AI systems that's a... [/INST] Yeah, yeah. Yeah, yeah. Basically, an AI is just some software and singularity, and AI is just some software process living in a container. That's input and output. There's a proxy that lives in that container along with the AI that handles interaction with the rest of SingularityNet. And then when one AI wants to contribute with another one in the network, they set up a number of channels and the setup of those channels uses the Ethereum blockchain. But once the channels are set up then data flows along those channels without having to be on the blockchain. All that goes on the </s><s>[INST] blockchain is the fact that some data went along that channel. So there's [/INST] not a shared knowledge. Well the identity of each agent is on the blockchain. If one agent rates the reputation of another agent that goes on the blockchain an agent can publish what APIs they will fulfill on the blockchain but the actual data for AI is not on the blockchain. Do you think it could be? Do you think it should be? In some cases it should be. In some cases maybe it shouldn't be but I think I think that using Ethereum, you can't do it. Using now, there's more modern and faster blockchains where you could start to do that in some cases. Two years ago, that was less so. </s><s>[INST] It's a very rapidly evolving ecosystem. I'll give you an example. So like one example, maybe you can comment on something I worked on a lot on is autonomous vehicles. You can see each individual vehicle as a AI system. And you can see vehicles from Tesla, for example, and then Ford and GM and all these as also like larger. I mean, they all are running the same kind of system on each sets of vehicles. So it's individual AI systems and individual vehicles, but it's all different. Statiation is the same AI system within the same company. So, you know, you can envision a situation where all of those AI systems are put on singularity net. Right? Yeah. And how do you see that happening and what would be the benefit? And could they share data? I guess one of the biggest things is that the power there is in a decentralized control, but the benefit would have been, is really nice if they can somehow share the knowledge in an open way [/INST] if they choose to. Yeah. Yeah, yeah, yeah. Those are all quite good points. So I think the benefit from being on the decentralized network as we envision it is that we want the AIs and the network to be outsourcing work to each other and making API calls to each other frequently. I got you. So the real benefit would be if that AI wanted to outsource some cognitive processing or data processing or data pre-processing, whatever, to some other AIs in the network, which specialize in something different. And this really requires a different way of thinking about AI software development, right? So just like object-oriented programming was different than imperative programming. And now object-oriented programmers all use these frameworks to do things rather than just libraries even. You know, shifting to agent-based programming where your AI agent is asking other like live real-time evolving agents for feedback and what they're doing. That's a different way of thinking. I mean, it's not a new one. There was loads of papers on agent-based programming in the 80s and onward. But if you're willing to shift to an agent-based model of development, then you can put less and less in your AI and rely more and more on interactive calls to other AIs running in the network. And of course, that's not fully manifested yet because although we've rolled out a nice working version of singularity in that platform, there's only 50 to 100 AIs running in there now. There's not tens of thousands of AIs. So we don't have the critical mass for the whole society of mind to be doing what we want. </s><s>[INST] Yeah, the magic really happens when there's just a huge number of agents. [/INST] Yeah, yeah, exactly. In terms of data, we're partnering closely with another blockchain project called Ocean Protocol. And Ocean Protocol, that's the project of Trent McConaughey who developed Big Chain DB, which is a blockchain-based database. So Ocean Protocol is basically blockchain-based big data and aims at making it efficient for different AI processes or statistical processes or whatever to share large data sets. Or one process can send a clone of itself to work on the other guy's data set and send results back and so forth. So by getting ocean and you have data lakes, so this is the data ocean, right? So again, by getting ocean and singularity net to interoperate, we're aiming to take into account of the big data aspect also. But it's quite challenging because to build this whole decentralized blockchain-based infrastructure, I mean, your competitors are like Google, Microsoft, Alibaba, and Amazon, which have so much money to put behind their centralized infrastructures plus they're solving simpler algorithmic problems because making it centralized in some ways is easier, right? So they're very major computer science challenges. And I think what you saw with the whole ICO boom in the blockchain and cryptocurrency world is a lot of young hackers who were hacking Bitcoin or Ethereum and they see, well, why don't we make this decentralized on blockchain, then after they raised some money through an ICO, they realized how hard it is. And so like, actually we're wrestling with incredibly hard computer science and software engineering and distributed systems problems, which can be solved, but they're just very difficult to solve and in some cases, the individuals who started those projects were not well-equipped to actually solve the problems </s><s>[INST] that they want to gain back is easier, right? So you think, would you say that's the main bottleneck? If you look at the future of currency, [/INST] you know, the question is, well- Currency, the main bottleneck is politics. So it's governance and the fans of armed thugs that will shoot you if you bypass </s><s>[INST] their currency of government restriction. That's right. So like your sense is that versus the technical challenges because you kind of just suggested [/INST] the technical challenges are quite high as well. For making a distributed money, you could do that on Algorand right now. I mean, so that while Ethereum is too slow, there's Algorand and there's a few other more modern, more scalable blockchains that would work fine for a decentralized global currency. So I think there were technical bottlenecks to that two years ago. And maybe Ethereum 2.0 will be as fast as Algorand. I don't know, that's not fully written yet, right? So I think the obstacle to currency being put on the blockchain is the- Is the other stuff you mentioned. I mean, currency will be on the blockchain. It'll just be on the blockchain in a way that enforces centralized control and government hedge money rather than otherwise. Like the ERMP will probably be the first global, the first currency on the blockchain. The E-Ruble maybe next, there already were- E-Ruble. Yeah, yeah, yeah. I mean- Oh, that's hilarious. Digital currency, you know, makes total sense, but they would rather do it in the way that Putin and Xi Jinping have access </s><s>[INST] to the global keys for everything, right? Is the other stuff you mentioned. I mean, the current- Oh, that's hilarious. Digital- So, and then the analogy to that in terms of singularity net. I mean, there's echoes, I think you've mentioned before that Linux gives you hope. And- [/INST] AI is not as heavily regulated as money, right? </s><s>[INST] Not yet, right? Not yet, right? [/INST] Not yet. Oh, that's a lot slipperier than money too, right? I mean, money is easier to regulate because it's kind of easier to define. Whereas AI is, it's almost everywhere inside everything. Where's the boundary between AI and software, right? I mean, if you're gonna regulate AI, there's no IQ test for every hardware device that has a learning algorithm. You're gonna be putting like hegemonic regulation on all software. And I don't rule out that that- Any adaptive software. Yeah, but how do you tell if software is adaptive? </s><s>[INST] And what every software is gonna be adaptive, I mean. Or maybe we're living in the golden age of open source that will not always be open. Maybe it'll become centralized control of software [/INST] by governments. It is entirely possible. And part of what I think we're doing with things like singularity net protocol is creating a tool set that can be used to counteract that sort of thing. Say a similar thing about mesh networking, right? Plays a minor role now, the ability to access internet like directly phone to phone. On the other hand, if your government starts trying to control your use of the internet, suddenly having mesh networking there can be very convenient, right? And so right now, something like a decentralized blockchain based AGI framework or narrow AI framework, it's cool, it's nice to have. On the other hand, if governments start trying to tamp down on my AI, interoperating with someone's AI in Russia or somewhere, right? Then suddenly having a decentralized protocol that nobody owns or controls becomes an extremely valuable part of the tool set. And we've put that out there now. It's not perfect, but it operates. And it's pretty blockchain agnostic. So we're talking to Algorand about making part of Singularity Run on Algorand. My good friend Tufi Saliba has a cool blockchain project called Toda, which is a blockchain without a distributed ledger. It's like a whole other architecture. So there's a lot of more advanced things you can do in the blockchain world. Singularity Net could be ported to a whole bunch of, it could be made multi-chain, important to a whole bunch of different blockchains. And there's a lot of potential and a lot of importance to putting this kind of tool set out there. If you compare the OpenCog, what you could see is OpenCog allows tight integration of a few AI algorithms that share the same knowledge store in real time in RAM, right? Singularity Net allows loose integration of multiple different AIs. They can share knowledge, but they're mostly not gonna be sharing knowledge in RAM on the same machine. And I think what we're gonna have is a network of network of networks, right? Like, I mean, you have the knowledge graph inside the OpenCog system. And then you have a network of machines inside a distributed OpenCog mind. But then that OpenCog will interface with other AIs doing deep neural nets or custom biology data analysis or whatever they're doing in Singularity Net, which is a looser integration of different AIs, some of which may be their own networks, right? And I think at a very loose analogy, you could see that in the human body. Like the brain has regions like cortex or hippocampus, which tightly interconnects like cortical columns within the cortex, for example. Then there's looser connection within the different lobes of the brain. And then the brain interconnects with the endocrine system and different parts of the body even more loosely. Then your body interacts even more loosely with the other people that you talk to. So you often have networks within networks, within networks with progressively looser coupling as you get higher up in that hierarchy. I mean, you have that in biology. You have that in the internet as just networking medium. And I think that's what we're gonna have </s><s>[INST] in the network of software processes leading to people. That's a beautiful way to see the world. Again, the same similar question is with OpenCog. If somebody wanted to build an AI system and plug into the singularity net, what would you recommend? [/INST] Yeah, so that's much easier. I mean, OpenCog is still a research system, so it takes some expertise to, and sometimes, we have tutorials, but it's somewhat cognitively labor-intensive to get up to speed on OpenCog. And I mean, what's one of the things we hope to change with the true AGI OpenCog 2.0 version is just make the learning curve more similar to TensorFlow or Torch or something. Because right now, OpenCog is amazingly powerful, but not simple to do one. On the other hand, singularity net, you know, as an open platform was developed a little more with usability in mind, although the blockchain is still kind of a pain. So I mean, if you're a command line guy, there's a command line interface. It's quite easy to take any AI that has an API and lives in a Docker container and put it online anywhere, and then it joins the global singularity net. And anyone who puts a request for services out into the singularity net, the peer-to-peer discovery mechanism will find your AI. And if it does what was asked, it can then start a conversation with your AI about whether it wants to ask your AI to do something for it, how much it would cost and so on. So that's fairly simple. If you wrote an AI and want it listed on like official singularity net marketplace, which is on our website, then we have a publisher portal and then there's a KYC process to go through because then we have some legal liability for what goes on on that website. So in a way that's been an education too. There's sort of two layers. Like there's the open decentralized protocol. And there's the market. Yeah, anyone can use the open decentralized protocol. So say some developers from Iran and there's brilliant AI guys in University of Isfahan in Tehran, they can put their stuff on singularity net protocol and just like they can put something on the internet, right? I don't control it. But if we're gonna list something on the singularity net marketplace and put a little picture and a link to it, then if I put some Iranian AI geniuses code on there, then Donald Trump can send a bunch of jackbooted thugs to my house to arrest me for doing business with Iran, right? So, I mean, we already see in some ways the value of having a decentralized protocol because what I hope is that someone in Iran will put online an Iranian singularity net marketplace, right? Which you can pay in the cryptographic token, which is not owned by any country. And then if you're in like Congo or somewhere that doesn't have any problem with Iran, you can subcontract AI services that you find on that marketplace, right? Even though US citizens can't buy by US law. So right now, that's kind of a minor point. You know, as you alluded, if regulations go in the wrong direction, it could become more of a major point. But I think it also is the case that having these workarounds to regulations in place is a defense mechanism against those regulations being put into place. And you can see that in the music industry, right? I mean, Napster just happened and BitTorrent just happened. And now most people in my kids' generation, they're baffled by the idea of paying for music, right? I mean, my dad pays for music. I mean, because these decentralized mechanisms happened and then the regulations followed, right? And the regulations would be very different if they'd been put into place before there was Napster and BitTorrent and so forth. So in the same way, we got to put AI out there in a decentralized vein and big data out there in a decentralized vein now so that the most advanced AI in the world is fundamentally decentralized. And if that's the case, that's just the reality the regulators have to deal with. And then as in the music case, they're going to come up with regulations </s><s>[INST] that sort of work with the decentralized reality, which is- And there's the market. Yeah. My dad pays for music, but I mean- Yeah, that's true. Beautiful. You were the chief scientist of Handsome Robotics. You're still involved with Handsome Robotics, doing a lot of really interesting stuff there. This is for people who don't know the company that created Sofia, the robot. [/INST] Can you tell me who Sofia is? I'd rather start by telling you who David Handsome is because David is the brilliant mind behind the Sofia robot. And so far he remains more interesting than his creation, although she may be improving faster than he is, actually, I mean, he's- Yeah. So, yeah, I- That's a good point. I met David maybe 2007 or something at some futurist conference we were both speaking at. And I could see we had a great deal in common. I mean, we were both kind of crazy, but we also, we both had a passion for AGI and the singularity. And we were both huge fans of the work of Philip K. Dick, the science fiction writer. And I wanted to create benevolent AGI that would create massively better life for all humans and all sentient beings, including animals, plants, and superhuman beings. And David, he wanted exactly the same thing, but he had a different idea of how to do it. He wanted to get computational compassion. Like he wanted to get machines that would love people and empathize with people. And he thought the way to do that was to make a machine that could look people eye to eye, face to face, look at people and make people love the machine and the machine loves the people back. So I thought that was very different way of looking at it because I'm very math oriented. And I'm just thinking like, what is the abstract cognitive algorithm that will let the system internalize the complex patterns of human values, blah, blah, blah, whereas he's like, look you in the face in the eye and love you, right? So we hit it off quite well and we talked to each other off and on. Then I moved to Hong Kong in 2011. So I've been living all over the place. I've been in Australia and New Zealand in my academic career. Then in Las Vegas for a while. Was in New York in the late 90s starting my entrepreneurial career. Was in DC for nine years doing a bunch of U.S. government consulting stuff. </s><s>[INST] Then moved to Hong Kong in 2011. Because he's a date neater, so actually, I mean, he's a- [/INST] Yeah, that's a good point. Mostly because I met a Chinese girl who I fell in love with and we got married. She's actually not from Hong Kong. She's from mainland China, but we converged together in Hong Kong. </s><s>[INST] Still married now, have a two-year-old baby, huh? So I went to Hong Kong to see about a girl, I guess. [/INST] Yeah, pretty much, yeah. And on the other hand, I started doing some cool research there with the Geno U and Hong Kong Polytechnic University. I got involved with a project called IDEA using machine learning for stock and futures prediction, which was quite interesting. And I also got to know something about the consumer electronics and hardware manufacturer ecosystem in Shenzhen across the border, which is like the only place in the world that makes sense to make complex consumer electronics at large scale and low cost. It's just, it's astounding the hardware ecosystem that you have in South China, like US people here cannot imagine what it's like. So yeah, David was starting to explore that also. I invited him to Hong Kong to give a talk at Hong Kong PolyU. And I introduced him in Hong Kong to some investors who were interested in his robots. And he didn't have Sophia then. He had a robot of Philip K. Dick, our favorite science fiction writer. He had a robot Einstein. He had some little toy robots that looked like his son, Zeno. So through the investors I connected him to, he managed to get some funding to basically port Hanson Robotics to Hong Kong. And when he first moved to Hong Kong, I was working on AGI research and also on this machine learning trading project. So I didn't get that tightly involved with Hanson Robotics. But as I hung out with David more and more, as we were both there in the same place, I started to think about what you could do to make his robots smarter than they were. And so we started working together. And for a few years, I was chief scientist and head of software at Hanson Robotics. Then when I got deeply into the blockchain side of things, I stepped back from that and co-founded SingularityNet. David Hanson was also one of the co-founders of SingularityNet. So part of our goal there had been to make the blockchain-based CloudMind platform for Sofia and the other... </s><s>[INST] Sofia will be just one of the robots in SingularityNet. [/INST] Yeah, yeah, exactly. Sofia, many copies of the Sofia robot would be among the user interfaces to the globally distributed SingularityNet CloudMind. And I mean, David and I talked about that </s><s>[INST] for quite a while before co-founding SingularityNet. By the way, in his vision, in your vision, was Sofia tightly coupled to a particular AI system? Or was the idea that you can plug... You could just keep plugging in different AI systems [/INST] within the head of it. You could just then head of it. David's view was always that Sofia would be a platform, much like, say, the Pepper robot is a platform from SoftBank. It should be a platform with a set of nicely designed APIs that anyone can use to experiment with their different AI algorithms on that platform. And SingularityNet, of course, fits right into that, right? Because SingularityNet, it's an API marketplace. So anyone can put their AI on there. OpenCog is a little bit different. I mean, David likes it. But I'd say it's my thing. It's not his. David has a little more passion for biologically-based approaches to AI than I do, which makes sense. I mean, he's really into human physiology and biology. He's a character sculptor, right? Yeah. So yeah, he's interested in... But he also works a lot with rule-based and logic-based AI systems, too. So yeah, he's interested in not just Sofia, but all the Henson robots as a powerful software and Henson robots as a powerful social and emotional robotics platform. And what I saw in Sofia was a way to get AI algorithms out there in front of a whole lot of different people in an emotionally compelling way. And part of my thought was really kind of abstract, connected to AGI ethics. And many people are concerned AGI is going to enslave everybody or turn everybody into computronium to make extra hard drives for their cognitive engine or whatever. And emotionally, I'm not driven to that sort of paranoia. I'm really just an optimist by nature. But intellectually, I have to assign a non-zero probability to those sorts of nasty outcomes. Because if you're making something 10 times as smart as you, how can you know what it's going to do? There's an irreducible uncertainty there, just as my dog can't predict what I'm going to do tomorrow. So it seemed to me that based on our current state of knowledge and the best way to bias the AGIs we create toward benevolence would be to infuse them with love and compassion the way that we do our own children. And if you want to interact with AIs in the context of doing compassionate, loving and beneficial things, and in that way as your children will learn by doing compassionate, beneficial, loving things alongside you in that way the AI will learn in practice what it means to be compassionate, beneficial and loving, it will get a sort of ingrained, intuitive sense of this which it can then abstract in its own way as it gets more and more intelligent. Now David saw this the same way. That's why he came up with the name Sophia, which means, which means wisdom. So it seemed to me making these like beautiful, loving robots to be rolled out for beneficial applications would be the perfect way to roll out early stage AGI systems so they can learn from people and not just learn factual knowledge, but learn human values and ethics from people while being their, you know, their home service robots, their education assistants, their nursing robots. So that was the grand vision though. If you've ever worked with robots, the reality is quite different, right? Like the first principle is the robot is always broken. I mean, I've worked with robots in the 90s a bunch when you had to solder them together yourself. And I'd put neural nets during reinforcement learning on like overturned solid bowl type robots. And in the 90s when I was a professor, things of course advanced a lot, but, but the principal stuff, the robot's always broken. So yeah. Yeah. So faced with the reality of making Sophia do stuff, many of my Robo AGI aspirations were temporarily cast aside. And I mean, there's just a practical problem of making this robot interact in a meaningful way. Because like, you know, you put nice computer vision on there, but there's always glare. And then, or you have a dialogue system, but at the time I was there, like no speech to text algorithm could deal with Hong Kong, Hong Kong, East people's English accents. So the, the speech to text was always bad. So the robot always sounded stupid because it wasn't getting the right text. </s><s>[INST] Right. So I started to view that really as what I mean, I've worked with the robots always broken [/INST] stoles. What in software engineering you call a walking skeleton, which is maybe the wrong metaphor to use for Sophia or maybe the right one. I mean, what a walking skeleton is in software development is if you're building a complex system, how do you get started? Well, one way is to first build part one well, then build part two well, then build part three well, and so on. Another way is you make like a simple version of the whole system and put something in the place of every part the whole system will need so that you have a whole system that does something. And then you work on improving each part in the context of that whole integrated system. So that's what we did on a software level in Sophia. We made like a walking skeleton software system where, so there's something that sees, there's something that hears, there's something that moves, there's something that there's something that remembers, there's something that learns you put a simple version of each thing in there and you connect them all together so that the system will do its thing. So there's a lot of AI in there. There's not any AGI in there. I mean, there's computer vision to recognize people's faces, recognize when someone comes in the room and leaves, try to recognize whether two people are together or not. I mean, the dialogue system, it's a mix of like hand-coded rules with deep neural nets that come up with their own responses. And there's some attempt to have a narrative structure and sort of try to pull the conversation into something with the beginning, middle, and end, and this </s><s>[INST] sort of story arc. I mean, if you look at the Lobner Prize and the systems that beat the towing test currently, they're heavily rule-based because, like you had said, narrative structure, to create compelling conversations, neural networks cannot do that well, even with Google Mina. When you actually look at full-scale conversations, it's just not... [/INST] Yeah, this is a thing. So I've actually been running an experiment the last couple of weeks taking Sophia's chatbot and then Facebook's Transformer chatbot, which they opened the model. We've had them chatting to each other for a number of weeks on the server just... That's funny. We're generating training data of what Sophia says in a wide variety of conversations, but we can see, compared to Sophia's current chatbot, the Facebook deep neural chatbot comes up with a wider variety of fluent sounding sentences. On the other hand, it rambles like mad. The Sophia chatbot, it's a little more repetitive in the sentence structures it uses. On the other hand, it's able to keep like a conversation arc over a much longer period, right? So there... Now, you can probably surmount that using Reformer and using various other deep neural architectures to improve the way these Transformer models are trained, but in the end, neither one of them really understands what's going on. And I mean, that's the challenge I had with Sophia, is if I were doing a robotics project aimed at AGI, I would want to make like a robo-toddler that was just learning about what it was seeing, because then the language is grounded in the experience of the robot. But what Sophia needs to do to be Sophia is talk about sports or the weather or robotics or the conference she's talking at. She needs to be fluent talking about any damn thing in the world, and she doesn't have grounding for all those things. So there's this just like, I mean, Google Mina and Facebook's chatbot don't have grounding for what they're talking about either. So in a way, the need to speak fluently about things where there's no non-linguistic grounding pushes what you can do for Sophia in the short </s><s>[INST] term a bit away from... It pushes you towards IBM Watson situation where you basically have to heuristic and hardcore stuff and rule-based stuff. I have to ask you about this. Okay, so because in part, Sophia is an art creation, because it's beautiful. She's beautiful because she inspires through our human nature of anthropomorphized things. We immediately see an intelligent being there. Because David is a great sculptor. He is a scary sculptor. That's right. So in fact, if Sophia just had nothing inside her head, said nothing, [/INST] if she just sat there, we already prescribed some intelligence to us. There's a long selfie </s><s>[INST] line in front of her after every talk. That's right. So it captivated the imagination of many [/INST] people. I was going to say the world, but yeah, I mean, a lot of people. Billions of people, </s><s>[INST] which is amazing. It's amazing, right. Now, of course, many people have prescribed much greater prescribed, essentially AGI type of capabilities to Sophia when they see her. And of course, friendly French folk like Jan LeCun immediately see that of the people from the AI community and get really frustrated because... It's understandable. And then they criticize people like you who sit back and don't say anything about, like basically allow the imagination of the world, allow the world to continue being captivated. So what's your sense of that kind of annoyance [/INST] that the AI community has? I think there's several parts to my reaction there. First of all, if I weren't involved with Hanson-Rabax and didn't know David Hanson personally, I probably would have been very annoyed initially at Sophia as well. I mean, I can understand the reaction. I would have been like, wait, all these stupid people out there think this is an AGI, but it's not an AGI, but they're tricking people that this very cool robot is an AGI. And now those of us trying to raise funding to build AGI, people will think it's already there and already works. On the other hand, I think even if I weren't directly involved with it, once I dug a little deeper into David and the robot and the intentions behind it, I think I would have stopped being pissed off, whereas folks like Yann LeCun have remained pissed off after their initial </s><s>[INST] reaction. That's his thing. That's his thing, yeah. That's his thing, yeah. [/INST] That in particular struck me as somewhat ironic because Yann LeCun is working for Facebook, which is using machine learning to program the brains of the people in the world toward vapid consumerism and political extremism. So if your ethics allows you to use machine learning in such a blatantly destructive way, why would your ethics not allow you to use machine learning to make a lovable theatrical robot that draws some foolish people into its theatrical illusion? If the pushback had come from Yoshua Bengio, I would have felt much more humbled by it because he's not using AI for blatant evil. On the other hand, he also is a super nice guy and doesn't bother to go out there trashing other people's work for no good reason. Shots fired, but I get you. If you're going to ask, I'm going to answer. </s><s>[INST] No, for sure. I think we'll go back and forth. I'll talk to Yann again. [/INST] I would add on this, though. I mean David Hansen is an artist and he often speaks off the cuff. I have not agreed with everything that David has said or done regarding Sophia, and David also has not agreed with everything that David has said or done about Sophia. David is an artistic Wild Man and that's part of his charm. that's part of his genius. So certainly there have been conversations within Hanson Robotics and between me and David where I was like, let's be more open about how this thing is working. And I did have some influence in nudging Hanson Robotics to be more open about how Sophia was working. And David wasn't especially opposed to this. And, you know, he was actually quite right about it. What he said was, you can tell people exactly how it's working and they won't care. They want to be drawn into the illusion. And he was a hundred percent correct. I'll tell you, this wasn't Sophia, this was Philip K. Dick, but we did some interactions between humans and Philip K. Dick robot in Austin, Texas a few years back. And in this case, the Philip K. Dick was just teleoperated by another human in the other room. So during the conversations, we didn't tell people the robot was teleoperated. We just said, here, have a conversation with Phil Dick, we're going to film you. And they had a great conversation with Philip K. Dick, teleoperated by my friend, Stefan Bugay. After the conversation, we brought the people in the back room to see Stefan, who was controlling the Philip K. Dick robot, but they didn't believe it. These people were like, well, yeah, but I know I was talking to Phil, like maybe Stefan was typing, but the spirit of Phil was animating his mind while he was typing. So even though they knew it was a human in the loop, even seeing the guy there, </s><s>[INST] they still believed that was Phil they were talking to, her illusion. And he was talking to her. A small part of me believes that they were right, actually, [/INST] because our understanding... Well, we don't understand the universe. That's the thing. I mean, there is a cosmic mind field that we're all embedded in that yields many strange synchronicities in the world, which is a topic we don't have time to go into </s><s>[INST] too much here. There is a cosmic universe. There's something to this where our imagination about Sophia and people like Yann LeCun being frustrated about it is all part of this beautiful dance of creating artificial intelligence. That's almost essential. You see with Boston Dynamics, I'm a huge fan of as well, the kind of... I mean, these robots are very far from intelligent. [/INST] I played with the last one, actually. </s><s>[INST] With the Spot Mini. [/INST] Yeah, very cool. I mean, it reacts quite in a fluid and flexible way. </s><s>[INST] Yeah, but we immediately ascribed the kind of intelligence. We immediately ascribed AGI to them. [/INST] Yeah, if you kick it and it falls down and goes out, you feel bad. You can't help it. </s><s>[INST] And I mean, that's going to be part of our journey in creating intelligence systems, more and more and more and more. As Sophia starts out with the walking skeleton, as you add more and more intelligence, I mean, we're going to have to deal with this kind of idea. Absolutely. [/INST] More and more ideas. Absolutely. And about Sophia, I would say, I mean, first of all, I have nothing against Jan LeCun. No, no, this is fine. He's a nice guy. If he wants to play the media banter game, I'm happy to play him. He's a good researcher and a good human being, and I'd happily work with the guy. The other thing I was going to say is, I have been explicit about how Sophia works, and I've posted online, and what, H Plus magazine, an online webzine. I mean, I posted a moderately detailed article explaining, like, there are three software systems we've used inside Sophia. There's a timeline editor, which is like a rule-based authoring system, where she's really just being an outlet for what a human scripted. There's a chat bot, which has some rule-based and some neural aspects. And then sometimes we've used OpenCog behind Sophia, where there's more learning and reasoning. And, you know, the funny thing is, I can't always tell which system is operating here, right? I mean, so whether she's really learning, or thinking, or just appears to be over half hour, I could tell, but over, like, three or four minutes of interaction. </s><s>[INST] So even having three... No, no, no, this is fine. This is all for fun. He's a nice guy. So even having three systems that's already sufficiently complex, [/INST] where you can't really tell right away. Yeah, the thing is, even if you get up on stage and tell people how Sophia is working, and then they talk to her, they still attribute more agency and consciousness to her than is really there. So I think there's a couple levels of ethical issue there. One issue is, is should you be transparent about how Sophia is working? And I think you should. And I think we have been. I mean, there's articles online, there's some TV special that goes through me explaining the three subsystems behind Sophia. So the way Sophia works is out there much more clearly than how Facebook's AI works or something, right? I mean, we've been fairly explicit about it. The other is, given that telling people how it works doesn't cause them to not attribute too much intelligence agency to it anyway, then should you keep fooling them when they want to be fooled? And I mean, the whole media industry is based on fooling people the way they want to be fooled. And we are fooling people 100% toward a good end. I mean, we are playing on people's sense of empathy and compassion so that we can give them a good user experience with helpful robots and so that we can fill the AI's mind with love and compassion. So I've been I've been talking a lot with Hanson Robotics lately about collaborations in the area of medical robotics. And we haven't quite pulled the trigger on a project in that domain yet, but we may well do so quite soon. So we've been talking a lot about robots can help with elder care, robots can help with kids. David's done a lot of things with autism therapy and robots before. In the COVID era, having a robot that can be a nursing assistant in various senses can be quite valuable. The robots don't spread infection. And they can also deliver more attention than human nurses can give, right? So if you have a robot that's helping a patient with COVID, if that patient attributes more understanding and compassion and agency to that robot, then it really has because it looks like a human. I mean, is that really bad? I mean, we can tell them it doesn't fully understand you. And they don't care because they're lying there with a fever and they're sick, but they'll react better to that robot with its loving, warm facial expression that then they would to a pepper robot or a metallic looking, looking robot. So it's, it's really, it's about how you use it, right? If you made a human looking like door to door sales robot that used its human looking appearance to, to scam people out of their money, then you're using that, that connection in a bad way, but you could also use it in a good way. But then that's the same, the same problem with every technology, right? </s><s>[INST] Yeah, beautifully put. So like you said, we're living in the era of the COVID. This is 2020, one of the craziest years in recent history. So if we zoom out and look at this pandemic, the coronavirus pandemic, maybe let me ask you this kind of thing. In viruses in general, when you look at viruses, do you see them as, as a kind of intelligence system? [/INST] I think the concept of intelligence is not that natural of a concept in the end. I mean that, I think human minds and bodies are a kind of complex self-organizing adaptive system. And viruses certainly are that, right? They're a very complex self-organizing adaptive system. If you want to look at intelligence, as Marcus Hutter defines it, as sort of optimizing computable reward functions over computable environments, for sure viruses are doing that, right? And I mean, in doing so, they're causing some harm to us. And so they're, you know, they're, you know, the human immune system is a very complex self-organizing adaptive system, which has a lot of intelligence to it. And viruses are also adapting and dividing into new mutant strains and so forth. And ultimately the solution is going to be nanotechnology, right? I mean, I mean, the solution is going to be making little nanobots that fight the viruses or, well, people will use them to make nasty viruses, but hopefully we can also use them to just detect, combat and kill the viruses. But I think now we're stuck with the biological mechanisms to combat these viruses. And yeah, we've been AGI is not yet mature enough to use against COVID, but we've been using machine learning and also some machine reasoning in OpenCog to help some doctors to do personalized medicine against COVID. So the problem there is given the person's genomics and given their clinical medical indicators, how do you figure out which combination of antivirals is going to be most effective against COVID for that person? And so that's something where machine learning is interesting, but also we're finding the abstraction we get in OpenCog with machine reasoning is interesting because it can help with transfer learning when you have not that many different cases to study and qualitative differences between different strains of a virus or people of different ages who may have COVID. </s><s>[INST] So there's a lot of different disparate data to work with in a small data sets and somehow integrating them. [/INST] You know, this is one of the shameful things that's very hard to get that data. So, I mean, we're working with a couple of groups doing clinical trials and they're sharing data with us like under non-disclosure, but what should be the case is like every COVID clinical trial should be putting data online somewhere like suitably encrypted to protect patient privacy so that anyone with the right AI algorithms should be able to help analyze it and any biologist should be able to analyze it by hand to understand what they can, right? Instead, instead that data is like siloed inside whatever hospital is running the clinical trial, which is completely asinine and ridiculous. So why the world works that way? I mean, we could all analyze why, but it's insane that it does. Look at this hydrochloroquine, right? All these clinical trials were done or reported by Sergesphere, some little company no one ever heard of, and everyone paid attention to this. So they were doing more clinical trials based on that. Then they stopped doing clinical trials based on that. Then they started again. And why isn't that data just out there so everyone can analyze it and see what's going on, right? </s><s>[INST] Do you have hope that we'll move, that data will be out there eventually for future pandemics? I mean, do you have hope that our society will move in the direction of? [/INST] It's not in the immediate future because the US and China frictions are getting very high. So it's hard to see US and China as moving in the direction of openly sharing data with each other, right? There's some sharing of data, but different groups are keeping their data private until they've milked the best results from it, and then they share it, right? So yeah, we're working with some data that we've managed to get our hands on, something we're doing to do good for the world, and it's a very cool playground for putting deep neural nets and OpenCog together. So we have a bio-adm space full of all sorts of knowledge from many different biology experiments about human longevity and from biology knowledge bases online. And we can do like graph to vector type embeddings where we take nodes from the hypergraph, embed them into vectors, which can then feed into neural nets for different types of analysis. And we were doing this in the context of a project called the REJUVE that we spun off from SingularityNet to do longevity analytics. I can understand why people live to 105 years over and other people don't. And then we had this spin-off Singularity Studio where we're working with some healthcare companies on data analytics. So there's bio-adm space that we built for these more commercial and longevity data analysis purposes. We're repurposing and feeding COVID data into the same bio-adm space and playing around with like graph embeddings from that graph into neural nets for bioinformatics. So it's both being a cool testing ground for some of our bio AI learning and reasoning. And it seems we're able to discover things that people weren't seeing otherwise. Because the thing in this case is for each combination of antivirals, you may have only a few patients who've tried that combination. And those few patients may have their particular characteristics. Like this combination of three was tried only on people age 80 or over. This other combination of three, which has an overlap with the first combination, was tried more on young people. So how do you combine those different pieces of data? It's a very dodgy transfer learning problem, which is the kind of thing that the probabilistic reasoning algorithms we have inside OpenCog are better at than deep neural networks. On the other hand, you have gene expression data where you have 25,000 genes and the expression level of each gene and the peripheral blood of each person. So that sort of data, either deep neural nets or tools like XGBoost or CatBoost, these decision forest trees are better at dealing with it than OpenCog because it's just these huge, huge, messy floating point vectors that are annoying for a logic engine to deal with, but are perfect for a decision forest or neural net. So it's a great playground for like hybrid AI methodology. And we can have singularity net have OpenCog in one agent and XGBoost in a different agent and they talk to each other and feeding. But at the same time, it's highly practical, right? Because we're working with, for example, some physicians on this project and the physicians in the group called Nth Opinion based out of Vancouver in Seattle, who are these guys are working every day like in the hospital with patients dying of COVID. So it's quite cool to see like neural symbolic AI, like where the rubber hits the road, trying to save people's lives. I've been doing bio AI since 2001, but mostly human longevity research and fly longevity research, trying to understand why some organisms really live a long time. This is the first time like race against the clock and try to use the AI to figure out stuff that, like if we take two months longer to solve the AI problem, some more people will die because we don't know what combination of antivirals to give them. </s><s>[INST] Yeah. At the societal level, at the biological level, at any level, are you hopeful about us as a human species getting out of this pandemic? What are your thoughts in general? [/INST] Well, the pandemic will be gone in a year or two once there's a vaccine for it. So I mean, </s><s>[INST] a lot of pain and suffering can happen in that time. So that could be reversible on me. [/INST] I mean, I think if you spend much time in sub-Saharan Africa, you can see there's a lot of pain and suffering happening all the time. Like you walk through the streets of any large city in sub-Saharan Africa and there are loads, I mean, tens of thousands, probably hundreds of thousands of people lying by the side of the road, dying mainly of curable diseases without food or water and either ostracized by their families or they left their family house because they didn't want to infect their family. I mean, there's tremendous human suffering on the planet all the time, which most folks in the developed world pay no attention to. And COVID is not remotely the worst. How many people are dying of malaria all the time? I mean, so COVID is bad. It is by no mean the worst thing happening. And setting aside diseases, I mean, there are many places in the world where you're at risk of having like your teenage son kidnapped by armed militias and forced to get killed in someone else's war fighting tribe against tribe. I mean, some humanity has a lot of problems, which we don't need to have given the state of advancement of our technology right now. And I think COVID is one of the easier problems to solve in the sense that there are many brilliant people working on vaccines. We have the technology to create vaccines and we're going to create new vaccines. We should be more worried that we haven't managed to defeat malaria after so long and after the Gates Foundation and others putting so much so much money into it. I mean, I think clearly the whole global medical system, global health system and the global political and socio-economic system are incredibly unethical and unequal and badly designed. And I mean, I don't know how to solve that directly. I think what we can do indirectly to solve it is to make systems that operate in parallel and off to the side of the governments that are nominally controlling the world with their armies and militias. And to the extent that you can make compassionate peer-to-peer decentralized frameworks for doing things, these are things that can start out unregulated. And then if they get traction before the regulators come in, then they've influenced the way the world works, right? SingularityNet aims to do this with AI. Rejuve, which is a spinoff from SingularityNet, you can see it, Rejuve.io, Rejuve.io, that aims to do the same thing for medicine. So it's like peer-to-peer sharing of medical data. So you can share medical data into a secure data wallet. You can get advice about your health and longevity through apps that Rejuve will launch within the next couple of months. And then SingularityNet AI can analyze all this data, but then the benefits from that analysis are spread among all the members of the network. But I mean, of course, I'm going to halt my particular projects, but I mean, whether or not SingularityNet and Rejuve are the answer, I think it's key to create decentralized mechanisms for everything. I mean, for AI, for human health, for politics, for jobs and employment, for sharing social information, and to the extent decentralized peer-to-peer methods designed with universal compassion at the core can gain traction, then these will just decrease the role that government has. And I think that's much more likely to do good than trying to like explicitly reform the global government system. I mean, I'm happy other people are trying to explicitly reform the global government system. On the other hand, you look at how much good the internet, or Google did, or mobile phones did, even you're making something that's decentralized and throwing it out everywhere, and it takes hold, then government has to adapt. And I mean, that's what we need to do with AI and with health. And in that light, I mean, the centralization of healthcare and of AI is certainly not ideal, right? Like most AI PHGs are being sucked in by, you know, a half dozen to a dozen big companies. Most AI processing power is being bought by a few big companies for their own proprietary good. And most medical research is within a few pharmaceutical companies. And clinical trials run by pharmaceutical companies will stay solid within those pharmaceutical companies. You know, these large centralized entities, which are intelligences in themselves, these corporations, but they're mostly malevolent, psychopathic, and sociopathic intelligences. Not saying the people are involved, but the corporations as self-organizing entities on their own, which are concerned with maximizing shareholder value as a sole objective function. I mean, AI and medicine are being sucked into these pathological corporate organizations with government cooperation and Google cooperating with British and US government on this as one among many, many different examples. 23andMe, providing you the nice service of sequencing your genome and then licensing the genome to GlaxoSmithKline on an exclusive basis, right? Now, you can take your own DNA and do whatever you want with it, but the pooled collection of 23andMe sequence DNA is just to GlaxoSmithKline. Someone else could reach out to everyone who had worked with 23andMe to sequence their DNA and say, give us your DNA for our open and decentralized repository that we'll make available to everyone, but nobody's doing that because it's a pain to get organized and the customer list is proprietary to 23andMe. So yeah, I mean, this I think is a greater risk to humanity from AI than rogue AGIs turning the universe into paperclips or computronium because what you have here is mostly good hearted and nice people who are sucked into a mode of organization of large corporations, which has evolved just for no individual's fault, just because that's the </s><s>[INST] way society has evolved. It's not altruistic, it's self-interested and becomes psychopathic, [/INST] like you said. The human operation is psychopathic, even if the people are not. And that's really the disturbing thing about it because the corporations can do things that are quite bad for society, even if nobody has a bad intention. </s><s>[INST] Right, exactly. No individual member of that corporation has a bad intention. [/INST] No, some probably do, but it's not necessary that they do for the corporation. I know a lot of people in Google and with very few exceptions, they're all very nice people who genuinely want what's good for the world. And Facebook, I know fewer people, but it's probably mostly true. It's </s><s>[INST] probably fun young geeks who want to build cool technology. I actually tend to believe that even the leaders, even Mark Zuckerberg, one of the most disliked people in tech, also wants to do [/INST] good for the world. What do you think about what you think about Jamie Dimon? Who's Jamie Dimon? </s><s>[INST] The heads of the great banks may have a different psychology. Oh boy, yeah. Well, I tend to be naive about these things and see the best. I tend to agree with you that I think the individuals want to do good by the world, but the mechanism of the company can sometimes be its own intelligence [/INST] system. I mean, my cousin Mario Goetzel has worked for Microsoft since 1985 or something. And I can see for him, I mean, as well as just working on cool projects, you're coding stuff that gets used by like billions and billions of people. And you think if I improve this feature, that's making billions of people's lives easier, right? So of course, of course that's cool. And you know, the engineers are not in charge of running the company anyway. And of course, even if you're Mark Zuckerberg or Larry Page, I mean, you still have a fiduciary responsibility. And I mean, you're responsible to the shareholders, your employees who you want to keep paying them and so forth. So yeah, you're enmeshed in this system. And you know, when I worked in DC, I worked a bunch with INSCOM, U.S. Army Intelligence, and I was heavily politically opposed to what the U.S. Army was doing in Iraq at that time, like torturing people in Abu Graha. But everyone I knew in U.S. Army and INSCOM, when I hung out with them, was a very nice person. They were friendly to me. They were nice to my kids and my dogs, right? And they really believed that the U.S. was fighting the forces of evil. And if you ask them about Abu Graha, they're like, well, but these Arabs will chop us into pieces. So how can you say we're wrong to waterboard them a bit, right? Like that's much less than what they would do to us. It's just in their worldview, what they were doing was really genuinely for the good of humanity. Like none of them woke up in the morning and said like, I want to do harm to good people because I'm just a nasty guy, right? So yeah, most people on the planet setting aside a few genuine psychopaths and sociopaths. I mean, most people on the planet have a heavy dose of benevolence and wanting to do good, and also a heavy capability to convince themselves whatever they feel like doing or whatever is best for them </s><s>[INST] is for the good of humankind. So the more we can decentralize control of decentralization, [/INST] decentralization, you know, the democracy is horrible, but this is like Winston Churchill said, you know, it's the worst possible system of government except for all the others, right? I mean, I think the whole mess of humanity has many, many very bad aspects to it. But so far, the track record of elite groups who know what's better for all of humanity is much worse than the track record of the whole teeming democratic participatory mess of humanity, right? I mean, none of them is perfect by any means. The issue with a small elite group that knows what's best is even if it starts out as truly benevolent and doing good things in accordance with its initial good intentions, you find out you need more resources, you need a bigger organization, you pull in more people, internal politics arises, difference of opinions arise, and bribery happens like some, some opponent organization takes a second in command now to make some the first in command of some other organization. And I mean, that's, there's a lot of history of what happens with elite groups thinking they know what's best for the human race. So yeah, if I have to choose, I'm going to reluctantly put my faith in the vast democratic decentralized mass. And I think corporations have a track record of being ethically worse than their constituent human parts. And, you know, democratic governments have a more mixed track record, but there are at least that's the best we got. Yeah, I mean, you can, you can, there's Iceland, very nice country, right? I've been very democratic for 800 plus years. Very, very benevolent, beneficial government. And I think, yeah, there are track records of democratic modes of organization, Linux, for example, some of the people in charge of Linux are overtly complete assholes, right? And trying to reform themselves in many cases, in other cases, not, but the organization as a whole, I think it's done a good job. Overall, it's been very welcoming in the third world, for example, and it's allowed advanced technology to roll out on all sorts of different embedded devices and platforms in places where people couldn't afford to pay for proprietary software. So I'd say the internet, Linux, and many democratic nations are examples of how sort of open decentralized democratic methodology can be ethically better than the sum of the parts rather than worse. And corporations, that has happened only for a brief period, and then it goes sour, right? I mean, I'd say a similar thing about universities, like university is a horrible way to organize research and get things done, yet it's better than anything else we've come up with, right? A company can be much better, but for a brief period of time, and that stops being so good, right? So then I think if you believe that AGI is going to emerge sort of incrementally out of AIs doing practical stuff in the world, like controlling humanoid robots or driving cars or diagnosing diseases or operating killer drones or spying on people and reporting under the government, then what kind of organization creates more and more advanced, narrow AI verging toward AGI may be quite important because it will guide like what's in the mind of the early stage AGI as it first gains the ability to rewrite its own code base and project itself toward super intelligence. And if you believe that AI may move toward AGI out of this sort of synergetic activity of many agents cooperating together, rather than just to have one person's project, then who owns and controls that platform for AI cooperation becomes also very, very important, right? And is that platform AWS? Is it Google Cloud? Is it Alibaba? Or is it something more like the internet or Singularity Net, which is open and decentralized? So if all of my weird machinations come to pass, right? I mean, we have the Hanson robots being a beautiful user interface, gathering information on human values and being loving and compassionate to people in medical home service robot office applications. You have Singularity Net in the backend, networking together many different AIs toward cooperative intelligence, fueling the robots, among many other things. You have OpenCog 2.0 and true AGI as one of the sources of AI inside this decentralized network, powering the robot and medical AIs helping us live a long time and cure diseases among other things. And this whole thing is operating in a democratic and decentralized way, right? I think if anyone can pull something like this off, whether using the specific technologies I've mentioned or something else, I mean, then I think we have a higher odds of moving toward a beneficial technological Singularity rather than one in which the first super AGI is indifferent to humans and just considers us an inefficient use of molecules. </s><s>[INST] That was a beautifully articulated vision for the world. So thank you for that. [/INST] Thank you for that. Let's talk a little bit about life and death. I'm pro-life and anti-death. For most people, there's few exceptions and I won't mention here. </s><s>[INST] I'm glad just like your dad, you're taking a stand against death. By the way, you have a bunch of awesome music where you play piano online. Well, one of the songs that I believe you've written, the lyrics go, by the way, I like the way it sounds. People should listen to it. It's awesome. I was, I considered, I probably will cover it. It's a good song. Tell me why do you think it is a good thing that we all get old and die? It's one of the songs. I love the way it sounds, but let me ask you about death first. Do you think there's an element to death that's essential to give our life meaning? The fact that [/INST] this thing ends? Let me say, I'm pleased and a little embarrassed you've been listening to that music I put online. That's awesome. One of my regrets in life recently is I would love to get time to really produce music well. I haven't touched my sequencer software in like five years. I would love to rehearse and produce and edit, but with a two year old baby and trying to create the singularity, there's no time. So I just made the decision to, when I'm playing random shit in an off moment. Just record it. Just record it. Whatever. Maybe if I'm unfortunate enough to die, maybe that can be input to the AGI when it tries to make an accurate mind upload of me. Death is bad. I mean, that's very simple. It's baffling. We should have to say that. I mean, of course people can make meaning out of death. And if someone is tortured, maybe they can make beautiful meaning out of that torture and write a beautiful poem about what it was like to be tortured, right? I mean, we're very creative. We can milk beauty and positivity out of even the most horrible and shitty things. But just because if I was tortured, I could write a good song about what it was like to be tortured. It doesn't make torture good. And just because people are able to derive meaning and value from death doesn't mean they wouldn't derive even better meaning and value </s><s>[INST] from ongoing life without death. Which I very... So if you could live forever, that's awesome. It was like a death is bad. I mean, that's very which I very definite. Yeah. Yeah. So if you could [/INST] live forever, would you live forever? Forever. I, my, my goal with longevity research is to abolish the plague of involuntary death. I don't think people should die unless they choose to die. If I had to choose forced immortality versus dying, I would choose forced immortality. On the other hand, if I chose, if I had the choice of immortality with the choice of suicide, whenever I felt like it, of course, I would take that instead. And that's the more realistic choice. I mean, there's no reason you should have forced immortality. You should be able to live until you get sick of living, right? And that will seem insanely obvious to everyone, 50 years from now, and they will be so... I mean, people who thought death gives meaning to life, so we should all die, they will look at that 50 years from now, the way we now look at the Anabaptists in the year 1000, who gave away all their positions, went on top of the mountain for Jesus to come and bring them to the ascension. I mean, it's ridiculous that people think death is good because you gain more wisdom as you approach dying. I mean, of course it's true. I mean, I'm 53, and the fact that I might have only a few more decades left, it does make me reflect on things differently. It does give me a deeper understanding of many things, but I mean, so what? You could get a deep understanding in a lot of different ways. Pain is the same way. We're going to abolish pain, and that's even more amazing than abolishing death, right? I mean, once we get a little better at neuroscience, we'll be able to go in and adjust the brain so that pain doesn't hurt anymore, right? And that, you know, people will say that's bad because there's so much beauty in overcoming pain and suffering. Oh, sure, and there's beauty in overcoming torture too, and some people like to cut themselves, but not many, </s><s>[INST] right? I mean, to the ascension. That's an interesting, but to push back again, this is the Russian side of me. I do romanticize suffering. It's not obvious. I mean, the way you put it, it seems very logical. It's almost absurd to romanticize suffering or pain or death, but to me, a world without suffering, without pain, without death, it's not obvious what that [/INST] you can say in the people's Zoo. The people torturing each other. </s><s>[INST] I guess what I'm trying to say, I don't know if I was presented with that choice what I would choose [/INST] because to me, this is a subtler matter, and And I've posed it in this conversation in an unnecessarily extreme way. So I think the way you should think about it is what if there's a little dial on the side of your head and you could turn to how much pain hurt? Turn it down to zero, turn up to 11, like in spinal tap if it wants, maybe through an actual spinal tap, right? So I mean, would you opt to have that dial there or not? That's the question. The question isn't whether you would turn the pain down to zero all the time. Would you opt to have the dial or not? My guess is that in some dark moment of your life, you would choose to have the dial implanted </s><s>[INST] and then it would be there. And then it would be there. Just to confess a small thing, don't ask me why, but I'm doing this physical challenge currently where I'm doing 680 pushups and pull ups a day. And my shoulder is currently as we sit here in a lot of pain. And I don't know, I would certainly right now, if you gave me a dial, I would turn that sucker to zero as quickly as possible. But I don't, I think the whole point of this journey is, [/INST] I don't know. Well, because you're a twisted human being. </s><s>[INST] I'm a twisted, so the question is, am I somehow twisted because I created some kind of narrative for myself so that I can deal with the injustice and the suffering in the world? Or is this actually going to be a source of happiness? [/INST] Well, this is a, to an extent is a research question that humanity will undertake, right? </s><s>[INST] So I mean, human beings, humanity, right? [/INST] So I mean, human do have a particular biological makeup, which sort of implies a certain probability distribution over motivational systems, right? So I mean, we, and that is there, that is there. Now the question is, how flexibly can that morph as society and technology change, right? So if we're given that dial and we're given a society in which say we don't have to work for a living and in which there's an ambient decentralized benevolent AI network that will warn us when we're about to hurt ourself. You know, if we're in a different context, can we consistently with being genuinely and fully human, can we consistently get into a state of consciousness where we just want to keep the pain dial turned all the way down and yet we're leading very rewarding and fulfilling lives, right? No, I suspect the answer is yes, we can do that, but I don't know that for certain. Yeah, no, I'm more confident </s><s>[INST] that we could create a non-human AGI system [/INST] and that is there, I don't know that, which just didn't need an analog of feeling pain. And I think that AGI system will be fundamentally healthier and more benevolent than human beings. So I think it might or might not be true that humans need a certain element of suffering to be satisfied humans, consistent with the human physiology. If it is true, that's one of the things that makes us fucked and disqualified to be the super AGI, right? I mean, this is the nature of the human motivational system is that we seem to gravitate towards situations where the best thing in the large scale is not the best thing in the small scale according to our subjective value system. So we gravitate towards subjective value judgements where to graph ourselves in the large, we have to un-graph ourselves in the small. And we do that in, you see that in music, there's a theory of music which says the key to musical aesthetics is the surprising fulfillment of expectations. Like you want something that will fulfill the expectations of listed in the prior part of the music, but in a way with a bit of a twist that surprises you. And I mean, that's true not only in outdoor music like my own or that of Zappa or Steve Vai or Buckethead or Christoph Pundericke or something. It's even there in Mozart or something. It's not there in elevator music too much, but that's why it's boring, right? But wrapped up in there is, we want to hurt a little bit so that we can feel the pain go away. Like we want to be a little confused by what's coming next. So then when the thing that comes next actually makes sense. </s><s>[INST] It's so satisfying, right? That's the surprising fulfillment of expectations. Is that what you said? Yeah, yeah, yeah. So beautifully put. Is there, we've been skirting around a little bit, but if I were to ask you the most ridiculous big question of what is the meaning of life, [/INST] what would your answer be? Yeah, yeah, yeah. It's so beautiful. Three values, joy, growth, and choice. I think you need joy. I mean, that's the basis of everything. If you want the number one value. On the other hand, I'm unsatisfied with a static joy that doesn't progress, perhaps because of some elemental element of human perversity, but the idea of something that grows and becomes more and more and better and better in some sense appeals to me. But I also sort of like the idea of individuality that as a distinct system, I have some agency. So there's some nexus of causality within this system rather than the causality being wholly evenly distributed over the joyous growing mass. So you start with joy, growth, and choice as three basic values. </s><s>[INST] Those three things could continue indefinitely. That's not, that's something that can last forever. Is there some aspect of something you called, which I like super longevity that you find exciting? That's what, is there in research wise, [/INST] is there ideas in that space that? I mean, I think, yeah, in terms of the meaning of life, this really ties into that because for us as humans, probably the way to get the most joy, growth, and choice is transhumanism and to go beyond the human form that we have right now, right? I mean, I think human body is great and by no means do any of us maximize the potential for joy, growth, and choice imminent in our human bodies. On the other hand, it's clear that other configurations of matter could manifest even greater amounts of joy, growth, and choice than humans do, maybe even finding ways to go beyond the realm of matter that as we understand it right now. So I think in a practical sense, much of the meaning I see in human life is to create something better than humans and go beyond human life, but certainly that's not all of it for me in a practical sense, right? Like I have four kids and a granddaughter and many friends and parents and family and just enjoying everyday human social existence. Well, we can do even better. Yeah, yeah, and I mean, I love, I've always, when I could live near nature, I spend a bunch of time out in nature in the forest and on the water every day and so forth. So, I mean, enjoying the pleasant moment is part of it, but the, you know, the growth and choice aspect are severely limited by our human biology. In particular, dying seems to inhibit your potential for personal growth considerably as far as we know. I mean, there's some element of life after death perhaps, but even if there is, why not also continue going in this biological realm, right? In super longevity, I mean, you know, we haven't yet cured aging. We haven't yet cured death. Certainly, there's very interesting progress all around. I mean, CRISPR and gene editing can be an incredible tool. And I mean, right now, stem cells could potentially prolong life a lot. Like if you got stem cell injections of just stem cells for every tissue, if your body injected into every tissue, and you can just have replacement of your old cells with new cells produced by those stem cells, I mean, that could be highly impactful at prolonging life. Now, we just need slightly better technology for having them grow, right? So using machine learning to guide procedures for stem cell differentiation and trans differentiation, it's kind of nitty gritty, but I mean, that's quite interesting. So I think there's a lot of different things being done to help with prolongation of human life, but we could do a lot better. So for example, the extracellular matrix, which is the bunch of proteins in between the cells in your body, they get stiffer and stiffer as you get older. And the extracellular matrix transmits information both electrically, mechanically, and to some extent, biophotonically. So there's all this transmission through the parts of the body, but the stiffer the extracellular matrix gets, the less the transmission happens, which makes your body get worse coordinated between the different organs as you get older. So my friend Christian Schaffmeister, at my alumnus organization, my alma mater, the Great Temple University, Christian Schaffmeister has a potential solution to this, where he has these novel molecules called spiral ligamers, which are like polymers that are not organic. They're specially designed polymers so that you can algorithmically predict exactly how they'll fold very simply. So he designed the molecular scissors that have spiral ligamers that you could eat and would then cut through all the glucosapine and other cross-linked proteins in your extracellular matrix, right? But to make that technology really work and be mature as several years of work, as far as I know, no one's funding it at the moment. So there's so many different ways that technology could be used to prolong longevity. What we really need, we need an integrated database of all biological knowledge about human beings and model organisms, like basically, hopefully, a massively distributed open-cock bioadm space, but it can exist in other forms too. We need that data to be opened up in a suitably privacy-protecting way. We need massive funding into machine learning, AGI, proto AGI, statistical research aimed at solving biology, both molecular biology and human biology based on this massive data set, right? And then we need regulators not to stop people from trying radical therapies on themselves if they so wish to, as well as better cloud-based platforms for automated experimentation on microorganisms, flies and mice and so forth. And we could do all this. You look, after the last financial crisis, Obama, who I generally like pretty well, but he gave $4 trillion to large banks and insurance companies. You know, now in this COVID crisis, trillions are being spent to help everyday people in small businesses. In the end, we'll probably will find many more trillions of being given to large banks and insurance companies anyway. Like, could the world put $10 trillion into making a massive holistic bio AI and bio simulation and experimental biology infrastructure? We could, we could put $10 trillion into that without even screwing us up too badly, just as in the end, COVID and the last financial crisis won't screw up the world economy so badly. We're not putting $10 trillion into that. Instead, all this research is siloed inside a few big companies and government agencies. And most of the data that comes from our individual bodies personally that could feed this AI to solve aging and death, most of that data is sitting </s><s>[INST] in some hospital's database doing nothing, right? I got two more quick questions for you. One, I know a lot of people are gonna ask me, you're on the Joe Rogan podcast wearing that same amazing hat. Do you have a origin story for the hat? [/INST] Does the hat have its own story that you're able to share? The hat story has not been told yet. So we're gonna have to come back and you can interview the hat. So we'll leave that for the hat's own interview. It's too much to pack into it. </s><s>[INST] Is there a book? Is there a book? Is the hat gonna write a book? Okay. [/INST] It may transmit the information through direct neural transmission. </s><s>[INST] Okay, so actually there might be some neural link competition there. Beautiful, we'll leave it as a mystery. Maybe one last question. If you build an AGI system, you're successful at building the AGI system that could lead us to the singularity and you get to talk to her and ask her one question, [/INST] what would that question be? We're not allowed to ask, what is the question I should be asking. Yeah, that would be cheating, but I guess that's a good question. I'm thinking of a, I wrote a story with Stephan Bugai once where these AI developers, they created a super smart AI aimed at answering all the philosophical questions that had been worrying them. Like, what is the meaning of life? Is there free will? What is consciousness and so forth? So they got the super AGI built and it turned a while, it said, those are really stupid questions. And then it puts off on the spaceship </s><s>[INST] and left the Earth, right? Yeah, that would be cheating, but I guess that's a good question. So you'd be afraid of scaring it off. [/INST] That's it, yeah. I mean, honestly, there is no one question that rises among all the others, really. I mean, what interests me more is upgrading my own intelligence so that I can absorb the whole world view of the super AGI. But I mean, of course, if the answer could be like, what is the chemical formula for the immortality pill? Like, then I would do that or emit a bit string, which will be the code for a super AGI on the Intel i7 processor. So those would be good questions. </s><s>[INST] So if your own mind was expanded to become super intelligent, like you're describing, I mean, there's kind of a notion that intelligence is a burden, that it's possible that with greater and greater intelligence, that other metric of joy that you mentioned becomes more and more difficult. [/INST] What's your... Pretty stupid idea. </s><s>[INST] So you think if you're super intelligent, [/INST] you can also be super joyful? I think getting root access to your own brain will enable new forms of joy that we don't have now. And I think, as I've said before, what I aim at is really make multiple versions of myself. So I would like to keep one version, which is basically human like I am now, but keep the dial to turn pain up and down and get rid of death, right? And make another version, which fuses its mind with superhuman AGI, and then will become massively transhuman. And whether it will send some messages back to the human me or not will be interesting to find out. The thing is, once you're a super AGI, like one subjective second to a human might be like a million subjective years to that super AGI, right? So it would be on a whole different basis. I mean, at very least those two copies will be good to have, but it could be interesting to put your mind into a dolphin or a space amoeba or all sorts of other things. You can imagine one version that doubled its intelligence every year, and another version that just became a super AGI as fast as possible, right? So, I mean, now we're sort of constrained to think one mind, one self, one body, right? But I think we actually, we don't need to be that constrained in thinking about future intelligence after we've mastered AGI and nanotechnology and longevity biology. I mean, then each of our minds is a certain pattern of organization, right? And I know we haven't talked about consciousness, but I sort of, I'm panpsychist. I sort of view the universe as conscious. And so, you know, a light bulb or a quark or an ant or a worm or a monkey have their own manifestations of consciousness. And the human manifestation of consciousness, it's partly tied to the particular meat that we're manifested by, but it's largely tied to the pattern of organization in the brain, right? So, if you upload yourself into a computer or a robot or whatever else it is, some element of your human consciousness may not be there because it's just tied to the biological embodiment. But I think most of it will be there, and these will be incarnations of your consciousness in a slightly different flavor. And, you know, creating these different versions will be amazing. And each of them will discover meanings of life that have some overlap, but probably not total overlap with the human Ben's meaning of life. The thing is to get to that future where we can explore different varieties of joy, different variations of human experience and values and transhuman experiences and values. To get to that future, we need to navigate through a whole lot of human bullshit of companies and governments and killer drones and making and losing money and so forth, right? And that's the challenge we're facing now is if we do things right, we can get to a benevolent singularity, which is levels of joy, growth and choice that are literally unimaginable to human beings. If we do things wrong, we could either annihilate all life on the planet or we could lead to a scenario where say, all humans are annihilated and there's some super AGI that goes on and does its own thing unrelated to us except via our role in originating it. And we may well be at a bifurcation point now, right? Where what we do now has significant causal impact on what comes about. And yet most people on the planet aren't thinking that way whatsoever. They're thinking only about their own narrow aims and aims and goals, right? Now, of course, I'm thinking about my own narrow aims and goals to some extent also, but I'm trying to use as much of my energy and mind as I can to push toward this more benevolent alternative, which will be better for me, but also for everybody else. And it's weird that so few people understand what's going on. I know you interviewed Elon Musk and he understands a lot of what's going on, but he's much more paranoid than I am, right? Because Elon gets that AGI is gonna be way, way smarter than people. And he gets that an AGI does not necessarily have to give a shit about people because we're a very elementary mode of organization of matter compared to many AGIs. But I don't think he has a clear vision of how infusing early stage AGIs with compassion and human warmth can lead to an AGI that loves and helps people rather than viewing us as a historical artifact and a waste of mass energy. But on the other hand, while I have some disagreements with him, like he understands way, way more of the story than almost anyone else in such a large scale corporate leadership position, right? It's terrible how little understanding of these fundamental issues exists out there now that may be different five or 10 years from now though, because I can see understanding of AGI and longevity and other such issues is certainly much stronger and more prevalent now than 10 or 15 years ago, right? So I mean, humanity as a whole can be slow learners relative to what I would like, but on a historical sense, on the other hand, </s><s>[INST] you could say the progress is astoundingly fast. But Elon also said, I think on the Joe Rogan podcast that love is the answer. So maybe in that way, you and him are both on the same page of how we should proceed with AGI. I think there's no better place to end it. I hope we get to talk again about the hat and about consciousness and about a million topics we didn't cover. Ben, it's a huge honor to talk to you. Thank you for making it out. Thank you for talking today. [/INST] I really love it. Thanks for having me. This was really, really good fun. And we dug deep into some very important things. </s><s>[INST] So thanks for doing this. Thanks very much. Awesome. Thanks for listening to this conversation with Ben Gertzel. And thank you to our sponsors, The Jordan Harbinger Show and Masterclass. Please consider supporting the podcast by going to jordanharbinger.com slash Lex and signing up to Masterclass and masterclass.com slash Lex. Click the links, buy the stuff. It's the best way to support this podcast [/INST] and the journey I'm on in my research and startup. </s><s>[INST] And if you enjoy this thing, subscribe on YouTube, review it with Five Stars Nine Up podcast, support on Patreon or connect with me on Twitter, Alex Friedman spelled without the E, just F-R-I-D-M-A-N. I'm sure eventually you will figure it out. And now let me leave you with some words from Ben Gertzel. Our language for describing emotions is very crude. That's what music is for. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] shopping for you. And so you would have potentially federal support of that kind of thing. So these kinds of efforts [/INST] and identify where they're vulnerable people live. It's really challenging in multigenerational homes in La County, for instance, there's a lot of, of older people living together with younger people in relatively crowded there there, there, there, it's really quite a challenge. But there again, you can use resources. So if grandma uh, is worried that grandson has come home but is potentially being exposed, grandson calls grandma says, I mean, I might have been at a party where I might either, where covid was grandma calls public health, public health and says, okay, you can have this hotel room for a couple of days until you check turned </s><s>[INST] negative in case it wasn't clear. The idea of focus protection is the people that are vulnerable, protect them and everybody else goes on with their lives. Open up the economy, just do as it was before. [/INST] And there was still fear abroad. So there still would be some restrictions people would pose on themselves. They probably would go to parties. Last grandsons probably wouldn't go so many parties, right? Um, uh, that there would be less, less participation in big gatherings. Uh, and you may even save like big gatherings in order to, to restrict community spread again, I'm not against any of that. Um but you shouldn't be closing businesses, you shouldn't be closing churches and synagogues, you shouldn't be closing, you shouldn't be forcing people to not go to school. You should, you should not be shuttering businesses, You should just let just allow society go on some disease will spread. But as we've seen the lockdown didn't stop the disease from spreading anyways. </s><s>[INST] Right? So what do you make of the criticism that this idea, like all good ideas cannot actually be implemented in a heterogeneous society where there's a lot of people inter mixing and once you open it up, people like that, the younger people will just forget that this is even existing and they'll stop caring about the older people and mess up the whole thing and the government will not want to fund and you kind of the great efforts you're talking about, about food delivery and then the food delivery services will be like, why the heck am I helping out on this anyway? Because like it's not making me much money and so therefore like all good ideas, it will collapse. [/INST] That might be true. I mean, I think it's always a risk with policy thing, but I think like think back to the moment, but we actually felt like we were in this together to some extent, right? I think I think that that um that empathy that we had that was used to like tell people to stay in and like happily not not going happily, but like stay in um to to like wear a mask or two to do all these things that we thought would help other people could have been redirected to actually helping the people who </s><s>[INST] most needed to be. Especially I do remember March. So this is even way, way before Barrington, all all that kind of stuff. March april may there was a feeling like if if we all just work together will solve this and that maybe started to, when did that start breaking down? I mean unfortunately the election is mixed into this that it became politicized but I think it lasted quite a long [/INST] time. I think into the summer. I think there was some, some, some of that sense. I don't know, obviously varied among different people but but I think that it's true. It would have been challenging. It's also true. That is heterogeneous exactly the way you said. Um But what that means is you need a local response a response. So like, like my vision of a public health officer is someone that understands their community, not not necessarily the nation at large but their community and then works within their community to figure out how to deploy the resources that they have available. Um to do, to do the kind of protection policies we're talking about. That's that's what should have happened instead. They spent a huge amount of efforts closing, making sure businesses stay closed businesses that I mean there are, you know, like hardware stores that closed. What good did it closing a hardware store do for the spread of Covid? If it has an effect on spread? Covid spread, I mean, it's going to be checking to make sure that plexiglass was put up everywhere, which now in retrospect, turns out to probably made the disease worse. You know, masking enforcement. So shaming around mask, I mean, a huge amount of effort on things that were only tangentially related to focus protection. What if we turned our energy, </s><s>[INST] that [/INST] enormous energy put into that instead into focus protection of the vulnerable. That's essentially the conversation I was calling for. And it wasn't I mean, I didn't think of it as we had every single idea. I mean, we gave some concrete proposals, but the criticism we got was that those concrete proposals weren't enough. And the answer to that I have is that's true. They weren't enough. I wasn't thinking of them as enough. I was thinking that I wanted to involve an enormous number of people in local public health to help think about how to do focus protection in their communities. </s>

<s>[INST] My favorite app for learning new things bleakest, takes the key ideas from thousands of nonfiction books and condenses them down into just 15 minutes that you can read or listen to. I'm a big believer in reading every day. As part of that, I use blankets to try out a book. I mean otherwise never have a chance to read and in general, it's a great way to broaden your view of the ideal landscape out there and find books that you may want to read more deeply with Bling kissed, you get unlimited access to read or listen to a massive library of condensed nonfiction books. I also use blink ists short casts to quickly catch up on podcast episodes I've missed right now, Bling kissed has a special offer just for the listeners of this podcast. They say that to all the girls, go to blankets dot com slash legs to start your free seven day trial and get 25% off of a blink is premium membership. That's blankets dot com slash lex And now here's my conversation with eric Weinstein, you often talk about getting off this planet and I think you don't often talk about extra terrestrial life, intelligent life out there. Do you wonder about this kind of thing about intelligent civilizations out there? [/INST] I do, but I try to not wonder about it in a particular way, um in a certain sense, I do find that speculating about Bigfoot and the loch ness monster and space aliens is kind of a recreation for when things aren't going very well. Uh at least it gives us some meaning and purpose in our lives. So, I worry about, for example, the simulation hypothesis is taking over from religion. You can't quite believe enough to go to church or synagogue or the mosque on the weekend. So then you just take up an interest in uh in the simulation theory, because that's something like what you do for your job coding. I do think that in some sense, the issue of aliens is a really interesting one, but has been spoiled by too much sort of recreational escapism. The key question that I find is, let's assume that it is possible to look at the night sky and see all of these distant worlds and then go visit. If that is possible, it's almost certainly possible through some uh as yet unknown or not accepted theory of physics beyond Einstein, and I mean, it doesn't have to be that way, but probably is if that theory exists, there would be a percentage of the world that have life in sort of a drake equation, kind of a way that would have encountered the ability to escape soon enough after unlocking the power of the atom at a minimum, and whatever they have that is probably analogous to the cell on that world. So, assuming that life is a fairly generic thing that arises, uh probably not carbon based, probably didn't have DNA, but that something that fits the pattern of Darwinian theory, which is descent with variation, um differential success </s><s>[INST] and thereby constantly improving and so on. That through time there will be a trajectory where there'll be something increasingly complex and fascinating and beautiful like us humans, but much more [/INST] that can also off gas, whatever entropy it creates to give an illusion that you're defeating thermodynamics. Right? So, whatever, whatever these things are, probably has an analog of the by lipid layer, so that cells can get rid of the chaos on one side of the barrier and keep order on the other, whatever these things are, that create life, assuming that there is a theory to be found that allows that civilization to diversify. Um we would have to imagine that such a civilization might have taken an interest in its concept of the universe and have come here, </s><s>[INST] they would come here, they would have a deep understanding of the physics of the universe, sufficient to have arrived here. [/INST] Well, there's two questions whether they could arrive physically and whether their information could be sent here and whether they could gain information from us, it's possible that um they would have a way of looking into our world without actually reaching it. I don't know, but yes, if my hope, which is that we can escape this world is can be realized. If that's if that's feasible, then you would have to imagine that the reverse is true and that somebody else uh should be here. First of all, I want to say this my purpose when I come on to your show and I reframe the questions is not to challenge you. I can sit inside all of this is to give you better audio and video because I think we've been on an incredible roll. I really love what you do and so I am trying to honor you by being as disagreeable about frame breaking as possible. I think some of your listeners don't understand that it's actually a sign of respect as opposed to some sort of a complex dynamic, which is I think you can play outside of some of the frames and that these are sort of offerings to get the conversation started. So let me try to break that frame and give you something different. Beautiful. I think what's going on here is that um I can prove effectively that we're not thinking about this in very deep terms. As soon as I say, we've got to get off this planet, the number of people who assume that I'm talking about faster than light travel is very high and faster than light travel assumes some sort of Einsteinian paradigm that then is broken um by some small adjustment and I think that that's fascinating, it shows me that our failure to imagine what could be being said is profound. We don't have an idea um of all of the difference, different ways in which we might be able to visit distant worlds um all we think about is okay, it must be, it must be Einsteinian space times and then some means of exceeding the speed limit and it's just, it's fascinating to me that we don't really have, we've lost the ability to just realize we don't know the framework and what does, what does it even mean? So one of the things I think about a lot is worlds with more than one temporal dimension. It's very hard to think about more than one temporal dimension. </s>

<s>[INST] Yeah, it's quite interesting to see human civilization as a network, as a graph and our relationships as kind of edges in that graph. And to then do outside of just pandemic, do interesting inferences based on that this this is true for like twitter social networks and so on. How we expand the kind of things we talk about. Think about sort of politically, if you have this little bubble, quote unquote of ideas that you play with, it's uh nice from a recommend er system perspective, how do you jump out of those bubbles? It's really fascinating. Uh Youtube was working on that uh twitter is working on that but not always so successfully. But there's a lot of interesting work from a mathematical and a psychological sociological perspective there within those grass. But if we look at the cleanest formulation of that of looking at a problem from a different perspective, you are also involved with the International Mathematics olympiad. Which takes a small clean uh problems that are really hard, but once you look at them differently can become easy. But that little jump of innovation is a is the entire trick. So maybe at the high level, can you say, what is the international mathematical olympiad? [/INST] Sure. So this is uh the competition for people who aren't yet in college uh math competition, which is the most prestigious one in the entire world. It's the olympics of mathematics, but only for people who aren't yet in college now, the kinds of questions that they ask you to do Are not computational. Usually you're not supposed to find that the answer is 42 instead, you're supposed to explain why something is true. And the problem is that at the beginning, when you look at each of the questions, first of all, you have 4.5 hours to solve three questions and this is one day and then you have a second day which is 4.5 hours three questions. But when you look at the questions, they're all asking, you explain why the following thing is true, which you've never seen before. And by the way, even though there are six questions, if you solve any one of them, you're a genius and you get an honorable mention. So this is, this is hard. For example, heart problems. </s><s>[INST] What about is that one person is a team? [/INST] Uh so it's, each country can send six people and the score of the country is actually unofficial. There's not an official country versus country system, although everyone just adds up the points scores of the six people and they say, well now which country stacked up where? </s><s>[INST] Yeah, so maybe it's a side comment, I should say that there's a bunch of countries including the former soviet union and Russia where I grew up where this is one of the most important competitions that the country participates in. Like it was a source of pride for a lot of the country. You look at the olympic sports like uh wrestling, weightlifting, there's certain sports and hockey that Russia and the soviet union truly took pride in and actually the mathematical olympiad, it was one of them for many years, it's still one of them and that's kind of fascinating. We don't think about it this way in the United States, maybe you can correct me if I'm wrong, but it's not nearly as popular in the United States in terms of its integration into the culture, into just basic conversation into the pride. Like, you know, if you win an olympic gold medal or if you win the Super Bowl, you can walk around proud. I think that was the case with the mathematical olympiad in Russia, not, not as much the case in the United States, I think so, I just want to give that a little aside because beating anybody from Russia from the Eastern Republic or from china is very, very difficult, like if I remember correctly, you know, there's people, this was a multiyear training process, they train hard and this is, this is everything that they're focused on. My uh, my dad was, was uh was a participant in this and it's, I mean, it's uh it's as serious as olympic sports, you think about like gymnastics, like young athletes participate in gymnastics, this is as serious as that, if not more serious. So I just want to give that a little bit of context, because we're talking about serious, high level meth, uh athletics almost here. [/INST] Yeah, actually, I also think that it made sense from the soviet Union's perspective, because if you look at what these people do eventually, even though let's look at the U. S. S. R. S. Olympic, international math olympiad record, even though they say, even though they won a lot of awards at the high school thing, many of them went on to do incredible things in research mathematics or research other things and that's showing the generalization, generalized ability of what they were working on. Because ultimately we're just playing with ideas of how to prove things. And if you get pretty good at inventing creative ways to turn problems apart, split them apart, observe neat ways to turn messy things into simple crystals. Well, if you're going to try to solve any real problem in the real world, that could be a really handy tool to. So I don't think it was a bad investment. I think it clearly it worked well for soviet union. </s>

<s>[INST] Yeah I mean there's uh because I've even even for help with this this podcast here but also at M. I. T. And so on. I've done a bunch of hiring and uh I was always looking for you said brain teasers all kinds of simple questions that can they can reveal a lot of information and it's always been challenging. I used to I still ask this question but uh do you think it's better to work hard or work smart And you know I had this idea that I've that I think of matured about which is I kind of believe that people who say work smart on that question don't actually work smart. So the right textbook answer is what is better to work smart, but the reality is it's, it's people that haven't actually ever done anything that say work smart, they're like uh they haven't really struggled because my general belief at the time was in order to discover what it means to work smart to be efficient to you have to work your ass off, you have to really fail a lot and failure feels like hard work. And so I was always suspicious of people that would say work smart, I would want to interrogate that question, but then I also, you know, have, have learned that there is people that are just exceptionally um exceptionally efficient. They really do know what it means to work smart even at a young age. And so like you can't just disqualify based on that you have to dig in deeper. But some of the most interesting people I've ever worked with would say work hard unapologetically and they're usually the ones that know how to be efficient, which is, it's just getting interesting thing like that. And I've always searched for questions of that nature to see, can I, can I get, can I get a person to reveal something profound about them in as brief of a question as possible. And I, you know, and then of course there's basic attention in detail and brain teasers and stuff like that depending on the role programming and so on to see can they can they solve a tricky puzzle and do so like one that doesn't require a lot of effort but requires a certain um not only your way of thinking, is there is there [/INST] some, I </s><s>[INST] mean maybe you don't want to reveal, but is there some questions that you sometimes find yourself leaning on? She said like how did you solve the hard problem in your past and have them talk to it? That's one, [/INST] you know, we started with brain teasers periodically at Coinbase and we got we got away from that relatively quickly and I think one it's a tough one because I actually think it does show how somebody kind of performs under pressure, but it's I don't think it's a super reliable indicator because there's some people who are really good in what the typical work situation, but that's not a typical work situation where somebody puts you on the spot like in a live interview and sometimes people get nervous and they can't think clearly and like they don't have their computer in front of them or whatever they normally use. So yeah, I'm a little skeptical now of the of the brain teaser thing. Um there is a whole yeah, there is a whole question by like a lot of universities are getting rid of you know entrance exams. So if you're hiring right out of universities um sometimes it's becoming a less reliable indicator. Like are they in that university? And I've heard some companies we haven't done this yet, but I've heard some companies are actually creating their own, like for college grads, like their own um basically exams, like standardized testing almost to get people in the door because the degree almost like doesn't mean what it used to, which is the whole topic, but </s><s>[INST] yeah. Um It's fascinating both for that because it's not just about you trying to hire a great team. It's also to help them find the right place to work at it. It's like it's like a two way street. Um Alright, so once you found the product market fit, how did coin base become what it is today? So let me ask on an engineering question actually sort of from the ruby wallet days, what are some of the interesting challenges there or are they not engineering? Just the things that had to be solved? What were they engineering regulation? Um Financial hiring lawyers. What was it? [/INST] So post product market fit. Yeah, a lot of its scaling and you gotta build up build out an actual company. So I remember I was still writing a lot of code there for a while and we were, we were hiring in, we had like maybe 25 people or something. Remember one of our investors came by one day and he was like brian, how much of your time are you spending writing code? And I was like maybe maybe 50% or something. He was like how much time are you spending hiring people? I was like probably 20%. He was like, I think you need to flip those numbers like this company is not gonna scale, you're the ceo, you don't need to be writing code every day, you're gonna have to transition that stuff. Like even if people can't all that stuff locked in your head, so maybe they're not gonna do it as well as you for the first six months or something. But like if you don't start transition it, you're never going to build a real company. It's just gonna be, you're gonna be the bottleneck. So you know like a lot of founders that took me a while to like really internalize that lesson, I'd always heard people say that and you know, but I still was holding on too much to, to decision making and I probably still am by the way like even to this day at Coinbase where we continue to push down decision making in the or like even 5000 people like who are the owners of each of these things. And so the temptation is people to push it up and you become a bottleneck and anyway, um so yeah, you want, you basically need to make sure you have enough money where you don't die if there's in some, some kind of a downturn or hit break even profitability. We, we were in a position where we were periodically profitable during a period, but then crypto would go down and we were unprofitable and so we had to kind of manage our own psychology and the balance sheet to make sure we didn't like die in the downturn, which a lot of crypto companies did. Um we had to basically professionalize a whole bunch of services that had been just very quickly thrown together by like 20 year olds right? Whether that was cybersecurity, it's like okay how do you get like a really senior experienced cybersecurity person but not someone who's so senior that they can't get their hands dirty and they can come into a company with 25 50 people. How do you get a finance person to come in and do that? Our finances were a mess. Like we didn't even really know how much money we had at certain times and stuff. I mean this was embarrassing to say, but it was true. Like I remember there was a point where um we had raised, I think like our series C. Or something like that and I think we had our bank accounts, I just put like $25 million business, I was like. you know our operations were so messy and we needed to hire a new finance person and I was like this I've heard horror stories of actually startups where they thought they had X amount of money and then it turned out they had way less and then the whole thing was insolvent in like three weeks. So </s>

<s>[INST] you can go to war with black belts, but then you can also play around with the purple and the blue bells and so on and that was really beneficial for growth and you know, you can pick which because everybody has a style and you can pick which style you really want to work on, right. And then I came to boston broadway jitsu with john Clark who I love, he's a good friend but you know the, it's a little bit more formal and I found myself it's a very interesting journey if I would be training with black belts the whole time and uh it was a very different experience. I found myself exploring much less. I found myself uh learning much less. I mean part of that is um on me, but part of it was also realizing that uh, wow, there's a value to training with people that are much worse than [/INST] you, </s><s>[INST] Is there, is there a philosophy you could speak to on that. [/INST] Um you probably know it already um you know from your studies and artificial intelligence that all human beings are naturally risk averse. This is a bias which is deeply seated in all of us. I'm sure you're, you know, well read on people like diversity and et cetera, you talk about this all the time. Um for your viewers, there are numerous psychological experiments that are showing that most people, to the point of irrationality, fear loss more than they Are excited at the prospect of an equivalent gain. So for example, if you have $100 in your wallet, You're more worried about the idea of losing the $100 that you have now than you would be excited by the prospect of gaining $100 that I could potentially offer you. Um this comes out whenever you get a black belt versus black belt confrontations or any kind of similar skill level, whenever you get similar skill levels, the chances of defeat get very, very high, interestingly, if you're a white belt and you're going to get a black belt, you'll take risk. Why? Because there's no shame in losing to a black belt when you're a white belt, So you'll, you'll play more lightheartedly and you're having more fun role, but when you have very similar skill levels, you're going to come back to what the techniques that are most likely to get your win, that number of techniques is usually pretty small and if you're always battling with the same tough opponents every day where if you make even a single error, it will cost you that match inspiring. And you don't like losing, you're going to stay with a very small set of moves, You might get slightly better at their execution over time, but you, as an individual will not grow growth as it does in organic life forms. Comes from small beginnings and builds over time. You can't take an untested untried move and get on a world champion black belt, it's gonna get crushed, so it's not ready for that. It's like a lion cub being thrown out into the Serengeti Plains. The lion cub is just too small and too ineffective. It's a lion, but it's a cub and it's not until it grows into maturity that it can be a line that can dominate the Serengeti Plains. Um That's why I always encourage my students to play with a variety of belt types, um and spend the majority of their time with lesser belts for development purposes. When you're getting closer to a competition, you obviously want to change that. You want to be getting more competitive sense of hard work, but you must learn to divide up your training cycles and to non competition cycles where you're presumably working with people who are slightly lower level in yourself and in some cases quite a bit lower than yourself and then um, competition cycles where you're working with people much closer to your own skill level, </s><s>[INST] is there something to be said about the flip side of that? Which is um when you're training with people at the same skill level being okay losing to them. [/INST] Yes, you have to see training for what it is. Training is about skill development, not about winning or losing. You've got to you've got to understand that you don't need to win every battle. You only need to win the battles that count and the battles that count are in the World Championship Finals. Okay, that's the one that counts. Think about that win. Okay, that's the one you're gonna be remembered for. You're not going to be remembered for the battle you lost on Tuesday afternoon at three p.m. And some nameless jim with some guy that no one cares about, no one's gonna remember that you're going to be remembered for your peak performances, aren't your everyday performances. Focus your everyday performances on skill development so that your peak performances, you can focus on winning. </s><s>[INST] You know, I just this is not a therapy session, but if I could just speak [/INST] every sessions of therapy session, </s><s>[INST] there is still an ape thing in there. [/INST] Of course you think, I don't feel it. You think everyone in the room doesn't feel it </s><s>[INST] because for example, you have never seen your role uh you know, when there's people, you know, I've seen the look in people's eyes when they see me train and they could see maybe it's me projecting, but they think I thought you were supposed to be good that you're supposed to be a black belt [/INST] like that. Look, they're like study. I'm going to give you some therapy. Okay. Do you know how many people have come up to me over the years who have visited the training halls that I work in and they come up to me to go, man, I rolled with Gary tonin. I did really well with him like, like really well. Oh my God, it's very, very good, very impressively. And then I see them talking to their friends like man, I tapped out Gary toner and I'm sitting there going, yeah, and you can see that they're just like, </s>

<s>[INST] What where would you put love on? [/INST] I would not say he's a specialist. I'd say his topical is the strongest move, topical is the strongest move. Yeah. Uh And and the the interesting thing about the the specialist versus the blender, there's a counter, right? Every move has a move that theoretically should be the right choice. So if you're a single move guy, uh there's going to be a guy out there who will get you. Yeah, it'll be very difficult for you to beat that guy. Um But like when you come to like a tournament, typically specialists do much better in tournament scenarios because their singular move can get them through a tournament very quickly and efficiently. Whereas you get a blender in a tournament, they typically will have longer and more difficult matches and by right. Yeah, yeah, but in super match format, typically blenders do better. </s><s>[INST] So we uh offline also talked about arm sumo or uh freedom arm wrestling, I don't know how you want to call it, but freedom uh Exactly North american way. Uh So what, so this is this is this idea and I watched a few videos and it looks fun, is basically removing the restriction of having to keep your elbow on the pad and just being able to arm wrestle over the whole table. I think you've mentioned that the criticism that gets is it might be injury prone or something like that. So, can you describe this arm sumo freedom arm wrestling idea [/INST] when you come to freedom arm wrestling, basically, it removes the limitation of a standard arm wrestling table. Right? So basically every single thing is a freedom arm wrestling table. Some are better than others. All right, So, looking for that nice table where we can kind of stand apart from each other and were anatomically, you know, in a fairly safe position. And the rules and freedom, the way you win is like the knuckles must either touch the tabletop or you hold it off the edge for for a three count, Right? So this is the main way to win. Yes, you can foul, like if you lift your eyeball up, it's still a foul, but you have the entire playing surface, so your elbow is no longer limited to Eur seven by seven or seven by nine pad. So you can move it all over the table. You can move your body around the table a </s><s>[INST] bit too. And if it's a big table, your body can largely be on the [/INST] table. Yeah, so basically, it's like adjusting your ring size. So arm wrestling, you're fighting in a phone booth, right? So you get your fighting in a field, you're fighting, you know, just bigger. So it just it just makes the sport bigger. Yeah, this is Japan, </s><s>[INST] but even on a small table there, even in a slightly larger phone booth, you can get a a lot more fun and variety. [/INST] Very interesting. I love it. I think it makes the sport bigger. I actually believe that it's the future of the sport. I really do </s><s>[INST] because it makes it more accessible. Like you don't need the equipment to do it at the bar, all that kind of stuff. [/INST] Less, less equipment requirements. Kids, most kids start freedom. Like most kids aren't wrestling school desks. Yeah. And like if you see a guy on the street and you're like whatever, like you cannot wrestle anywhere. You don't need to bring your table around with you </s><s>[INST] if uh we talked about the elite level, if somebody was interested in starting an arm wrestling or like going from just like, you know, I go to the gym, you kind of lift, you've arm wrestled a few times trying to get better at it, trying to learn how do you advise? Like getting better to where you can beat your closest buddies, that first step, [/INST] first step, I'd say fine. People find people find good people volume get, we'll get with the club, get with people who know what they're doing, who can mentor you </s><s>[INST] and that's really cool. I got to meet, I realized there's a club in Austin, sure there's been a lot of places [/INST] where we got this app called arm back. Mhm </s><s>[INST] Yeah, which is a app that helps you find other people. They're [/INST] very easy, but I mean, they're all over social networks. I mean, it's kind of widespread now, but yeah, find people, find people and it's just much easier to learn with another person and you'll get stronger that way. But I mean, do this, do the lifts. If you go to the gym, just start doing the lifts uh and right away, those will technically prepare you. </s><s>[INST] What are what are the lists can we describe? [/INST] So, I'd say if you want to just keep it very very simple. Let's just talk about three. There's more there's much more than three. But like when you talk about energy allocation, these three lifts in my opinion should be like 90% of your investment is very big. These three lifts. Ah And the exact percentages can you can argue about it, but we'll we'll we'll start off with the cupping of the wrist. Just this, it's a simple thing and do it with a cable. Uh You know, you can get a thicker diameter. So it kind of, you know, is more on your fingers where an armrest is going to attack you, right? Because any good arm rest is gonna attack your fingers. So like </s><s>[INST] open hand. [/INST] No, no. Well, I mean, for health, yes, you could. But like if you want to be really specific, you train exactly the way you would at a table in the position that you actually start that </s><s>[INST] hatch and then you're just doing this kind of [/INST] to your center. Your people people, one of the, one of the big misconceptions in arm wrestling is that you're aiming for that pin pad. </s>

<s>[INST] but eventually you just have to be a Scientologist. So you mentioned the eight dynamics, but you also mentioned survival. So that seems to be a core principle that human existence is about survival. Can you elaborate what is meant by survival? We're talking about survival of the human species, survival of the individual humans, survival of the, the manifestation of thetans in human form. [/INST] What do we, what's survival? So it would be all of that because survival is the, the dominant force across all the dynamics that, I mean, Elrond Hubbard, it was either Dianetics or Science of Survival. He says he discovered the principle upon which all life exists. And that is all life, no matter what it is trying to do. Are you ready, Lex? </s><s>[INST] It's trying to survive. That's pretty powerful. That's pretty powerful. I gotta tell you, I gotta- You might get me back in, Lex. No, I'm not, I'm not trying to get you back in. I'm trying to take you, get you to take seriously the power of the ideas behind Scientology, because I think those ideas are not bad ideas. They resonate with a lot of ideas throughout philosophy, throughout religions, throughout the history of human civilization. [/INST] The interesting aspect is how it goes wrong. But here's the thing, Lex, here's the thing. It is consistent with prior efforts or studies. It's just that Elrond Hubbard said this was a watershed breakthrough that it was being discovered for the first time. </s><s>[INST] That's kind of what I'm mocking, really. Yeah, but you can mock Nietzsche for saying man has will to power. You can mock Roy. But did he claim to be the first person to ever say it? Well, Nietzsche had a bit of an ego, so. Like, and he's full of contradictions, but I'm pretty sure the imply thing is that he was the first to say it. There's a lot of scientists. There's one of the people I really admire, Stephen Wolfram, who wrote a book called A New Kind of Science that explores complex systems and cellular automata and these mathematical systems that have been explored before. But he boldly kind of defined, I am presenting to you a whole new way to look at the world. And if you just set a little bit of the ego behind that aside, there's actually beautiful ideas in there. They have of course been done before and explored before, but sometimes people declare this is the coolest. [/INST] That's the only thing I'm really mocking is that this discovery that life is trying to survive is greater than the discovery of fire. I mean again, it's a little silly. But that's fine. We can agree that the fact that life is trying to survive has meaning and is meaningful and can be, is valuable. And it's true. </s><s>[INST] I mean, life is trying to survive. Also, there's a non-trivial definition of what is life here? So this idea of a thing, the thing that permeates through lifetimes through people, there's some fabric that is bigger than the individual biological bags of meat. That's a philosophically-interesting idea. Of course, if it's not grounded in a little bit more [/INST] of physical reality, then it becomes a little too woo-woo. And the way L. Ron Hubbard in Scientology defines survival is very much intertwined with how they define ethics. Ethics, anything to be ethical is pro-survival. To be unethical is counter-survival. But we were talking about just the concept of the dynamics, like what does survival refer to? And it actually does refer to all of them, but just keep in mind, when it comes to the seventh dynamic, Thetans collectively, involved in here is the idea that a Thetan cannot die. There's no such thing as killing a Thean. A Thean can only survive. And so, anyway, this concept of the dynamics is one of the most fundamental and important concepts in Scientology. But because I mentioned that it also gets tied up with ethics, and this probably speaks to what you're just talking about is you can have the ideas and the concepts and you can have how do they go wrong. Because they hold that Scientology, applying Scientology, getting people into Scientology is the key to basically saving every spiritual being in existence. When you're analyzing what is ethical, it becomes whatever's good for Scientology becomes by definition ethical, because anything that's good for Scientology, which is a third dynamic, is inherently good for all the dynamics. So that's where you get the ends justifying the means to do anything possible, use any means necessary to afford the aims of Scientology. </s><s>[INST] That's kind of where a lot of Soviet implementation of communism went wrong, is the ends justify the means. The equality, the justice for the workers, if we have to kill, murder, imprison, censor, in the name of that, then it's for the greater good [/INST] in the long term to achieve the ideal of communism. In some respects, Scientology created a near-perfect communist experiment in its organization. What is it, from everyone according to their ability to each according to their need, or something like that? Scientology, C Organization, is damn near a perfect communist experiment. Coming from someone who doesn't necessarily know what a perfect communist experiment really is, because I'm not grew up in a cult, Lex. You can't keep it as an excuse. It's a funny tagline I use in my videos. It is interesting that an organization that is so hyper capitalist and so money hungry and is known to be very wealthy at its core is run by these group of Syorg members that live a communist lifestyle. </s><s>[INST] You can't keep using it as we're gonna jump around. Let's go, what is C or What is C organization? What is this organization? [/INST] Yeah. What is this organization? The organization is the most dedicated version, the most dedicated brand of Scientologists. So there's three like echelons of Scientologists, there's public who just live normal lives in the real world and they pay to do Scientology courses and auditing. Then there's staff members who also live in the real world, but work on two and a half year contracts or five year contracts at their local Scientology organization and then once they finished their contract, their debt is paid or whatever, and then there's the seorg members, these are the guys who signed the billion year contracts. They don't have lives in the outside world. They don't own property, they live in Scientology provided housing, they eat Scientology run cafeterias. </s>

<s>[INST] And all of that was stuff you were explicitly having a discussion about like those designs and argument. You were having like arguments discussions. I'm [/INST] sure it was arguments. I mean I'm not sure if you've met my co founder mike but he's a pretty nice guy. He's very reasonable and uh and we both just saw eye to eye and we're like yeah it's like make this fast. Really seem fast. It'll be great. I mean honestly I think the most contentious thing and he would say this to initially was um I was on an iphone three G. So like the not so fast one. He had a brand new iphone four. I was </s><s>[INST] cheap. [/INST] Um And his feet loaded super smoothly. Like when he would scroll from photo to photo buttery smooth right? But on my phone every time you got to a new photo it was like a chunk ka chunk allocate memory like all this stuff right? I was like mike that's unacceptable. And he's like oh come on man just like upgrade your phone basically. You didn't actually say that he's nicer than that but I could tell he wished like I would just stop being cheap and just get a new phone. But what's funny is we actually sat there working on that little detail for a few days before launch and that polished experience plus the fact that uploading seemed fast for all these people who didn't have nice phones I think meant a lot because far too often you see teams focus not on performance, they focus on what's the cool computer science problem they can solve right, can we scale this thing to a billion users and they've got like 100 right? Yeah. You talked about loss functions So I want to come back to that. Like the loss function is like do you provide a great, happy, magical, whatever experience for the consumer and listen, if it happens to involve something complex and technical and great. But it turns out I think most of the time those experiences are just sitting there waiting to be built with like not that complex solutions but everyone is just like so stuck in their own head that they have to over engineer everything and then they forget about the easy </s><s>[INST] stuff and then also maybe to flip the last function there is you're trying to minimize the number of times, there's unpleasant experience, right? Like the one you mentioned where it, when you go to the next photo, it freezes for a little bit. So it's almost as opposed to maximizing pleasure. It's probably easier to minimize the number of like the friction. [/INST] Yeah. And as we all know, you just, you just uh, you just make the pleasure negative and then minimize everything. So </s><s>[INST] we're mapping this all back and your networks. [/INST] But actually can I say one thing that, which is, I don't know a lot about machine learning but I feel like I've, I've tried studying a bunch that whole idea of reinforcement learning and planning out more than the greedy single experience I think is is the closest you can get to like ideal product design thinking where you're not saying, hey, like can we have a great experience just this one time? But like what is the right way to onboard someone, what series of experiences correlate most with them hanging on long term. Right? So not just saying, oh did the photo load slowly a couple times or did they get a great photo at the top of their feed? But like what are the things that are going to make this person come back over the next week? Over the next month? And as a product designer asking yourself, okay, I want to optimize not just minimize bad experiences in the short run, but like how do I get someone to engage over the next month and I'm not going to claim at all that I thought that way at all at the time because I certainly didn't, but if I were going back and giving myself any advice would be thinking, what are those, what are those second order effects that you can create? And it turns out having your friends on the service enormous wind. So starting with a very small group of people that produce content that you wanted to see which we did, we seeded the community very well I think ended up mattering and </s><s>[INST] so yeah, you said that community is one of the most important things. So it's from a metrics perspective from maybe a philosophy perspective building a certain kind of community within the app. So I wasn't sure what exactly he meant by that when when I heard you say that maybe you can elaborate but as I understand now it's can literally mean get your friends onto the app. [/INST] Yeah, I think of it this way you can build an amazing restaurant or bar or whatever right? If you show up and you're the only one there is it like doesn't matter how good the food is, the drinks whatever. Um no um these are inherently social experiences that we were working on. So the idea of having people there like you needed to have that otherwise it was just a filter out. But by the way part of the genius, I'm going to say genius even though it wasn't really genius was starting to be marauding is a filter app. It was awesome. The fact that you can, so we talk about single player mode a lot which is like can you play the game alone and instagram, you could totally play alone, you can filter your photos and a lot of people would tell me, I didn't even realize that this thing was a social network until my friend showed up. It totally worked as a single player game and then when your friends showed up all of a sudden it was like, oh not only was this great alone, but now I actually have this trove of photos that people can look at and start liking and then I can like there's, and so it was this bootstrap method of how do you make the thing not suck when the restaurant is empty? </s>

<s>[INST] Yeah, like, is there advice that you give to Liv [/INST] how to get the videos done faster? Well, it would be the same exact thing. I mean, actually, I can give good procrastination advice. Panic monster, rate of panic monster. Yeah, well, we should do it together. It should be like, we have this date, but, you know, it's, yeah. We should actually just do another bet. I have to have my script done by this time. Yes. Yeah, so I gotta get the third part out. Because then you'll actually do it. And it's not, the thing is, the time in, but it's like, if you could take three weeks on a video and instead you take 10 weeks, it's not like, oh, well, I'm also, I'm having more fun in those 10 weeks. The whole 10 weeks are bad. Yeah, I told you. Bad, so you're just having a bad time and you're getting less work done and less work out. And it's not like you're enjoying your personal life. It's bad for your relationships. It's bad for your own. You keep doing it anyway. Yeah, well, a lot of people, why do people have troubles keeping a diet, right? </s><s>[INST] Primitive mind, primitive mind. Why'd you point at me when you said that? That was offensive. What's your procrastination weakness, do you have one? Everything, everything, everything. Everything, preparing for a conversation. I had your book, amazing book. I really enjoyed it. I started reading it. I was like, this is awesome. It's so awesome that I'm going to save it when I'm behind the computer and can take notes, like good notes. Of course, that resulted in like last minute, everything, everything, everything I'm doing in my life. [/INST] Not everyone's like- What's your procrastination weakness, do you have one? Everything, everything. Not everyone's like that. People self-defeaten different ways. Some people don't have this particular problem. Adam Grant is a, he calls himself a pre-crastinator where he gets an assignment. He will go home and do it until it's done and handed it, which is also not necessarily good. It's like you're rushing it either way, but it's better. But some people have the opposite thing where they will, the looming deadline makes them so anxious that they go and fix it. The procrastinator, I think, has a similar anxiety, </s><s>[INST] but they solve it in a totally different way. Or they don't solve it. They just live with the anxiety. [/INST] Right, right, right. They just live with the anxiety. Now, I think there's an even bigger group of people. There's these people, the Adam Grant's, there's people like me. And then there's people who have a healthy relationship with deadlines, but they're still part of a bigger group of people that actually, they need a deadline there to do something. So they actually, they still are motivated by a deadline. And as soon as you have all the things in life that don't have a deadline, like working out and working on that album you wanted to write, they don't do anything, either. So there's actually like, that's why procrastination is a much bigger problem than people realize, because it's not just the funny last second people. It's anyone who actually can't get things done </s><s>[INST] that don't have a deadline. You dedicate your book, quote, to Tannis, who never planned on being married to someone who would spend six years talking about his book on politics, but here we are. What's the secret to a successful relationship with a procrastinator? [/INST] That's maybe for both of you. </s><s>[INST] Well, I think the first and most important thing. You already started with a political answer, [/INST] I can tell, okay, go ahead. No, the first and most important thing is, because people who don't procrastinate, it's like people in their instinct is to judge it as like, they think they're just being like a loser, or they'll take it personally, and instead to see this as like, this is some form of addiction or some form of ailment. They're not just being a dick, right? They have a problem and some compassion, but then also maybe finding that line where you can maybe apply some tough love, some middle ground. On the other hand, you might say that you don't want the significant other relationship where it's like they're in the one nagging you. You don't want them even being part of that, and I think that maybe it's better to have a live do it instead. Right. Having someone who can create the infrastructure where they aren't the direct stick. You need a bit of carrot and stick, right? Maybe they can be the person who keeps reminding them of the carrot, and then they set up a friend group to be the stick, </s><s>[INST] and then that keeps your relationship in a good place. Yeah. In a good place. Like looming in the background. That's your friend group. Okay. At the beginning of the conversation, we talked about how all of human history can be presented as a thousand page book. What are you excited about for the 1,000th, what do you say, that first page? So the next 250 years. [/INST] What are you most excited about? I'm most excited about, have you read The Fable of the Dragon? Okay. Well, it's an allegory for death and it's, you know, Nick Bostrom. And he talks about the, he compares death to a dragon that eats 60 million people or whatever the number is every year. And you just, every year we shepherd those people up and they feed them to the dragon and that there's a Stockholm syndrome. When we say that's just a lot of man. And that's what we have to do. And anyone who says maybe we should try to beat the dragon, they get called vain and narcissistic. But someone who tries to, someone who goes, does chemo, no one calls them vain or narcissistic. They say they're, you know, good, good for you, right? You're a hero, you're fighting the good fight. So I think there's some disconnect here. And I think that if we can get out of that Stockholm syndrome and realize that death is just the machine, the human physical machine failing, and that there's no law of nature that says you can't, with enough technology, repair the machine and keep it going until, I don't think anyone wants to live forever. People think they do, no one does. But until people are ready. And I think when we hit a world where we can, we have enough tech that we can continue to keep the human machine alive until the person says, I'm done, I'm ready. I think we will look back and we will think that anything before that time, that'll be the real AD, BC. You know, we'll look back at BC before the big advancement and it'll seem so sad and so heartbreaking, barbaric. And people will say, I can't believe that humans like us had to live with that when they lost loved ones and they died before they were ready. I think that's the ultimate achievement, but we need to stop criticizing </s>

<s>[INST] The blood data, DNA data, fitness tracker data all to provide you with a customized personalized advice for diet and lifestyle changes. This is the future. The decisions about your body I think should be based on data from your body, not on generic population data. Let's go to john Mayer and say that your body is a wonderland. Let's go with that. It's a unique wonderland. And so lifestyle [/INST] changes </s><s>[INST] should be based on data from that unique wonderland that is your body. Go to inside tracker dot com and for a limited time you get special savings for being a listener of this very podcast. This is a luxury brand podcast support it. Please check out our sponsors in the description and now Dear Friends, here's Susan cain, you've written on your website that quote, I prefer listening to talking, reading, to socializing and cozy chats to group settings. So I think this conversation in the podcast is gonna be fun. What's a good definition of an introvert is something like those three things? A good start. [/INST] It is a good start in terms of how introverts experience day to day </s><s>[INST] life. [/INST] I think a good definition is one that some of your listeners will have heard many times before. The idea of where do you get your energy and for some people they get their energy more from quieter settings and for other people they get it more from being out there. So a good rule of thumb is to ask to imagine that you're at a party that you're really enjoying and you've been there for about two hours or so and it's with people you really like and it's in your favorite place. It's all good an extrovert in a setting like that is going to feel charged up and they're going to be looking for the after party and an introvert, no matter how good a time they're having and how socially skilled they are. There's this moment where you just wish that you could teleport and be back at home </s><s>[INST] and at the time before the start of the party to the the time when that moment happens is different for different people. So like the short of that is the more of an age where you are, is that that kind of thing? [/INST] The shorter the moment until you get to the place where you gotta teleport. </s><s>[INST] Yeah [/INST] and then for extroverts it's it's the opposite right? Like they're going to feel, you know, maybe they're working on I don't know like focused on producing a memo that's really intensely interesting to them. But if they're in that state of like solitary, the solitary mode of really focusing, they might get stir crazy a lot faster than an introvert would. And so it doesn't have so much to do with what you're good at as how you get your energy. </s><s>[INST] And so for an introvert the source of energy is what silence solitude and for an extrovert, its interaction with other people, [/INST] what I'd really say is that and this is neurobiological as well is that it has to do with how your nervous system reacts to stimulation. So for an introvert you're feeling in a great state of equilibrium when there are fewer inputs coming at you. So they could be social inputs. But that's why an introvert in general would rather hang out with one close friend at a time as opposed to you know big party full of strangers because that's too many inputs for the nervous system and for an extrovert, the nervous system needs more stimulants. So if they're not getting enough they get that listless and sluggish feeling. </s><s>[INST] So if you're just walking to the world, like people listening to this, but in general, how do you know if you're an introvert, like how do you empirically start to determine if you are in large part and introvert? [/INST] Well, I would start by just asking that question of what happens to you, you know, around the remark where you're </s><s>[INST] having a good time. Yeah, [/INST] like I mean imagine. But I also find curious if you have a different experience from this, but from all the years that I've been out there talking about this topic, I found that most people really seem to know once they're being honest with themselves. And maybe that's the question to ask is like if you imagine that you have a saturday or a whole weekend where you can spend your time exactly the way you want to with no professional obligations and no social obligations, who would you spend it with? How many people, what would you be doing and what is that picture that you're painting start to look like. </s><s>[INST] Yeah. So there's nuance to this though because I'm sure for extroverts to get energized by stimulation whether that stimulation with other people, like it depends what that stimulation is, right? Like maybe you're not surrounded by the kind of people that you enjoy being around. So you know um maybe that has to do less with whether some characteristics of your personality more has to do with the fact like what your environment is like that's always kind of the question, do you want to be alone because everybody around you is an asshole or do you want to be alone because you get energized from being? [/INST] Well I would hold the variables constant I guess I would say keep </s><s>[INST] the assholes constant and see um And then there's the other thing you kind of observed that there's a lot of people that will say they get energized from being alone. Like people are exhausting to them or something like that, but at the same time when you see them at a party they seem like the life of the party. [/INST] I know and I hear from those people all the time, there are so many people like </s><s>[INST] what would you classify them as exactly? Is it ultimately is the source of energy is the most important thing or like how the heck are they the life of the party? [/INST] It's a bunch of different things, you know? So first of all, just to say like a big caveat to all of this is humans are just amazingly complex. So you can't like explain every individual humans through these parameters even though I think the parameters are really valuable but that person at the party, it could be that they're more of an amber vert so they kind of are more in the middle of the spectrum. That's, it basically means someone who's not extremely introverted or extremely extroverted, they're kind of in the middle, so maybe at a party they're more extroverted side comes out. Um or it could be an introvert who's gotten really good at the skills of acting more like a pseudo extrovert and they pull that up at the moment that they need it. </s><s>[INST] So they've learned how to fake it. [/INST] Oh there's a lot of people like that and I know this because I think out of all the people on this planet you could be talking to, I've heard from the most number of those people like they will come and tell me about their experience out in the world, presenting a face that's different from what they feel. </s><s>[INST] So one of the things you talk about is at least in the West we've constructed a picture of success in that picture is usually one of an extrovert. Like when you imagine somebody who's a leader who is a successful person, that person has some of the qualities you would associate with an extrovert. And so there's a lot of incentive for faking it. [/INST] Yeah, </s><s>[INST] if you want to be successful, you got to be able to to fake it to sort of hang with the rest of the team, you have to, you have to be able to be outgoing and all those kinds of things and not be drained by the interaction. [/INST] Yeah, but I mean there are also a lot of introverts who figure out ways to draw on their own strengths and they're incredibly connecting and successful and they're great leaders and they're not actually faking it, they're more just figuring out ways to do it their own way. You see a lot of people like that </s><s>[INST] is their advice, is their lessons, you can draw from that from just observing how you can be an introvert and be in a leadership position. [/INST] Yeah, it's kind of like a mantra of figuring out what your own strengths are and how to draw on them. Um I think of uh a guy I know Doug Conant who had been the ceo of Campbell's soup for many years, he's very introverted, he's quite shy also by his own description and um and he really cares about people and so when he started At Campbell the employee engagement ratings of the company were all the way at the bottom of the Fortune 500 And by the time he stepped down 10 years later they were all the way at the top and he was, it wasn't that he was going out there and schmoozing people, but he really did care. So he would find out who are the people who had really been contributing and he would write to them personal letters of thanks and </s><s>[INST] and people [/INST] And these letters meant so much to people. They would carry them around with them. And during his time, the 10 years there, he wrote 30,000 of those letters. So like that was his way of doing it. That was his way of drawing on his own strengths and you know, like he did that together with of course you sometimes have to go outside your comfort zone no matter who you are. So he was doing plenty of that too. It's kind of combination. </s><s>[INST] Yeah, the writing process and focusing on the one on one interaction I can definitely relate. There's something deeply draining which concerns me about like zoom meetings because it's some weird brain manipulation. Well because you're not really engaged, but it wears on you the same way that there's a party, it feels like you're emptying that bucket for the introverts, even though they're not participating at all in the meeting. I mean, I suppose that's true for physical meetings too. But with zoom meetings, remote meetings, it's so much easier to invite a larger number of people into the meeting. So you're draining more and more of the introvert energy and I'm probably expert too. But the introvert, definitely. There's I mean it's interesting, I would love to understand that more because there's more and more push towards remote work without I think a deep understanding of why these meetings are so draining on people. I just anecdotally I've heard from that, but maybe that's because the managers, the people who arrange the meetings are just not sufficiently yet aware of the draining nature of them so that they pull into many people, they schedule them to regularly so they need to adjust that kind of thing. Probably, [/INST] I think people are starting to realize, but I would say one reason that zoom is so draining is because you can see your own self presentation the whole time if you choose to and when you go into in person you can't so you're kind of free to thinking about that. So it's like an extra cognitive load that you're bearing the whole time. </s><s>[INST] Oh yeah, [/INST] like you might want to, you know, turn off the camera so you can't see yourself, but then you feel like, well I have the ability to so I probably should be doing it and then that alone is a decision that you're making. Yeah, </s><s>[INST] that's probably studies on this now happening either have happened or are happening. The effect of seeing your own face on on camera because it's reminding you that you're supposed to be acting a certain way. And that's that is especially a stressful thing. [/INST] Yeah, you can't be in the moment as much. But I mean for you, you you make the decision to do all your podcast interviews in person, right? And so </s><s>[INST] and that's even when it's very costly if there's any kind of chemistry that contributes at all to the conversation, which I think most conversations have come to even the boring work meanings, there's something there because yes, you're trying to solve a particular problem at this particular time, but underneath it, there's a team building that's happening. And [/INST] honestly, </s><s>[INST] people also have told me about this, why they enjoyed the zoo meetings during the pandemic. It's like they're lonely, [/INST] like </s><s>[INST] they, you know, it's annoying to have to sit and listen to folks talk about nothing and so on, but they tune in anyway because it's kind of lonely to sit there by yourself and that I mean there's a deep connection there when you're with other people and that is especially true when they're they're in personal, which is a huge concern for me for like more and more offices from a capitalist perspective realizing, hey, why are we, why do we have these large office spaces? Why do we have to get people together? But I think in some, in some deep sense we do. But then you also talk about that once we do we want to protect the introverts, like you don't want the open space, which office space, which was a big fad for a while. I don't know where people stand on that point. [/INST] I think people are figuring it out in a post pandemic context, but I mean I know what you mean. So Before I became a writer, I was a corporate lawyer for like seven years and literally the only thing I miss from those years is hanging out with people at the office like I don't know just some of the funniest moments I've had in my life came from being at the office until midnight with the other people I was working with. So I know exactly what you're talking about. So I will say the offices there at that firm and at most firms in those days everybody had their own office. So it was like a dorm room. You know where it was like a long hallway with everybody in their own little dorm room. Um So you had tons of privacy but you would also come out and you can </s><s>[INST] roam whenever you want. [/INST] Yeah </s><s>[INST] and whenever you're Rome that means you're kind of open, you're looking for trouble, [/INST] your </s><s>[INST] interaction [/INST] and the extent to which you would keep your door open. You know. Is it wide open or was it half a jar or just a little bit? Those are all signals. </s><s>[INST] So is there because you said reenergized? Is there? Do you like to think? And again the human mind is complicated but do you like to think of it as like a bucket that gets refilled for introverts in terms of energy of um social interaction that that able to handle it is do you think of it like that as as a bucket that gets empty and needs to be refilled. [/INST] I think of it. Yeah more or less because I use the metaphor of a battery that gets recharged or not. It's basically the same thing. Different metaphor. But yeah, but just to add on that there is a layer of complexity to that because you could be somebody who doesn't want the kind of social life let's say where you have to be like on and presenting and interacting with tons of people all the time but you get really lonely if you were just by yourself, you know? So what you want is to maybe be in the company of a couple of people you know really well. Like for me the pandemic was not actually that hard for me personally. I mean I lost family but I mean from the point of view of what we're talking about it wasn't that hard because I live with my husband and my kids. So I knew it was hard on the kids and I felt badly for them. But for me I was like you know what, I kind of have a lot of social life right here in the house </s><s>[INST] and do [/INST] my work. Yeah. That's </s><s>[INST] the cool thing about the pandemic. I think it helped people figure out how much they love their family. [/INST] That's </s><s>[INST] true and and while you give you a chance to really reconnect with kids with your kids, like really spend time with them. It's just fascinating to watch. Like people actually it did strengthen the family unit and and often beautiful way, which just sucks to, to, to have to leave behind at this point. [/INST] Yeah. And I think that's part of what people are not gonna want to go back to that we need to solve for. Um to the extent that work becomes non remote. Again, I think people have just realized how precious those aspects of their lives are and you know, for somebody who's in a conventional office job where you're going home and seeing your kids for an hour before bedtime and that's your interaction with them. That's that's kind of a ridiculous way to set things up. </s><s>[INST] It's cool that you can get, I think a lot of places give you the option now, which is interesting. You get to optimize that element of your life. You take the commute and the office work and the social interaction there. Do you focus on the work at home, it's also lonely at home, but then you get to see your kids if you have kids, that's part of the optimization is like I have some options now and I'm going to try to optimize solitude loneliness, happiness, productivity, seeing family uh seeing coworkers. The chemistry with the team building with the coworkers versus just the raw exchange of information with coworkers is fascinating to see how that how that kind of evolves. [/INST] Yeah. And then there's the big the third space idea of, you know, the spaces where you are um in a coworking space or a cafe or something like that. You've got other people around you, but you're not exactly interacting with them, but they're very much there and that's huge too. I don't think we think about that enough. </s><s>[INST] Yeah. That energy is there. [/INST] Yeah, yeah. I I lived in Manhattan for 17 years before we had kids and I absolutely loved it. Like I loved it. The feeling of all that energy all around you, but you could be anonymous within it. It was perfect, </s><s>[INST] beautiful. Like I, I worked this morning for a few hours programmed for a few hours at a Starbucks. And first of all, like wearing wearing suits, it's like Manhattan is the one place you can kind of fit into that because everyone's wearing suits. Uh well these days, unfortunately, cause I get recognized, I I wear usually not suits when I just on my own life. But yeah, I love it. I love the way it feels. I don't know, and the way I think about the world when I wear a suit, I take it seriously as if my life is gonna end today, like this is what I would want to wear, not for, not physical appearance, but just for some reason, it makes me feel like focused, I don't know. [/INST] So even if you're not going to see anyone, you would still put the suit on when you're doing your work </s><s>[INST] especially, [/INST] especially, </s><s>[INST] Yeah, I really, I really love doing that, [/INST] like tells you seriousness of purpose. Something </s><s>[INST] like yeah, [/INST] like everything, </s><s>[INST] I don't know what it is. I don't know what I imagine exactly, but it's some kind of platonic form of like a mixture of James bond and like I don't know who else. Richard Feinman can I think about when I think about a suit, [/INST] you know, I think of Leonard Cohen, but he was always wearing suits. But you know, </s><s>[INST] Leonard Cohen is definitely one of my as a tragic human is a beautiful human being through his words through his own private life. Yes, I definitely would think about uh Leonard Cohen. So small talk. That's another thing. Um is that part of the equation of introvert versus [/INST] extrovert? How </s><s>[INST] much people enjoy? Small talk? [/INST] I kind of went into this whole thing thinking that it was, but from what I've seen most people of studies that most people don't like small talk, I think that's why people like your podcasts because you're </s><s>[INST] like forget the [/INST] small talk. I'm going deep into it from the very beginning. </s><s>[INST] Yeah, so it's actually the picture you're painting is like the way you started like with your with the book quiet and the way you are today is you realize the picture may be more complicated. [/INST] Yeah, everything's more complicated. I will say with the small talk thing that I'm curious if you have this experience, but I find it fantastic to have a career where I'm known for anti small talk kinds of topics because it means that anywhere I go like if I show up at a conference or something like that, no one does small talk with me, they're like telling me about the deep truth of their lives from the first Hello and I love that. And in normal life you have to like wade through a lot before, you know, if people are ready to go there, </s><s>[INST] do [/INST] you have that experience to </s><s>[INST] know definitely, definitely with people that know me for sure. [/INST] But you forgot how many people feel like they know you because of your podcast. </s><s>[INST] That's what know that counts of podcasts. And I feel it like before I ever became friends with joe Rogan, I felt like I was friends with him because I was a fan of his podcast. So like it was, I don't, I feel like it's a friendship. I know it's a one way friendship with all the people I listened to in podcasts and even people who are no longer with us like writers, I feel like I have a relationship with them. Maybe I've been saying [/INST] no, I totally feel that way. That's the whole reason I became a writer. </s><s>[INST] Like I'm friends with Leonard Cohen [/INST] Yeah, </s><s>[INST] and he's not aware of [/INST] it. No, I can't. But I think that's the whole reason for writing or making music or whatever people do, it's be able to have those kinds of connections that don't require having to be in a room together because there's only so many people you can be in a room </s><s>[INST] with in [/INST] your lifetime. </s><s>[INST] The hard thing is, is unfortunately [/INST] because </s><s>[INST] I value human connections so much and I only have just like you mentioned sort of a small circle of people I'm really close with by design. It always hurts me a lot to say goodbye to people. Like you meet people and you can tell they're beautiful people. They're amazing. There's something so fascinating about them there. They've had a complicated life. Like you could see in their eyes and the way they tell their story in just a few sentences. They've gone through some ship, but they also found some elements of beauty. And then you get to realize, okay, well there's a fascinating human here and all you get to say is a few words here and there. It was like a a funny little joke. Maybe a dark joke here and there and then you just say goodbye. Maybe hug it out and you go on your way. So like that's a hello and a goodbye and your paths will never cross again. That makes me like a sad walk walk away. But I guess I wouldn't have it any other way I suppose is the reality is in in your new book, you talk about that sorrow, that sadness not being such a bad thing. [/INST] Yeah. And when you just said that I just thought of this one moment in my life that I haven't thought of for 2030 years or something. But it was when I was in law school and and a classmate of mine had his friend come to visit for the weekend and the three of us hung out a lot and we just had, you know, like an amazing time and, and then this other guy who wasn't going to be coming back anytime soon if at all sent a postcard to me. And the only thing written on the postcard was this quote from Oscar Wilde and I don't remember the exact words, but it basically said that there's no pain as intense as the sorrow of parting from someone to whom you've just been introduced. There's something so intense about that and so true. </s><s>[INST] I think [/INST] partly also because when you've just been introduced to somebody, you don't yet know their difficulties. So you're seeing, you're seeing the most sparkling version of them, you're seeing like a platonic version of love and friendship </s><s>[INST] and your imagination fills in the rest and some some beautiful way that matches perfectly the kind of thing you're interested in. That's how I feel about one school, like one spoonful of ice cream and that's why you always finish the whole tub and you regret all of it. Um, do you do you do say, what did you write this? I think this is on your website that one of the best things in the world is that sublime moment when a writer, artist or musician manages to express something you've always felt, but never articulated or at least never quite so beautifully. So that's the Oscar wilde line is one line like that, but just a line from a song or maybe a piece of art that just grabs you. Is there something that jumps out into memory like that for you? I [/INST] don't know if I have an exact line though, I mean that feeling that you just quoted happens to me all the time, I'm just bad at recalling exact instances, but but the writer Alain de botton regularly makes me feel that way. He's just this beautiful essayist and observer of human nature and he's just constantly expressing things um in this gorgeous way </s><s>[INST] that [/INST] you've experienced yourself and you feel like it's just this grand act of generosity. You know, you feel less lonely, you feel like this deep sense of communion, it's such an elevating experience. </s><s>[INST] Even when it's like a melancholy line [/INST] maybe, especially when it is. </s><s>[INST] What is that? There's a so Jack, correct on the road, definitely makes me feel that way. Like every other line in there, forlorn rags of growing old, [/INST] you know, I never read that book. So what was it about that book that made you feel that way? </s><s>[INST] Well, okay, well, since you [/INST] asked, I'm going to linger, </s><s>[INST] I'm going to linger on this? Uh so this story is kind of the book? The kind of defining book of the, of the beats of the beat generation and it's basically a story of a writer who takes a road trip across the United States a couple of times and experiences a few close friends and a few strangers along the way. And there's a lot of just those melancholy goodbyes along the way. You meet all these people with interesting lives. Some of them are defined by struggles, some of them are defined by drugs drinking women, all that kind of stuff. And still, he just kind of dances around all of that and is defined by the goodbyes and the passing of time. So a lot of the really powerful lines are basically like, there's one on there again, I don't remember exactly, but he meets a beautiful girl at rest up and the girl is getting or or a woman is getting on a different bus than he's getting on and it's that like, feeling of falling in love for like a second and realizing that like, fate is just ripping that out, which is similar uh to this idea of it sucks to say goodbye just when you met, but it's especially true when you fall in love just a little bit with that stranger with all the possibilities that could lay there. Um so there's a few lines of written down, I went down this whole rabbit hole of thinking, what are the lines that grabbed me A couple of lines from on the road. So one is, what is that feeling when you're driving away from people and they received on the plane until you see the respects dispersing its the two huge world vaulting us and it's goodbye. But we lean forward to the next crazy venture beneath the skies. So this is him talking about leaving a particular city. The spoiler alert towards the end of the book rather The end of the book line never turned too often. It's more poetry, but it's a feeling that captures the book. I would say the evening star must be drooping and shedding her sparkler dims on the prairie, which is just before the coming of complete night that blesses the earth, darkens all the rivers, cups the peaks and folds the final shore in and nobody, nobody knows what's going to happen to anybody besides the forlorn rags of growing old. And it just captures this kind of in the moment appreciation of the beauty of the world and the sadness over the fact that time passes and you leave the people you love behind. You leave the places you love behind or at least the way they were at the time that you really enjoyed them. And you just leave that all that just the sadness you feel when you do something about it. Like look at her picture, looking at your kids grow up, looking at old friends getting old. something makes you realize that time passes and somewhere deep in there is probably a realization of your mortality and it just makes you somehow for sad that everything comes to an end and then that's immediately um followed by sort of an appreciation of the moment. Like a gratitude that you get to experience this moment. [/INST] Yeah, I know it exactly. I mean that's that's the whole reason that I wrote bittersweet. It's all about that. So I know intensely what you're talking about and by the way, my husband loves the book, a moveable feast by Ernest Hemingway, which I also haven't read. But it talks about that same thing, you know, groups of people traveling around together and the group coalesces into some magical formation and then one person leaves the group and it's never going to be the same again and then they move on to the next one. Yeah. I mean I think that's the deepest essence of human nature. Um the feeling of longing for some kind of state of perfect, complete completeness completion. Perfect Love the garden of </s><s>[INST] Eden all of it. [/INST] Um and the feeling that you're never going to quite attain it, but you get glimpses of it here and there and that those glimpses are some of the best things that ever happened to us and they're suffused with sadness because they're not the real thing or they're not the full thing. They're just a glimpse. It's a glimpse of what we long for. </s><s>[INST] So the sadness that we might feel is always connected to the ways in which we fall short from the perfect thing that we're like, there's always a thing you're longing for and the sadness has to do with the getting a glimpse of it, but not quite getting a hold of it. [/INST] Yeah. Yeah. </s><s>[INST] It was always losing, Its [/INST] always losing, but it's also always that. But it's not that sounds really depressing but it's it's not, you know, it's not depressing because you experience this all the time. It's also those are the most beautiful moments I think life offers. I mean, it's intense, intense beauty in those moments because it's getting closer to the real thing that we long for. </s><s>[INST] So what about like, loss. Losing love is that also Mhm. A beautiful thing. [/INST] Well, the moments you're talking about, I think it's easier to appreciate the beauty of it all in the moment because you're experiencing, you're kind of experiencing the loss and the love all at the same time. Um Whereas if you're talking about straight up loss, like a betrayal or a bereavement or whatever it is, that's it's different. It's quite overwhelming. </s><s>[INST] So losing loved one kind of thing, [/INST] Losing a loved one. I mean I will say that the truth that I think that we can come to after a lot of time on this earth is the idea that love exists. Not only in its particular forms. So not only in the form of the one person, you know that one person we love or that other person we love. But love itself is a state that we have access to. And so over time the loss of person a can heal and you can tap into a kind of bigger river of love. </s><s>[INST] Yeah, I mean I had this. This comes from Louie Louie C k [/INST] uh had a </s><s>[INST] show damn. I love that line. I mean there's uh he's talking to an older gentleman and louise is all like sad about um Losing a loved one. Or like getting rejected essentially like a breakup. And then the older gentleman give us some advice saying like basically criticizes louis for saying why are you moping around because this is the most this is the best part. Like losing love is the best part because that's the real losses when you forget like feeling shitty about having gone through a breakup is when you most intensely appreciate what that person meant to you. Like you most intensely feel love and some strange way by realizing that you've lost it by missing it, wishing at this moment. I wish I had that like that feeling that that's when you feel that love the most the absence of it. So the the older gentleman gives advice that that that's the best part and it can if you're good with it, it can last for the longest. [/INST] It </s><s>[INST] could be the most sort of, for a long experience of deep appreciation and emotion and so on. So that, that's kind of um that's a nice way to look at loss, which is a reminder of how much somebody meant to us. [/INST] Yeah, I mean, I think there's a lot of truth in that. Um, because yeah, you wouldn't care so much if it weren't something that mattered to you. So it's always a signpost to the direction you really want to go to. That's always what it is. Um, and it's interesting to see the way that the mystical versions of many of the great religions all point in this direction, whether you're looking at Sufism or, or the Kabbalah or in christian mysticism. You see this idea that the longing for what you lack is the very thing that gives you what you're longing for. So the longing is the cure. I mean that's the way the Sufi poet, Rumi puts it. The longing is the cure. Um, you know, and he says be thirsty like be as thirsty as you possibly can. That's what you want to be. </s><s>[INST] The good stuff is the wanting not to having [/INST] Yeah, </s><s>[INST] yeah, of course, tell that to a person that just broke up and they'll be like shut up asshole devices, sucks, I wish I had her him back. [/INST] Yeah, No, absolutely. Those are the kinds of life lessons that only work when you kind of step away for a while. They don't work in the moment, in the moment of excruciating. There is something about the fact of knowing that all humans are in that experience together. That is also incredibly uplifting. </s><s>[INST] Well, that takes time for people to realize like, um you know, like heartbreak in your early teenage years or something like that could feel like this is completely the most novel [/INST] in </s><s>[INST] the most dramatic uh pain that any human has ever felt, right, or maybe even when you're younger. And then one of the process, one of the things you realize is that everybody goes through this, that can be an awakening to the fact that we're all in this together. This human condition is not just a personal experience experience we all share. And that's uh that's the kind of love, the unity of [/INST] it. You </s><s>[INST] can get to experience [/INST] really deep kind of love. And I feel like, you know, we're we are prevented from perceiving that love as it's actually like the most obvious kind of love. And it's right there and it happens all the time, but were prevented from perceiving it because we're not really supposed to talk about things like that. It's like there's something unseemly about it. </s><s>[INST] Well, it's also in the in the West is this individualist society. So like there's a pressure to sort of see the individual as a distinct sovereign entity that um experiences things. And it's not the unity between people is not it's not obviously sort of communicated and talked about as part of the culture. [/INST] Yeah, it's not part of the culture and yet, you know, you see it in our behaviors because we're humans. So you know, why do people listen to sad music? I mean one reason is they're hearing expressed for them. Like the musician is basically saying to them this thing that you have experienced, I've experienced it too. So have lots of other people and but they're saying it all without words and it's transformed into something beautiful and there's something about that that's just incredibly elevating and people don't know it. But there's one study that I have in bittersweet, they found that people whose favorite songs are there happy songs play it on their playlist about 100 and 75 times with people who love sad music, play them about 800 times and they say that they feel connected to the sublime when when they're listening to that music. </s><s>[INST] What what do you think that is? So what is that, what what is it in music that connects us to the sublime through sadness? [/INST] I mean I have a bunch of different theories like the whole reason I started writing this book is because I kept having this reaction reliably to sad music and I realized that for people who I knew who were religious believers, the way they described their experience of God was what I was experiencing when I would hear that music, like all the time it happens over and over again. </s><s>[INST] So you you wonder what that is [/INST] and yeah, so I started wondering what that is. Um and lots of people have tried to figure out, you know what that's all about. Um and there are different theories that it's expressing, its like a kind of catharsis for our difficult emotions, that it's as we were saying, a sense of being in it together, we don't react in that sort of uplifted way when you just see um like a slideshow of sad faces, which is something researchers have actually tested. No one really cares when they're seeing the slide to the sad faces, but the sad music, they're really reacting and also they don't really react when they're hearing music expressing other negative emotions like martial music or something like that. It's just the sad music that gives people this elevated sense of wonder. So I think it's the combination of the sadness and the beauty and I think it's just tapping into the essence of the human source code, which is a kind of spiritual longing, whether we're atheists or believers, there's this feeling of longing for a state and a place of perfect love and perfect unity and perfect truth and all of it and like an acute awareness that we're not there in this world and, you know, in religions, we expressed that through the longing for Mecca or Eden or Zion and artistically we express it with Dorothy longing for somewhere over the rainbow or you know, harry potter enters the story at the precise moment that he's become an orphan. Um, so he's now going to spend the rest of his life longing for these parents who he can never remember. And that's, there's something about that state that's at our very core and I think that's why we love it so much. </s><s>[INST] Well, it could be, you know, you could have the Ernest Becker theory of denial of death where at the core of that the warm of the core, as young said, is the fear of death. So were the longing for the perfect thing. It has to do with sort of becoming immortal is um reaching beyond the absurdity, the cruelty of life that all things come to an end for, for no particularly good reason whatsoever. One we can rationally explain. [/INST] I know, you know, I wonder about that all the time. Like I know obviously there's that idea from Becker and throughout philosophy and the tale of Gilgamesh about the idea that the thing we're longing for most of all is immortality, </s><s>[INST] but [/INST] I feel like it's not only that, I think it's more so or also let's say, um, a longing for the lions to lay down with the lambs. Finally, you know, for like the, the fundamental calculus of the universe to be different where life doesn't have to eat life in order to survive. And yeah, just a completely different situation, immortality would not solve. </s><s>[INST] I wonder that could be a very kind of modern thing because surely so much of human history is defined by violence and glorified violence. That doesn't give inklings of this lions and the lambs so much. Yeah, I [/INST] mean, I know all the other stuff is in the bible, </s><s>[INST] there's other stuff in the bible and the bible and that particular aspect doesn't necessarily reveal the fundamental motivation of human nature. That could be deeper stuff, you know? But yeah, that is a beautiful, that is a beautiful picture. But is it just about humans or is it all about about all of life? And you have to think about what is the perfect world look like? It's not just the lines and lambs laying together is you know, how many lines and how many [/INST] lambs and </s><s>[INST] uh you know what, having just had a few very technical conversations about marxian economics versus kensi and economics versus neoclassical economics. [/INST] What does the </s><s>[INST] economic and the government system look like for the lines and the labs that were longing [/INST] for? So </s><s>[INST] that then you start to build society on top of all those things and but you still, you return to this, what is it? What are we longing for and what's the role of love and that what's the role of that sad melancholy feeling? The feeling of loneliness is the feeling of loneliness fundamental to the human condition, Like? Are we always striving sort of channel that feeling of loneliness um to connect with others? Like we want that feeling of loneliness, otherwise we wouldn't be connecting. Is that fundamental? That feeling like you're alone in this, even when you're with other people sort of alone together, you're born alone, you die alone. Maybe loneliness is fundamental. [/INST] I think the longing for Union is fundamental. It's just that it looks so different for different people, you </s><s>[INST] know? Yeah. [/INST] Coming back to where we were, what we were talking about at the beginning, you know, Union looks incredibly social for a lot of people and hardly social at all for others, but everybody needs some version of union. </s><s>[INST] Yeah. People have been telling me recently about polyamory and all those kinds of things. Some having probably grown up in a certain part of the world. I'm very, I think I'm very monogamy centric, not in a judgmental way. Just for me, what makes me happy is one person feel for my whole life, basically just dedication. Because I've just seen through relationships with people and objects in my life, the longer we stay together, the deeper the tie. So that's just the empirical thing. And yes, that probably it's a personalized thing that's just true for me, it could be very different for others. Maybe it's connected to the introverted thing? Maybe not Who knows before before, because you mentioned songs, sad songs? What um what are we talking about? What's a good what what what song do you remember last crying two? [/INST] Oh gosh, well, I mean that's you know, I I </s><s>[INST] literally [/INST] literally dedicated my book to Leonard Cohen, he's played such a huge role in my life, I love him, I love him. Um and I loved him with this crazy love that I've never been able to understand for decades. I think I understand it a little better now, </s><s>[INST] but you guys, so you're better friends with him than me? I'm so um does it make you, is it the the musician or the human too? Because the human is a is a tortured soul in a way, [/INST] I'd say it's the musician, it's the musician and I actually was thinking about this the other day. I mean obviously he's not alive anymore, but I was kind of running the thought experiment, you know, if he were alive still and I had the chance to meet him </s><s>[INST] in person, [/INST] would I want to do that? And I'm not really sure that I would because he represents for me symbolically everything. Well everything else in the sentence right there. And so and I think that's okay, you know, I think people can can express something through their art that they might or might not express if you were just like hanging out with them and having a coffee and I'm happy to know him that way. </s><s>[INST] He can express himself. I'm sure in the way that you know him as over coffee too, it just requires um like a focus of remembering of like a deep focus of connection, that's why like when I interact with folks Mhm. Um it's so draining for me because I'm putting all my whatever weapons I got in terms of like deeply trying to understand the person in front of me and doing that dance of human interaction, the humor, the intense kind of delving into who they are. Um So which requires like navigating around like small talk type of stuff and and just like compliments and so on in general, like people depending on the culture, depending on the place, they'll sometimes flower stuff was smiling and like compliments, like oh yeah, I love you, this is great, like this, that's all great, but you want to get to the core of like what are the demons in the closet, let's talk about it, you know? And that could be exhausting. That can be really exhausting. So from a Leonard cohen perspective you get more and more famous, it can be hard sometimes because he probably is also an introvert. [/INST] Oh yeah, I know he was an introvert because he actually tweeted about my book when it came out. Um So that was a precious moment for me. Something about we should all be listening to the quiet, I can't remember exactly what he said. Um But yeah, yeah, he definitely was, </s><s>[INST] he struggled with depression, which I wonder if that's something that's also connected to introversion? Um but perhaps not actually. Perhaps that they're very disjointed and also it's [/INST] connected to sensitivity and many sensitive people are introverts. So it's kind of like a Venn </s><s>[INST] diagram about [/INST] 80% of highly sensitive people are introverted but then some are extroverts and then not all introverts are sensitive so it's complicated but he was definitely a sensitive type. Well </s><s>[INST] there's on top of that you see like the percent of artists relative to the average. I suffer from depression. So creative people is very high. Crazy. [/INST] Yeah. And then the number of artists and successful artists who were orphaned when they were young who lost one parent or both parents like an astronomical number. I have it in the book and remember the percentage but huge and he was one of them. He lost his father When he was nine and his first act of poetry was was he took, his father made suits. That's why I thought of him when we were talking about you in your suit. And and he took one of his father's bow ties and wrote a poem in his honor and buried the poem and the bow tie in the backyard. And that was like his first creative act. </s><s>[INST] You know that song um Chelsea Hotel #2 [/INST] where he </s><s>[INST] met I guess it's about Janis [/INST] Joplin, Joplin </s><s>[INST] uh what a fun, intense and cruel person she is. [/INST] Yeah. </s><s>[INST] Uh I guess um have [/INST] you ever seen, I'm sorry to interrupt you but have you ever seen his son Adam Conan and Lana Del Rey perform that song together? </s><s>[INST] Oh no, it's [/INST] incredible. I have to send it to you. </s><s>[INST] Yeah. So that for people who don't know, I don't, I mean maybe, I don't know it goes. I remember you well in the Chelsea hotel you're talking so brave and so sweet giving me head on the unmade bed while the limousines wait in the street. There's a good line in there about um about being ugly. [/INST] Oh yeah, we are ugly. But we have the music. </s><s>[INST] No, before that from a guy's perspective it was [/INST] oh you told me again you preferred handsome men but for me you would make an exception. Yeah. So good. Well she </s><s>[INST] continued that threat in later because I think she said that he was lousy in bed. [/INST] Is that right? </s><s>[INST] Yeah. She publicly said that. Which is like man, they're just okay for people that don't know. I think this is a true story about them interacting and and being together for a very brief time. I don't know dating but just connecting falling in love [/INST] or </s><s>[INST] in this only in this very particular way that I think famous musicians poets can can which is like it's impossible for that kind of thing to last. But they did for a brief um moment is like a sadness to it because it's so momentarily but it's so [/INST] epic </s><s>[INST] that these two paths cross and then you just look at that we know these famous people and it's it's interesting to watch [/INST] and you don't even have the impression that they're thinking it's going to last. They more know that it's like a blaze of an intersection and the you know, the limousine is already waiting while they're in the moment in the middle of it and then it's done. Um Yeah </s><s>[INST] but he's [/INST] talked about how his music, he said he said something like some people are more inclined to say hello with their music but I'm rather more valedictory. That's what he said </s><s>[INST] like [/INST] saying goodbye. Like the Victorians address </s><s>[INST] interesting, [/INST] you know, so many of the songs really are about some form of partying or goodbye or an imperfection or something or like the broken hallelujah. Um </s><s>[INST] But the [/INST] thing that's so incredible about him is is the way that he's taking all of that and pointing it in the direction of transcendence. Like it's not it's not just pure sadness, it's sadness and beauty and that's the thing. </s><s>[INST] Yeah, there is a feeling of transcendence and a lot of the songs, it's like sadness and transcendence. You're right, it's a goodbye but you're moving on to some bigger thing but in a in a sort of ethereal way. Like not like a proud arrogant way. [/INST] Yeah. So his favorite poet was Garcia Lorca, he actually named his daughter after him, his daughter's name is Lorca. And </s><s>[INST] and [/INST] he talks about how there's some poem that Lorcan had written that made him realize that the universe itself was aching. But the ache was okay because that's the way you embrace the sun and the </s><s>[INST] moon. [/INST] And that's what I think is. That's why I think there's this whole Witteveen in </s><s>[INST] in this [/INST] bittersweet tradition that he embodies. That's like the essence of beauty. You know, it's the way you embrace the sun in the moon, </s><s>[INST] the song Hallelujah. I returned to that often have been me to play it. Um have now a friend who wants to sing it with me. Are you [/INST] a singer? </s><s>[INST] Mm [/INST] What </s><s>[INST] what when when somebody says they're a singer, do they have to be [/INST] good because </s><s>[INST] then no, but I would say yes, I was in a band for a while. I sang for a while. I was always bad. But I I enjoy it. I enjoy it. I enjoy lyrics. I enjoy awards when sung or spoken. They capture something that could get in that moment. Tom waits is a huge favorite of mine for that reason. Um Although he often his lyrics are often not that not that simple. I'd rather have a bottle in front of me than a frontal Lobotomy. He's always playing with just like these [/INST] weird </s><s>[INST] word play. That's especially in the english language is trickier to do. I'm fortunate enough to know another language which is Russian. So I get to understand that certain languages allow for more world wordplay than others english for that reason, I don't think has a, like, like a, like a culture of, you know what, I need to push back on what I'm about to say. But there was no culture of wordplay until hip hop came along. [/INST] So </s><s>[INST] like, like distorting words in interesting ways for there to be a rhythm rhyme and at the same time you're capturing some really powerful message. Um plus humor, all of that mixed in actually hip hop does does a really good job of this. But there wasn't a tradition, if you look at poetry in the 20th century, there wasn't really a tradition of that in the United States, but there was in other parts of the world and certainly in Russia empowered, also, not just by language by the fact that you go through a world war where tens of millions of people die. Something about mass mass death of civilians that inspires great literature and music and art. [/INST] Yeah, absolutely. Because you start telling the real truth. I think </s><s>[INST] there's no more reason for small talk. Um [/INST] that's funny. I always have thought that if I could choose any other medium besides writing it would be singing. Um but </s><s>[INST] then [/INST] no, I mean, like I'm really not, I just love the idea of it. But then I also think, you know, I'm fundamentally a shy person. So I think it's much better that my medium is writing instead of singing. So like it all worked out </s><s>[INST] that said you're also an exceptionally good public speaker and you're not supposed to be mathematically speaking [/INST] mathematically speaking, </s><s>[INST] you're not supposed to be a good public speaker. [/INST] Oh, you mean because of shyness or </s><s>[INST] because of shyness? Because of introversion? Because of all those kinds of things. [/INST] Oh yeah, but lots of introverts or public speakers actually, like this is one of I I knew this from the studies, but then also when I started going out on the lecture circuit, I realized that all my fellow speakers and all these conferences I was going to, they're all introverts because, you know, there are people who like spent years figuring out some idea and now they're out there talking about it </s><s>[INST] in their head, figuring out the idea. How do you explain? How do you explain that the the public speakers, would you say the good public speakers are usually introverts? [/INST] No, I think there's just different styles of it and I think that we just have when we hear the word public speaker, we have a really limited idea of who that person would be. Um so for me, like I used to be very phobic about public speaking and part of the reason for it was because I thought that being the kind of person I was didn't equal being able to be a good public speaker because you're you're only imagining, you know, like that super kind of out </s><s>[INST] there showman. [/INST] Um but I think there's another style of public speaking that's more reflective and thoughtful and conveying ideas and people like that too, </s><s>[INST] is there advice you can give on how to overcome that? Like, if you're a shy person, how to be a public speaker, [/INST] totally give that advice because I used to before I would give speeches, you know, if I had to do it in law school, if I knew, like today was the day when I was going to get called on in a law school class, I literally one time vomited on my way to class. That's how nervous I used to be. Um and yeah, the way to do it is through desensitization, you know, it's like been figured out, it's the way to overcome any fear. You have to expose yourself to the thing you fear, but in very small doses, so you can't start by giving the ted talk, you have to start, I I started by going to this class for people with public speaking anxiety, where the first day all we had to do was stand up and say our name </s><s>[INST] and sit [/INST] down and like, that's the victory. </s><s>[INST] That's fun to watch all those people with anxiety. Just okay, that's the first step in the step, one step at a time. [/INST] Yeah, and then, like with this class, you go back the next week and and he would have us come to the front of the room and stand up with other people standing next to us so that you didn't have the feeling of being all alone in the spotlight, through others, sharing it with you. And you would answer some questions about where do you grow up? You know, where do you go to school and you declare victory and you're done? And then little by little by little, you keep ratcheting up the exercises until you get to the point </s><s>[INST] where you can do it [/INST] and then you start having successes and you realize, oh, actually I can do this. </s><s>[INST] What about like writing versus improvising? Because I knew a few people sort of the colleagues of mine that were working on Ted talks and she was like, you're supposed to like write the thing like way ahead of time and you practice it and they help you and all that kind of stuff. I don't think I've ever practiced the speech once in my life or a lecture or any of that. Like I know it's really good to do. But do you find that relief some of your anxiety preparing well or are you not able to do not preparing well at all? [/INST] Um I definitely like to prepare before, but the kind of preparation that I've done for my Ted talks is completely different from what I've done for everything else because Ted talks are more like a theatrical event where it's like a one person show, and of course, if you were gonna go on broadway with a monologue, you would know every word, So it's kind of like that, </s><s>[INST] um and [/INST] so I would rehearse it over and over the way you would do that. </s><s>[INST] Isn't that more anxiety knowing every single word. It's [/INST] so much anxiety because yeah, you're you're not even so freaked out about being on stage so much as what if I forget something. Yeah, absolutely, I mean, they do things like the last Ted talk I gave actually did forget something halfway through, like, I just couldn't remember the next line um and so I had to walk over, like over there were my notes and so I did that and the audience like, very kindly clapped, I did that, and then I came back to the spotlight and kept going and they edit that out. </s><s>[INST] So there's a family, it seems really, it seems really stressful, like, I'm now, I'm not sure if I'll ever publish it, but I've been mostly is for personal journey, but I've been working on a series on wait for it, Hitler and the third Reich sort of looking at the historical context of everything because of my family was so much affected by that whole part of history. So, for me to rigorously, I've read a lot about Stalin and Hitler, and for me to force myself one of the best ways to force yourself to really consider material is to have to talk about [/INST] it. </s><s>[INST] And so that's that's why I'm doing it. But I'm playing with ideas of Some of it, maybe like 20 is written down on paper. But the rest of it is my thoughts in the moment. That's a difficult balance to strike because if you write a lot, you're going to be more precise, you're going to be more accurate, but you're going to miss some of the deep, like honest emotion. The silences won't be correct, or the silences between the words won't capture the depth of feeling unless if you're somebody like me, if you're like, I guess that's what actors and actresses have to do. Like, basically, even though the script is fully written, you improvise between the words between the lines. [/INST] Yeah, </s><s>[INST] that's a skill. [/INST] It also takes so much time. I mean, I experienced that with the ted talks, it's like you get to a stage, so you're memorizing everything word for word. And at first in that process, it comes out in a really wouldn't weigh the way you're saying, like, the emotion's gone, but once you really know it. So you've internalized the words, then all the emotion comes back and you can say them in a completely different way. You know, when you're really speaking from the heart, but you have to know it so well before you can do that, </s><s>[INST] I would never [/INST] recommend it because it's just like, it's it's so time consuming, It's </s><s>[INST] an inch well it in your case. It works out beautifully when it, when it all comes together. It's, it is a theatrical thing. It's like a musical or whatever. Um I'm gonna, I think I'm going to come out with a one man show on broadway um singing No, I'm inspired [/INST] for real. Where are you going to, where are you going to talk about Hitler and Stalin and everything you're learning </s><s>[INST] to? [/INST] Have you ever thought of using the medium of just speaking into a microphone? But without the video, I'm curious about this because like, I fell in love with podcasts originally before there was ever this whole video component to it and I realized there's something so primal and magical about having someone's voice in your ear. Um my favorite kinds of interviews. Still very few people do it this way nowadays, but my favorite kind or when you're just talking into the microphone, so it's not over zoom. </s><s>[INST] It's not [/INST] in person, it's just you and the microphone and the other person in the microphone and they're in your ear, it's like the ultimate and intimacy. </s><s>[INST] Oh, you mean from the interviewer perspective, that's yeah, [/INST] but but it would be interesting also is with the kind of thing you're talking about just speaking. Like just you and the mic. </s><s>[INST] I would love to be in person, but you can't see the person. I wonder what that's like, [/INST] what do you mean? Like they're all there but behind a curtain, </s><s>[INST] just have your eyes closed, [/INST] just </s><s>[INST] talk and you have your eyes closed or whatever you have, because I think you still have to get the same kind of chemistry because it's not just the visuals. I don't even know that because obviously have trouble making eye contact. Um but I don't know if the visual stimulation is the necessary thing. There's something about the way audio travels that captures the intimacy where some people actually have headphones on. Like joe does this have headphones on? That's really intimate? Like there's something about that sound going directly into your ear. Yeah, there is something primal there. Yeah, for sure. I've I've thought about it definitely, some of my favorite podcasts are like that uh WTF with Marc Maron, that's audio only. There's there's a few audio only podcasts that I just love and what is that? I still go on clubhouse that that was, it's a social media platform where it's audio only. And so interesting that people, the interesting thing about clubhouse in particular is people from all walks of life can tune in and they just have, it's uh somebody needs to do some research in terms of introversion on that one because I don't feel any of my introvert um like triggers happening because so nobody can see is just the audio and nobody is offended if you're just sitting there quietly, just listening so you can, you can participate whenever you want or not. [/INST] Yeah, it's like the ultimate social freedom. You can listen as much as you'd like, you can participate if you want, but you don't have to, it's no big deal. </s><s>[INST] Nobody can. Like if I'm actually a physical party, somebody's gonna look at me and be like, why? You know, there'll be that pressure to speak. But you don't have to in the, in that kind of audio setting and there's that intimacy like you can when it's audio only, it feels like you can reveal a lot more of yourself in some kind of honest way. I don't know what that is, what is that, [/INST] but I don't know. But I assume it's tapping into something really ancient. Like we used to tell stories around the fire, like our whole storytelling tradition was oral originally. </s><s>[INST] So maybe it's that. But we used visual stuff like that. [/INST] You can actually see the person. It seems like it seems like </s><s>[INST] the visual elements so fundamental to social interaction, but there's something primal about audio. I wonder what that is. And still, that's why, I mean, most people listen to podcasts, I think audio only they have it in their ears while they're doing stuff. [/INST] That's how I do it. </s><s>[INST] And then there's that's how I do it and that's what, that's where the friendship, like it's formed. It's weird. The deep connection with other humans. It's formed because they're in your ear. Uh and you get to see them grow, you get to see them be bored, experience excitement and anger and fear and all those kinds of things. It's fascinating. It's fascinating. The world of podcasting is fascinating because we're in this world of a century radio even though we all of all this high definition content, all this like Tiktok style fast stuff and still [/INST] I know and we still choose to do this at the end of the day. I think that's really what people want most is just to talk to each other and to know what people really think and podcasting of all the media that I've ever seen is the one where people come closest to telling you the truth and you know, telling you like the good and the bad and the bitter and the sweet and all of it, </s><s>[INST] especially long form, there's not enough time. [/INST] Yeah, exactly. </s><s>[INST] I had to explain this to people like you talk to Ceos and stuff they don't understand and their understanding, they're starting to understand much better now as a hard requirement with like C E O. S and stuff. It has to be three hours. I say like [/INST] wow </s><s>[INST] because there's something they can't be doing marketing stuff for three hours. [/INST] They break, they </s><s>[INST] start being human, they [/INST] start joking and </s><s>[INST] relaxing and if they can't that also tells the kind of story. But I do that kind of torture for Ceos only anyway [/INST] when I was there getting my publishing house, did media training with me before bittersweet came out and and they were preparing me for like the 5 to 7 minute interview that you might have, you know, if you go on some quick tv thing or something like that and God, I hate that. It's like it feels like there you're basically having to not tell the full truth somehow because you can't you can't tell it in such a short amount of time </s><s>[INST] the other. [/INST] So to me, podcasting is just the best thing that's ever happened. </s><s>[INST] The other downside of the seven minute interview is I think you can do a really good job with that, but the dance partner has to be very good. It's actually challenging for everybody involved, it's much harder for everybody involved because if you can do you know, I can imagine like a Christopher, Hitchens type character who is just super witty then, that you could do a seven minute thing you can get to the core bittersweet, you can get to the core of the book without asking those generic small talk questions because too many people in that short form interview are just asking very generic questions they're doing small talk for. So [/INST] it's </s><s>[INST] like alright, like you only get seven minutes, you only get one interesting question, go ask the weirdest, the deepest question that that also energizes the other person. It's an art form um that people don't take seriously, I think the seven minute thing, five minutes or even less. And then the commercials, which I [/INST] yeah, and I've noticed that many </s><s>[INST] of the [/INST] best podcasters are ones where when you're on my side of the table, you feel like it's more of a conversation and less like an interview where you're answering all the same questions you've asked, you've answered a million times before. It's really interesting how different the experiences </s><s>[INST] and you're right, the uh the audio thing um if you can lose yourself in that, the intimacy of that and you don't even remember what stupid stuff you [/INST] said </s><s>[INST] people, I've seen that. I mean people don't give them enough credit as um you might, you might not be aware might not be a fan, but joe Rogan is a incredible conversationalist in that he makes you forget that anything is being recorded, that you're talking at all. It makes you forget time and you just enjoy yourself and that's whatever that is. Um and that plugs then you plug into that primal connection to other humans. What's your favorite Leonard Cohen song? [/INST] Famous Blue raincoat? Do you know that </s><s>[INST] one? [/INST] Yeah. </s><s>[INST] For people [/INST] who don't know Leonard Cohen, and this is your first introduction to him. It's gonna sound so gloomy, but it's so good. </s><s>[INST] He's got this deep, rich voice Tory most covering famous Blue in Korea. No, we want what the original, just like Hallelujah, Jeff buckley covered Leonard Cohen. That was a really good one. [/INST] That was a really good one. Yeah. And I also really like Rufus Wainwright's cover um but famous blue raincoat for people who don't know it it's basically about a love triangle and it's told from the perspective of a man whose wife has just been with another guy who is also his friend and he's writing a letter to that other guy and and he's reflecting on the way that all the relationships have changed in the wake of this this event. So they're still well, he he refers to him as my brother, my killer, which is such a Leonard cohen thing to do because it's always like, you know, it's light and it's dark all at once. Nothing is ever all one thing. Yeah, </s><s>[INST] I love this song. [/INST] Yeah, right. I mean, he </s><s>[INST] just speaks in the end of december, [/INST] The fact that it's 4:00 AM and it's the end of December. Those are transitional moments. It's night going in today </s><s>[INST] and [/INST] it's december going into the new Year. </s><s>[INST] It's [/INST] not an </s><s>[INST] accident. There's something about december. This, whatever there's certain scenes you can paint in your mind. Um there's a there's a poem by Charles Bukowski called Nirvana. It's a young man traveling through the middle of nowhere in the snow. There's something about the snow and the rain or the [/INST] snow can </s><s>[INST] put you in a certain kind of mood. That just uh what is it? James, Joyce's the dead, The snow is falling and Dublin. Yeah, I [/INST] mean, David Yaden who's a he's a researcher and psychedelics and consciousness at johns Hopkins. He's a great guy and he's done research that has found that when people are in their transitional moments of life, you know, and it could be a career change, it could be a divorce, it could be that they're nearing the end of their life, that they very often will say those are their most meaningful moments and their most spiritual moments. And so I feel like that's what Leonard Cohen knows how to tap into instinctively. The year after he died, his son Adam Cohen made a memorial concert for him where all these famous musicians came to Montreal where they had lived um and performed his music. And my husband who's not a Leonard Cohen fan and he's not a bittersweet type at all, but he knows how I feel about him. He's like, you know, you should really go to that concert. I felt so ridiculous. The whole family went all the way to Montreal on a monday monday. It was just like a random monday. And we got on the plane, like, so like everyone's out of school, just so I can go to this concert and and I got there and at the beginning I was feeling like this was all a terrible mistake because it's all these other musicians playing this music and I don't actually really want to hear them. Like I'd rather listen to him on Youtube and then and then a musician named Damien Rice came and played famous blue raincoat and he's saying it and he, he did the most amazing thing at the end, the whole thing was amazing. But then at the end he sang this musical riff that was like all I could say is it was like a musical lamentation of the ages and the whole audience just rose silently to its feet and it was one of the greatest moments that I've ever had. There is sometimes </s><s>[INST] certain artists in a cover can capture in some kind of deeper way like carrying the thread of the power of the song. So been listening a lot to johnny cash Heart which is a nine inch nails [/INST] you talked about on your podcast with rick Rubin </s><s>[INST] which is when I reached [/INST] out to you. Like I love that interview and I love that song also. </s><s>[INST] Yeah, yeah, yeah. So there's, there's that, there's the Kennedy center honors where they, they celebrated certain artists that did that for led Zeppelin and I forgot what her name is but the lead singer of heart um perform stairway to [/INST] heaven. And </s><s>[INST] it's like if you like, all right, you take one of the great sort of rock songs of all time, what do you do in front of oh the cool thing is um you get to perform this in front of the artists while they're still there. You know, they're still alive. So you get to watch, you sort of perform in that case the president, President Obama is there and she just knocked it out of the park. But at the same time without um out doing the original somehow, [/INST] you're just making it your own, </s><s>[INST] you're making your [/INST] own. It's not, but </s><s>[INST] not department completely not departing from the spirit of the original. It's tough because the original Hallelujah by Leonard Cohen, it's just not, it's it's so powerful, but it's just not as good as some of these covers. [/INST] Well, I think, </s><s>[INST] I think [/INST] it's the words and the melody and then the covers take it to a different place. </s><s>[INST] The thing that Leonard Cohen seems to do well, he, I don't think he did it on on Hallelujah because he was almost being playful on Hallelujah, like, I don't know, uh as opposed to that deep melancholy, like painful longing thing that Jeff buckley did and others do too. [/INST] I wonder if it's because in a way he, I mean that he over edited it, but he apparently worked on that song for years and went through gazillions of verses and checked most of them out. So I wonder if we're hearing his version after, He's like a little tired with that process. </s><s>[INST] Yeah, well that's the other thing is like maybe from a book tour, you know, it's like you get tired of saying the same thing over and over and over and over, You forget you forget the, you [/INST] forget like the initial like the heart of it. </s><s>[INST] Yeah, but I actually, I got a chance to hang out with dan Reynolds, who's the lead singer of Imagine Dragons and this is um incredible band, Super Pop, the most played band on like Spotify or [/INST] something. Is that right </s><s>[INST] through [/INST] a huge Imagine Dragons Face. So we were listening to their music a lot. It </s><s>[INST] was so surreal to be hanging out with him and he's such a, it's such a good, like very few people I met in my life are just as good of a human being and that has to do with the fact that he struggled, he struggles, he still, I think struggles, but he struggled for a long time with depression and so out of that pain. You see born this really good human being. This really good relationship with his wife like that. Like when times are good, they leah and each other for like they they're deeply grateful for those precious moments. So it's beautiful to watch. But he said that it's really important to feel the song every time. Otherwise people know people are really good at detecting your bullshit. You can't fake [/INST] it. You </s><s>[INST] really have to feel it every time. You have to feel the emotion of it. Whatever the emotion is of the original time you wrote it. Um yeah, which is interesting because it put, I thought you could maybe fake it, but he's he believes person because he played in front of the gigantic crowds and over and over and over and over and over. He's like, no, every time you have to have to be there, but [/INST] there's got to be times when he's about to go out and he's not feeling it and he has to figure out some way of getting himself into that heart space. </s><s>[INST] That's what he's saying. You have to, otherwise you're just that's the job, [/INST] don't take the </s><s>[INST] job then. Uh and he loves it. He says the biggest struggle, in fact is the come down from that, which is like you have such a beautiful experience of connecting with this large number of people sharing a song that you love. And then it's just a rush of connection and then you have to, you know, when you get off stage, you're now back to normal life and that's why a lot of musicians get into heavy drugs and all that kind of stuff because you're you're looking for that rush again, it's very tough to like go into this Speaking of introvert because he probably is an introvert as is like you have to find that calmness and how do you find the calmness when you were just playing in front of tens of thousands of people or 100,000, whatever that number is. That rush of connecting everybody, there's love in the air and you still have to find that like inner peace and calm. [/INST] That's interesting because so I don't know if this is the introvert in me talking and the writer in me talking but I don't know like I I love most the moments where let's say, I'll get a letter from a reader who will tell me what something I wrote meant to them and they'll talk about having had that kind of moment of, you know, the communion between the writer and the reader and obviously I wasn't there physically when it happened. So I wasn't getting that kind of rush that a musician would get in a concert, </s><s>[INST] but just [/INST] the knowledge of that having happened out there in the world due to something that I added to it. Yeah, </s><s>[INST] but imagine imagine reading like thousands of those letters and then it's such a strong rush and everything else doesn't it could be overwhelming, I guess. But like anything else you have to come down and find a complex, like for example, the danger getting letters like that, you start taking yourself too seriously. You think like, you are a special person somehow, but that's you really want to avoid that feeling too. [/INST] Yeah, I don't I don't actually experiencing as that much different from when I'm on the other side of it. Like, if I'm the reader and some other writer has made me feel that way to me. It's the same thing. </s><s>[INST] Yeah, me too. Yeah. It's it's a cool it's a it's a virtual hug. [/INST] I think it's like I was just listening to something about </s><s>[INST] the [/INST] different Russian writers. I was mentioning him to you this academic, his name is Gary saul Morson and he studies Russian literature and he was talking about, I don't know if I'll be able to get this right, but basically that people misunderstand a work like anna Karenina, him </s><s>[INST] and that [/INST] we think of it as telling us that you're supposed to live, you know, you're supposed to have like these grand, tempestuous romances that might end in death or despair or whatever it is, but you know, you should be in it for the intensity of the emotion. And he's saying actually that's not that's exactly not what Tolstoy was saying, that actually it was the opposite that he was really making, he was really advocating for everyday life. He was saying, it seems from every day, like he was juxtaposing anna Karenina with all these other couples who are just living happily and quietly day by day. And that was what he believed was the ideal. So as as opposed to the grand rush and as opposed to the intensity, </s><s>[INST] is there a uh, a depth is there a romance? It's just the day to day. [/INST] Don't get don't </s><s>[INST] get distracted by the dopamine roller coaster ride of the Grant romantic notions [/INST] and enjoy it while it's happening because those are life experiences also, but not to mistake those for being everything. </s><s>[INST] Where is he from? [/INST] He's a professor at Northwestern </s><s>[INST] Northwestern [/INST] and apparently his lectures are like the most popular on campus </s><s>[INST] people love him. Gary saul Morrison is an american literary critic and Slob Ist. He's particularly in over his scholarly work on the great Russian novel lists Leo Tolstoy and Fyodor dostoevsky Morrison is Lawrence the professor in the arts and Humanities at Northwestern University. Wow. And there's a lot of incredible work and then I'm sure looking through the lens of Russian literature and the romance of all of that, he's looking at modern at the modern world. [/INST] Yeah, I think you should have him on your podcast </s><s>[INST] and quiet flows the vodka or when Pushkin comes to shove the curmudgeons guide to Russian literature and culture. This is one of the silly books he has on the list. Okay, cool. Oh [/INST] no, I was just saying, yeah, like I find that when I take photos on my phone, I hardly ever take photos at the moment, you're supposed to like everybody's gathered for some event. I'll forget to take the photo, but I take a lot of like scenes from everyday life because that's what I actually want to remember in the end. </s><s>[INST] Yeah, yeah, I'm the same the same. It's actually concerning because it's bad for productivity because I love everyday life so much. Then why why do any ambitious big thing. [/INST] Your productivity is pretty good. I don't know that you have to worry about </s><s>[INST] it. I do. So I want to launch a business, You know, I have, I have a dream outside, this is like a fun side thing that, you know, I wanna, there's been a lifelong passion anyway, that's a, I like building, I like building stuff and I haven't been doing that as much as I would like. That's because largely because I like sitting in silence and enjoying the beauty that is just nature and life and when there's people, there's people, I love people, I love everything. And so when you love everything, why go through hell to build a company? [/INST] Yeah, it's a valid question. I mean, I think you have to have a really good reason for wanting to do </s><s>[INST] it, but then your heart calls you for the certain, sometimes you look out into the mountains and you say for some reason I long to go there even if it means leaving the tribe and putting yourself in danger and doing stupid shit, that's a human, um, imperative for exploration. [/INST] Yeah, absolutely. Like when we were talking about, um, you know, this idea of longing being like the source code of humanity, I think that's also the source code of our creativity. It's a, it's the same logging for Eden. It's like you're always reaching for something that you want to get to you or that you want to build. It's the best of us. </s><s>[INST] What do you think? Uh, you've read about creativity and sadness, You practically speaking, how should we leverage sadness for creativity is that sort of in the artist domain, in the writer's domain, in the engineering domains and so on. [/INST] It's definitely in those domains, but it's in all domains. Um, you know, we're all going to face pain in this life at some point and we all have the ability to weather it and withstand, withstand it and live with it for a bit and then try to transform it into something that we find beautiful. Um, you know, and it's very easy to, to notice the grandeur of like the, you know, the painting hanging on the gallery wall or the new company that's just been created. But it takes 1000 different forms, right? You know, you could bake a cake or like </s><s>[INST] in [/INST] the wake of the pandemic, we've had more people applying to medical school and nursing school And after 9-11, you had people applying for jobs as firefighters and teachers. So there's something in the human spirit that takes pain and turns it into meaning when we're at our best and when we're not at our best, we deny the pain and then take it out on ourselves and on other people. So there's a kind of fork in the road what to do with it. But we know, I mean there's all these studies that I go through in the book. Like there was one, um, where the researchers had people watch different movies, like happy movies, sad movies, bittersweet movies and they found when people watched Father of the Bride, which is like the ultimate bittersweet. You know, you're walking your daughter down the aisle kind of feeling that was they would give them creativity tasks after watching these different movies. And the people who had been primed for bittersweetness were the most </s><s>[INST] creative, They [/INST] were like prime to remember Finality. You know, like love and finality basically love and impermanence. There's something about that that's that gets us to our most beautiful state. </s><s>[INST] I wonder if it is. I mean there's studies like that there's a, I don't know if you looked into terror management theory. [/INST] Yeah, that's really interesting stuff. </s><s>[INST] So they especially intensely have you focused on not just sad but traumatic like death primary with death and see how that changes your mind. Like both. Like I don't, I don't know if there's creativity studies but they have um interesting I think a little bit tainted by political bias but maybe not. I mean psychology is a complicated field but they study like who are you likely to vote [/INST] for if </s><s>[INST] you're primed by existential like by thinking about the [/INST] fear of mortality, </s><s>[INST] mortality. I forget what the conclusions are, but [/INST] I I think they find that people become more try ballistic, you know like there's one study where they found that after they primed people that way that um, that they would then give them the chance to put hot sauce on a meal that their political opponents were going to be eating and they put way too much hot sauce on after they've been primed to worry about death. </s><s>[INST] I think at the core we're simple creatures. [/INST] So I actually, like in the book, I I spent a bunch of time with people who are working on radical life extension, you know, or the quest to live forever. And people ask them a lot questions like, you know, the kinds of questions you were talking about earlier. Well like how are you gonna feed everybody and how is there gonna be space for everybody? Um if everyone really could live forever? And what about conflict when we have an intensified conflict? And their answer to that is they point to terror management theory. And they say because it's the fear of death, they're basically saying it's the fear of death that are causing our conflicts in the first place. And that if we remove the fear of death, we'd have less conflict to contend with and that I don't really buy that </s><s>[INST] it's possible that that's true. But are you also, how does the expression go throwing out the baby with the bathwater? Are you also going to remove basically any source of meaning and um happiness in the human condition? Like it's very possible that death is fundamental to the human condition? The finality. Yeah. [/INST] And that's the great philosophical question and I went to a conference of of people who are working on this, and I thought that they were going to be talking about those questions all through the conference, but the M. O. Is much more like we're so happy that we're here with people who have gotten past all those quibbles. Um, you know, we just know there's going to be meaning no matter what </s><s>[INST] the basic assumption is, let's try to extend life indefinitely. And then we'll figure out if that's a good decision, [/INST] or more like, we're sure it's a good decision. That was what I felt. </s><s>[INST] It's sure it's either we're sure it's a good decision or we're sure that is good to believe that it's a good decision, meaning, like there's no downside to to that, even if we find out it's wrong. But yes, there's a kind of certainty, obviously you want to extend human life. That's the kind of assumption that always seemed. Um now it could be true, but just like the people who over focus on colonizing other planets, it feels like you neglect the beauty and the struggle of our life here on earth. I have sort of the same kind of criticism whether it's thinking about valhalla or any other afterlife, as you can have if you're not careful Forget to to make this life A Great one whatever happens afterwards. So, yeah, definitely. But from an engineering, from a biology, from a chemistry perspective, it's very interesting to think how do we extend this thing because it does seem that nature the way it designed living organisms, it really wants us to die because that's part of the selection mechanism. This part this seems to be fundamental to evolution. It uh, gets because people are young, they need protection once they, once they're young brain, they get to explore a lot. Get to figure out the world to come up with their own novel ideas how to adapt, how to respond to that world. And then you get older and older, they get like stubborn and stuck in their ways. And so we need them to die. So we make room for new life that's able to adapt to the changing environment. If the old doesn't die, then you're not, you're going to get stale and not be adaptable to the changing environment, [/INST] but maybe it doesn't have to happen so soon. </s><s>[INST] Yeah, maybe it doesn't like pressing listen, I'm a big fan of pressing snooze on the alarm clock. I'm in the same way. Um, I do. I'm one of the people that believe it's or I don't definitely believe. Of course, I don't know. But I think death is a is a fundamental part of life. But yeah, I found on my deathbed, I would sure as hell press news as many times as possible. [/INST] Yeah, it's interesting because in some ways I really I share your instinct. Um, there was one scientist who I spoke to at that conference, He's one of the leading advocates and he said, you know, that's a story that we've invented for ourselves because we have no choice. And if you really believe that you have no choice, then you're then it's it's adaptive to tell that story that death gives meaning to life. But if you really think you don't, if you really think you could triumph over it, would you still be telling that same story? </s><s>[INST] And I've been [/INST] thinking about that question ever since. </s><s>[INST] Yeah. Yeah. No, [/INST] they got a good point. </s><s>[INST] They got a good point, no matter what, as an engineering and the scientific pursuits. It's a beautiful one in your own personal life. If you can go there, what have been some dark places you've gone in your own mind, grief, loss, sad moments, moments of sadness that have made you a better writer, better creator, a better human being. Well, [/INST] I mean, I've been through a lot of bereavement just in these last couple of years with Covid. But even before that, I mean, there's there's all kinds of stuff I write about in the book. And in some ways I feel like I can write about those kinds of things better than I can speak them. But I had a really complicated relationship with my mother growing up where we had a kind of garden of Eden during my childhood. We were intensely, intensely close. And and my mother because of some vulnerabilities that she had reacted with a lot of trouble to my adolescence and to growing independent from her and starting to have different religious views and different political views and all kinds of things. And and we had a pretty intense break that I described in the book and it was so intense that even though after that we still would get together for holidays and talk to each other on the phone and all that, there was a sense in which it was over at that point. The relationship was over. </s><s>[INST] The garden of Eden was no more. [/INST] Yeah. Yeah, it was like a feeling of like, yeah, like I know what Eden was like and it's not there anymore. Um and I think it was all the more confusing because like if you lose someone to actual bereavement, you go through a mourning process and people have thought for thousands of years about how to do that, but with something like this, there's no process because you're not even admitting to yourself, especially when you're like in your teens and twenties that your morning something. So, um but but it was the case that for decades, for decades, I could not answer even the simplest question about my mother, like where did she grow up without tears in my eyes or or more than tears in my eyes, like embarrassing tears. So I would just try to steer the subject in another place. Um but I will say, you know, two things happened. One is that I've spent the last 67 years writing this book about joy and sorrow and loss and love and all of it. And I've really come to terms with all of it. And then the second thing that happened is my mother now has Alzheimer's and in her Alzheimer's </s><s>[INST] she's [/INST] still actually the same person. Like she's forgotten most things, but she still has these conversational lanes that you can travel down that are like the way she always was and the way that she was when I was a kid, which was like so incredibly loving and so connected and so warm and sweet and funny and all of it, all the things I remembered, like it's all come back and for all these decades I had been wondering whether that garden of Eden I remembered had actually happened or whether, you know, that was just like the fantasy of a child and you know, maybe it was always difficult and I had not seen it. But I'm seeing her now and I realize that it was all true. Everything I remember it was all true. It all happened because it's happening again </s><s>[INST] and you return to the garden of Eden for a time and and to childhood. It's always a question of whether you can return to that place. [/INST] Well, I don't know, I don't even know if I'd say I've returned because I'm a different person now and I don't, I need her in. Are you sure </s><s>[INST] You're different than the than the 10 year old? [/INST] Well, no, I mean, I'm the same person in terms of my need for love and love of love and all of that, but I don't look, I'm not dependent on my mother for it the way I was then, and that makes the experience really different </s><s>[INST] younger. She's a she's a God figure what, what is that, that the roots to parents, such a funny civilization will live and there's a depth of connection to parents that's that's probably more powerful than anything else in terms of its formative effect on who you are. [/INST] I think it's the most powerful, and in fact, when this started happening, I got to college and I took a class in creative writing and I tried to write a story, a fictional story, a fictionalized version of what was happening. And I called it the most passionate love Because of what you just said. Um and the teacher actually said to me, she was like, you know, you should put this story in a drawer and not take it out again for 30 years because you're way too close to it. Um so I've now finally written it 30 years later. Mhm. </s><s>[INST] Uh Yeah, but you're probably still too close to it though. [/INST] I don't know though. I mean, I do think like, I think everybody goes through experiences in this life where you're experiencing like a fundamental pain of separation and desire for a union, and it takes so many different forms and this was my primal form of it. But, you know, for someone else, it's a betrayal or bereavement or an exile from a country of their birth or whatever it is, but </s><s>[INST] and then you get to solve that puzzle for for the for the rest of your life. [/INST] Yeah. Fact of like, I really do believe that, you know, that the original love that we long for like that one of the great things that you learn as you grow older is that the that the love exists in some plane, that's more general than the particular Rised form in which you first knew it. </s><s>[INST] Yeah, I mean, that's why despite all the creepy interpretations was why, even though Sigmund Freud is probably wrong in the details, He was the first one to sort of suggest that our experiences. I mean, he said that that that was really controversial at the time when young people, they start having sexual thoughts like at like age two or something, whatever the hell he said, so you develop this kind of connection to uh to the opposite sex or whatever it did to your mother, to your parents. And uh I think while a lot of that is shown to be probably not true, what is like a deeper truth, there is your first early experiences of love or depth of connection are probably somehow strongly formative of your conception of love and your definition of the perfect thing you're reaching for for the rest of your life. [/INST] Yeah, I think that's right. And you can really see it when you become a parent too. You know, you can just see like there's, you know, I have to say like, I mean knock on wood, I actually feel like like we're doing pretty well. Like my kids are teenagers now and I really had thought that I wasn't going to repeat the issues that I had been through with my mom. And I can say I really am not. Um yeah, like my mom for various reasons, just had a lot of trouble with my independence and I just don't feel that at all. So </s><s>[INST] yeah, there might there might be other things you're totally blind to, I [/INST] guess that's possible. </s><s>[INST] Is that the way of parenting is your you solve the problems of the past, but there's some [/INST] Other new one you can find out in 10 or 20 years, but like so far so good. </s><s>[INST] What wisdom about parenting can you, can you can you give from your own experience and from your writing? [/INST] Yeah. Well, oh my God, there's a lot to say so. On the bittersweet side of things, the wisdom that I would give is that especially for kids who are growing up in relative comfort with everything going pretty well. Um they get the idea that real life is when things are going well and when things don't go well, it's like a detour from the main road as opposed to understanding that it's all the main road. Um and I tell the story in the book of this time that we went on this family vacation where we rented a house in the countryside and the house was next to this field, where lived two donkeys that our kids fell in love with and they were really little at the time, two boys and they're spending all this time feeding carrots to the donkeys and it's all beautiful and then comes the day where they realize that we're leaving in like two days and they're never going to see these donkeys again and they start crying themselves to sleep. And the usual things that parents might say at a moment like that, of like, you know, maybe we'll come back or another family will feed them, will feed these donkeys. None of that made any difference. But when we said to them, you know, goodbye is part of life and this is this feeling you're having everybody has it, you've had it before, you're gonna have it again, you'll feel better in a couple of days. Um but this is the way it's supposed to be, this is natural, that's when they stopped crying because I think that's when they stopped resisting. Like it's one thing to feel the pain of goodbye and it's another thing to be feeling like this isn't supposed to be happening. It's the resistance part of this isn't supposed to be happening. That makes life really difficult </s><s>[INST] as opposed to a [/INST] more clear eyed view of what it really is. </s><s>[INST] This is indeed supposed to be happening. There's a show called Yellowstone that recently started watching. [/INST] Yeah, no, I've heard of it. We actually started watching it, but only a few minutes and didn't get into it. </s><s>[INST] So there's just a quick, it's not a spoiler of any kind. But there's a father taking out the sun for the first time to go hunting and you know, to shoot their first buck and the sun is really sad because he pulls the trigger and he took a life. And the father says that everybody gets killed in this life. That's the way of nature. That's the way each one of us is going to get killed. And it's interesting because I didn't really think of it that way because you think you die. But he really framed it as killed because he's like there's no such thing as dying of old age medically. Let's discuss that a bit. But basically there's something whether it's a truck or a bacteria some things go to kill you in the end. And that was an interesting way to look at it because we tend to think of humans aren't aren't supposed to be killed. We think of murder as one of the sins, sort of, one of the things that you don't do in society, but you know what we do. That's a more technical discussion whether we ultimately get killed by something in the end. But to some degree, that's true, at least for most of us that there's something that gets is whether it's cancer, those kinds of things. It's interesting. But yeah, that reframing of it's not, it's supposed to be this is the way of the world. [/INST] So it's funny. I mean, you know, at the same time that I just wrote a whole book about the fact that this is the way it is. Like I really do believe this is the way it is. And and with this reality, there's an intense beauty that comes along with it. </s><s>[INST] So [/INST] we have to accept the reality to get to the beauty. I believe that. And at the same time there's a part of me that's just like, yeah, but give me the magic wand to make the world different. </s><s>[INST] Yeah. I don't know, [/INST] I don't know how much of this is a female thing too. Like I was watching with my son, my 12 year old the other day, we were watching this show about the battle of thermopylae. Um, and and it was like all about, you know, valor and glory on the battlefield. And, </s><s>[INST] and I [/INST] said to him something like, gosh, don't you just wish we lived in a world where you didn't have to do all this in order for everyone just to live their lives. And he just looked at me completely puzzled. No, you know, like to him, it all just seemed self evident that the world would be structured that way, You know, and he had like the 12 year olds like admiration for the dollar of it all. </s><s>[INST] But you wonder if that's um, if that's nature or nurture. I wonder what that world looks like. Um, We do live in a world where murder is seen as bad. But you look at a lot of the human history, I don't know if they had the same kind of conception of that in terms of, you have to ask what kind of murder, uh, you know, for what purpose, you know, war was a way of life. Um, it's interesting, It's interesting if you can imagine properly a future that is different than ours in terms of operating under the different moral systems. But I'd like same with living um indefinitely or living in a society with no war. Like how fundamental is war? How fundamental is death? So [/INST] fundamental to our source code. I just wish that our source code were different basically. Like I can't </s><s>[INST] get people [/INST] that wish </s><s>[INST] there's, there's brain computer interfaces that try to merge so greater and greater with smartphones were already kind of cyborgs but greater and greater merger of computational power. So literally adding source code to our original source code. Just different. There's the mushy biology that run source code and then there's more cold electrical uh systems and then they integrate together and potentially one day we offload the magic that is human consciousness also into the machine and then we'll get to see maybe maybe there'll be a little bit less asshole ish about the whole war thing would be more. But there is, I think even when I when I think about engineering human intelligence or superhuman intelligence systems, I feel like they also need to have the ying yang of life. They have to be able to be afraid and to um to be sad and all those kinds of things. But maybe it's because I'm a product in this particular environment. Maybe sadness is a human, useful human invention but not a universal one. [/INST] This is what I don't know because this is where I come back to the, as I told you like the original reason that I wrote my whole book was the feeling that somehow in the expression of sad music is what other people see when they talk about God, like there's something so there's like an ultimate beauty there. But I don't know if we have access to you without that, but maybe we do. But I can say in this world it's a great way to get access to that state. </s><s>[INST] Is it within the reach of science to deeply understand this. You think to understand that, why you feel sad when you're listening to a song or [/INST] why you feel so much love when you're listening to a sad song, </s><s>[INST] sad song, right? Why, why the sad, why the sad song opens up some kind of deep connection to something you can call divine or something, whatever the heck that is. Yeah, [/INST] I do think, I mean, we have like really early signs of it from the research and I'm sure we're just at the scratching the surface stage. But I mean like we know for example, that, um, you know, the vagus nerve, which it's so fundamental that it governs our breathing and our digestion, </s><s>[INST] our [/INST] vagus nerve also activates when we see another being in distress. There's like an instinctive impulse to want to make it stop. Um, and, and the theory is that that's an evolutionary design. Because we had to be able to respond to the cries of her infants, you know, and from that ability grows the, the greater ability to respond to other people's cries too. So that's probably just, you know, the very first step in being able to understand what all that is. </s><s>[INST] We've already given plenty of advice. But broadly, what advice would you give to young folks today about career, about life? Whether they want to be writers, lawyers, scientists, musicians, and artists, whatever the heck they want to be, how can they live a life? They can be proud of. [/INST] Okay, here's what I think. Um, you should absolutely do that thing that you're dying to do, but you should always have a plan B. Um and like a backup plan and a way of earning a living, no matter what happens, because I feel like people, </s><s>[INST] we [/INST] have this narrative in our culture of like the glamorous thing is to yeah, figure out the thing you love and then risk everything to achieve that. But first of all, a lot of people aren't comfortable with that level of risk. And second when you're living with that level of risk, that's a cognitive load too. And so you don't have the full emotion and heart to be able to focus on the thing that you actually really love because you're like stressed out about it. So I'd say like get the backup plan in place and then do the thing. </s><s>[INST] My advice would be the opposite. [/INST] I'm </s><s>[INST] living the romantic. Well, I think the best the truth is B aware of the costs not having a plan B has to do it deliberately if you don't. But I, you know, I'm with Bukowski and find what you love and let it kill you. I think now you have to actually know your personality. I know if I have a plan B, I will not try as hard on plan A and I would likely take plan B. Because if if planning is the risky thing, I just work way much better when in the state of desperation. So with my back against the wall and you have to know that about yourself. I think that has to do. [/INST] So I think we can refine it to say you actually have to really know yourself and how you respond to different kinds of </s><s>[INST] risks. Like [/INST] I would not do well in that kind of situation. I'd be like up at two in the morning worrying about it. Whereas if I have some, like it doesn't have to be paying the rent in some grand way. But if there's some basic way of paying the rent, then my heart's free to do the thing I really love. </s><s>[INST] That's hilarious. I, for me, the only way I'm free is um when I don't know how I'm gonna be there, right. [/INST] Huh. </s><s>[INST] Yeah, because that's because otherwise I'll find a way to pay the rent that's not at all a source of deep fulfillment for me. So [/INST] I see. So it's like if you don't have like the, what's the expression, I don't know, something like the dog at your back. </s><s>[INST] You know, [/INST] you won't actually do it. </s><s>[INST] I create real or artificial deadlines, anxiety and so on. So yeah, you have to have, [/INST] yeah. So the advice is know your triggers but we're still saying that the same basic thing of like do the things you really love, but just set up the </s><s>[INST] strategy as the rest [/INST] of your </s><s>[INST] personality. [/INST] Exactly exactly. </s><s>[INST] What do you think is the meaning of life, the meaning of this whole thing? Um, [/INST] Probably </s><s>[INST] has something to do with whatever we feel when we listen to us, that song. [/INST] Yeah. Because two things come simultaneously to my mind when you ask that question and I've been asking it since I was four. I remember the first time I did </s><s>[INST] the question is more important than the answer. Probably [/INST] keep </s><s>[INST] asking. [/INST] I don't know. The first one is beauty and I don't know why beauty is so important, but I just know that it is </s><s>[INST] impossible to define. Perhaps [/INST] is it defendable? Um, other than, you know it when you see it? I don't know. I mean, just </s><s>[INST] has to do with that line. You feel something when you just see it or you hear it. [/INST] Yeah, you just see it. And it's like, like some it's whatever can deliver you to that mode of transcendence where you're no longer purely in your own self and you're in something higher. And when you're in those states of mind, you know it because you have the temporary sensation that that you could die at that moment, that the people you love could die and it will all be okay because there's something else. So that's my first answer. And then my second answer is </s><s>[INST] um [/INST] the need to relieve psychic pain, like other people's psychic pain. I don't know why that's just that's just like an impulse that I </s><s>[INST] have psychic pain is more like, like suffering of any form. [/INST] Yeah. But I mean, </s><s>[INST] what is there a particular [/INST] yeah, just making the world better and less pain, less pain to go around in general. Um </s><s>[INST] Hence your this sort of optimistic desire and longing for a world without sort of destruction without malevolent destruction. [/INST] Yeah. World where that wouldn't be necessary. Yeah, exactly, exactly. But yeah, like I so I had this moment, it wasn't so long ago, I was doing some interview and somebody asked me like, what are you longing for right now? And my answer at that moment was like, you know what I'm actually at this moment in life where I'm not longing for anything. I'm at this particular way station where everything is the way I wanted to be. Of course the minute you say something like that, you know, you're going to be proven wrong because like an hour later I get a letter from </s><s>[INST] a [/INST] reader who I've been in touch with over the years and he was telling me about like a psychic struggle that he's going through and and I just feel like, oh my gosh, if there was anything I could do to make it that his life wouldn't have been such that he would be in this position in the first place. Like his struggles had to do with a long life history. So I don't know why I feel </s><s>[INST] that, [/INST] but I do </s><s>[INST] those moments when you were just at peace. There's nothing else you want. Feel like that's like a temporary repose like a pause. You could you bet your ass [/INST] a </s><s>[INST] desire follows that at some point. But you get to enjoy those little moments [/INST] when I said, even when he asked me and I answered that way I said this is a way station. Like I knew it was temporary, but I didn't realize it would be disrupted like an hour later </s><s>[INST] and sort of to give you push back to your to your statement about the possibility of correct because the beauty and basically alleviating suffering, there's a quote I really like from Hunter s Thompson that pushes back against that, which is for every moment of triumph for every instance of beauty, many souls must be trampled. But that's a very hunter? S Thompson. And you know [/INST] how he ended up, he's </s><s>[INST] not the greatest philosopher all of all times, but it's certainly a a beautiful, chaotic human being. Well, [/INST] that's true. And I will tell you that my nickname for my husband is Gonzo kind of because of him. You know, he invented that form of Gonzo journalism where like the the writer is totally in the story and my husband, like that's his personality. He's in everything that he does. He's really in it. He's really present. He just lives that way. So, </s><s>[INST] so [/INST] His name is Kim but I call him so it was like 90% of the time. Well </s><s>[INST] then that's a beautiful way to end this season. This is thank you for your work. Thank you for being who you are thanking you for initially at least making me feel okay about being an introvert and educating and making the rest of us feel great about being introverts, it's like half the world or whatever the heck it is. It's a lot of people thank you for being you. Thank you for talking today, This is awesome, this is fun, [/INST] thank you so much. It was so great to talk to you and I think it was the, what I said to you when we first got connected is thank you for your way of being in the world. Really, really love it. </s><s>[INST] Thanks for listening to this conversation with Susan cain to support this podcast. Please check out our sponsors in the description and now let me leave you with some words from Susan CAin herself. The highly sensitive introvert tends to be philosophical or spiritual in their orientation rather than materialistic or hedonistic. They dislike small talk. They often describe themselves as creative or intuitive. They dream vividly and can often recall their dreams the next day. They love music, nature, art and physical beauty. They feel exceptionally strong emotions sometimes acute bouts of joy, but also sorrow, melancholy and fear, highly sensitive People also process information about their environments, both physical and emotional, unusually deeply. They tend to notice subtleties that others miss another person's shift in mood or a light bulb burning a touch too brightly. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] You know, human beings are able to conceive of the philosophical notion of death of mortality. Is there any of that for chimps? When they're thinking about violence? Is violence? Like what is the nature of their conception of violence? Do you think today? Do they realize they're taking another conscious beings life or is it some kind of like optimization over the use of resources or something like that. [/INST] Mhm. I don't think it's I can't think of any way to get an answer to the question of what they know about that. I think that the way to think about the motivation is rather like the motivation in sex. So when males are interested in having sex with a female, whether it's in chimpanzees or in humans, they don't think about the fact that what this is going to do is to lead to a baby. </s><s>[INST] You're right. [/INST] Mostly, what they're thinking about is I want to get my underway. Uh and um I think that it's a similar kind of process with the chimps, you know, what they are thinking about is I want to kill this Yes, individuals and it's hard to imagine that uh taking the other individuals perspective and thinking about what it means for them to die is going to be an important part of that. In fact, you know, there's reason to think it should not be an important part because it might inhibit them and they don't want to be inhibited, you know, the more efficient they are in doing this the better. But, you know, I think it's interesting to think about this whole motivational question because it does um produced at a rather haunting thought that there has been selection in favor of enthusiasm about killing and in our relatively gentle and deliberately moral society that we have today. It's very difficult for us to face the thought that uh in all of us, there might have been a residue and more than that sort of inactivity an active potential for that thought of really enjoying getting someone else. But I think, you know, one can sustain that thought fairly obviously by thinking of circumstances in which it would be true that the ordinary human male would be delighted to be part of a group that was killing someone. What you've got to do is to be in a position where you're regarding the victim as dangerous and thoroughly hostel. </s><s>[INST] But the pure enjoyment of violence, there's a uh I don't know if you know this historian dan Carlin, here's a podcast. He has an episode 34 hour episode that I recommend to others is quite haunting. But he takes us through an entire history. It's called painful tainment. The history of humans enjoying the murder of others in a large group. So like public executions were part of a long part of human history. And there's something that uh for some reason humans seem to have been drawn to just watching others die. And he ventures to say that that may still be part of us. For example, he said if it was possible to televise to stream online, for example, the execution and the murder of somebody or even the torture somebody that a very large fraction of the population on earth would not be able to look away, they'll be drawn to that somehow has a very dark thought that were drawn to that. So you think that's part of us in there somewhere that selection that we evolved for the enjoyment of killing and the enjoyment of observing uh those in our tribe [/INST] doing the killing. Yes, I mean, and that word you produced at the end is critical, I think because uh, it would be a little bit weird, I think, uh, to imagine a lot of enjoyment about people in your own tribe being killed. I don't think we're interested in violence for violence is sake that much. Um, it's um, it's when you get these social boundaries set up and in today's world happily, we kind of are already one world. You know, we you have to dehumanize someone to get to the point where they are really outside our recognition of a tribe at some level, which is the whole human species. But in ancient times, that would not have been true because in ancient times there are lots of accounts of hunters and gatherers, uh, in which the appearance of a stranger would lead to an immediate response of shooting on sight. Because what was human was the people that were in your society And the other things that actually looked like us and you know, we're human in that sense, were not regarded as human. So there was a kind of automatic dehumanization of everybody that didn't speak our language or hadn't already somehow become recognized as a sufficiently like us to escape the dehumanization contact. </s><s>[INST] And so hopefully the story of human history is that we our uh, that tribalism fades away, that are dehumanization. The natural desire to dehumanize or tendency to dehumanize groups that are not within this tribe decreases over time. And so then the desire for violence decreases over time. [/INST] Yeah, I mean that that that's the optimistic perspective and the great sort of concern of course is that um small conflicts can build up into bigger conflicts and then dehumanization happens and then violence is released as Hannah Aaron says. You know, there currently is no uh known alternative to war as a means of settling really important conflicts. </s><s>[INST] So if we look at the big picture, what role has violence or do you think violence has played in the evolution of homo sapiens? So we are quite an intelligent, got a beautiful particular little branch on the evolutionary tree. Um What part of that was played by our tendency to be violent? [/INST] Well, I think that violence was responsible for creating your homo sapiens. Um and that raises the question of what homo sapiens is. Yes. </s>

<s>[INST] Should the interaction be with like the way humans see it with the buttons and the language or should be with the html javascript and CSS what's what do you think is the better? So [/INST] today, all of this interaction is mostly on the level of html CSS and so on. That's done because of computational constraints. Uh but I think ultimately, um everything is designed for human visual consumption and so at the end of the day there's all the additional information is in the layout of the web page and what's next to you and what's a red background and all this kind of stuff and what it looks like visually. So I think that's the final frontier is we're taking in pixels and we're giving out keyboard mouse commands, but I think it's impractical Still today, </s><s>[INST] do you worry about bots on the internet, Given, given these ideas, given how exciting they are. Do you worry about bots on twitter being not the stupid bots that we see now with the crypto bots, but the bots that might be out there actually that we don't see that they're interacting in interesting ways. So this kind of system feels like it should be able to pass the I'm not a robot click button whatever um which you actually understand how the test works. I don't quite like there's there's a there's a check box or whatever that you click, it's presumably tracking like mouse movement and the timing and so on. So exactly this kind of system we're talking about should be able to pass that. So yeah, what do you feel about? Um bots that are language models plus have some interact ability and are able to tweet and reply and so on. Do you worry about that world? [/INST] Yeah, I think it's always been a bit of an arms race between sort of the attack and the defense. So the attack will get stronger but the defense will get stronger as well, our ability to detect that. How </s><s>[INST] do you defend? How do you detect? How do you know that your account on twitter is is human? [/INST] How </s><s>[INST] did you approach that? Like if people will claim you know uh how would you defend yourself in the court of law that I am a human? [/INST] Um </s><s>[INST] this account at some [/INST] point I think uh it might be I think the society society will evolve a little bit like we might start signing digitally signing some of our correspondence or you know things that we create right now. It's not necessary but maybe in the future it might be I do think that we are going towards the world where we share we share the digital space with a eyes, </s><s>[INST] synthetic beings. [/INST] Yeah. And they will get much better and they will share our digital realm and they'll eventually share our physical realm as well. It's much harder. But that's kind of like the world we're going towards and most of them will be benign and hopeful and some of them will be malicious and it's going to be an arms race trying to detect them. </s><s>[INST] So I mean the worst isn't the ai is the worst is the Ai is pretending to be human. So I don't know if it's always malicious. There's obviously a lot of malicious applications but it could also be, you know, if I was an ai I would try very hard to pretend to be human because we're in a human world. I wouldn't get any respect as an A I want to get some love and respect. [/INST] I don't think the problem is intractable. People are people are thinking about the proof of personhood and uh we might start digitally signing our stuff and we might all end up having like uh yeah, basically some some solution for proof of personhood. It doesn't seem to be intractable. It's just something that we haven't had to do until now. But I think once the need like really starts to emerge which is soon. I think people think about it much more </s><s>[INST] so, but that too will be a race because obviously you can probably uh spoof or fake the proof of personhood. So you have to try to figure out how to [/INST] probably. I </s><s>[INST] mean it's weird that we have like social security numbers and like passports and stuff. It seems like it's harder to fake stuff in the physical space, in the visual space. It just feels like it's gonna be very [/INST] tricky, very </s><s>[INST] tricky to out because it seems to be pretty low cost fake stuff. What are you gonna put an ai in jail for like trying to use a fake fake personhood proof? I mean, okay fine, you put a lot of guys in jail, but there will be more as our like exponentially more, the cost of creating bot is very low. Unless there's some kind of way to track accurately. Like you're not allowed to create any program without showing uh tying yourself to that program. Like any program that runs on the internet, you'll be able to uh trace every single human program that was involved with that [/INST] program. Yeah. Maybe you have to start declaring when uh you know, we have to start drawing those boundaries and keeping track of okay, what are digital entities versus human entities and uh what is the ownership of human entities and digital entities and uh something like that. Um I don't know, but I'm I think I'm optimistic that this is uh this is uh possible and in some, in some sense, we're currently in like the worst time of it because um all these bots suddenly have become very capable, but we don't have defenses yet built up as a society and but I think that doesn't seem to be intractable. It's just something that we have to deal with. </s>

<s>[INST] Let me zoom in on a very particular aspects of street fighting. Where with all due respect, I disagree with mr joe Rogan and George ST Pierre on, which is uh, the suit and tie situation. Now, to criticize GSB Yeah, he's very accomplished and everything, but to criticize him for a bit, he made claims about how dangerous the tie is in a street fighting situation without ever having used in in a fighting situation. So he made sort of broad proclamations without understanding the fundamentals. So I thought I would go to somebody who thinks in systems, what do you think? Is it um, is it dangerous to wear a tie or not in a grappling situation versus all the other weapons [/INST] in the street flight here? It's a strange wear a tie and a grappling competition. It </s><s>[INST] would be, it would be yes. In a in a street fight situation, joe Rogan thinks it is like the most dangerous. It's like it becomes your weakest point if you wear a tie because it's very easy to choke George ST, Pierre seemed to have agreed with that. Also, George added that you can grab the tie and pull the person down uh to a knee. Yeah, this is the go to general gonna go for the choke George ST Pierre will go for the tide to the knee, which I was saying is ridiculous. So what do you think? [/INST] Okay. Um, first up, I actually can speak with experience on this because they worked as a bouncer for over a decade and most of the clubs I worked that did not require a suit and tie, but occasionally they did. Um, okay, let's first differentiate between the kinds of threats when you wear a time. Yes, if you wear a tie, if there is going to be a threat by far the more important threat is not strangulation. Okay, Being strangled by your tie is possible, but it is a poor choice. There are many other ways to strangle people that are far more efficient if I strangle with by your tie, I'm literally in front of you. Um, that means as I go to apply the stranglehold, I can easily be eye gouges sarah sarah. If you're gonna strangle people on the street do it from behind and there's just much better ways to do it </s><s>[INST] in that here. That joe Rogan [/INST] uh, with regards to the snap down question, that's, that is more a problem. I always recommend if you are going to work as a bouncer with a tie. We're a clip on tie. So it just comes off immediately. If you don't like clip ons then you can use a bow tie. I used to work for years with uh in hip hop clubs, with members of the nation of Islam security team that were known, they had various factions, but the one I worked with the X men and they would always wear bow ties, which of course can't be grabbed now um the boat I was, it was a recognizable part of their brand, a security guard. So everyone knew that that's what they wore. If if I wore a bow tie and the security situation, people would probably think that I was some kind of nancy boy and uh and and want to fight with me, so I couldn't wear one, so I would always wear a tie, which you should become familiar with. Mr Freeman, that's the texas bola tie, which is a kind of shoe string tie, which is very, very thin, almost like shoestring and rather short and just has a simple pendant in the middle. Um this is perfect if you need to wear a tie in a situation where you believe there's a high likelihood, have you been grab because I can't be grabbed, there's nothing to grab. It's literally like string, like if you pull that, they were just slips through your hand. Um that tie that you're wearing now, that would give me tremendous control of your head and I could easily turn into a hockey fight situation where your head was being pulled down out of balance and you would have a hard time recovering. Um, so strangulation, not really a problem, getting pulled down, possible problem solutions, clip on tie, bow tie or if you don't want to look like a nancy boy, we're a bola tie. </s><s>[INST] Beautiful. So you disagree with joe Rogan, agree with George ST pierre. I love it. I feel like, I feel like this is an instruction we put together on uh, on street fighting and the tie. Speaking of [/INST] joe Rogan, let me ask uh, </s><s>[INST] the following question, he's uh, currently doing a podcast with Gordon [/INST] Ryan </s><s>[INST] and probably going to try to convince him and you as he's already been doing to move to Austin, what are the chances of the donna, her death squad coming in to Austin and opening a school in Austin and making Austin home so I can attend the classes there. [/INST] I would definitely have to think about that. Um, I do know that I personally love new york. Yes, but every single person in the squad despised new york and wanted to leave for a long time. So, um, uh, </s><s>[INST] what was the nature of your love in new york by the way, [/INST] it was truly an international city? Like I'm a big believer in the idea of breadth of experience and if you want breath of experience usually requires extensive travel, but training people means you have to be in a fixed location, working according to a schedule. And that pushed those two push in different directions. New york was the compromise where everyone from around the world came there. So you had breadth of experience of world culture, but at the same time you had a fixed location so you could run a training program that produced world champions. So it was the ideal compromise. Um it was a fascinating thing to teach classes of over 120 people were literally, the entire world was represented on the map and go outside and see the same thing. It was truly uh, the world's leading international city. It was um, it was like the world's unofficial capital, fascinating place to live. So I loved it, but the squad headed for them, it was like an expensive um things, they'd never actually lived in Manhattan, they always lived in New Jersey or Long Island had to commute. And so all they ever saw was the bridges and the tunnels, the expensive daily parking fees, they only saw the worst of new york. And despite my pleas for them to move into Manhattan, they never did. And so they hated it because when all you see in new york is the bridges and the tunnels and the parking garage, That's not a pleasant thing. So I, I understand where they're coming from. So uh then when Covid broke out, um they wanted to move to Puerto rico and and work there. Now Puerto rico is a beautiful alternative to new york. It's in many ways has many advantages over new york. It's physically beautiful. The people are wonderful and uh, it's a, it's a, it's a, just a wonderful place to spend time. Um, </s>

<s>[INST] in a sense that when they grow up it's uh some incredible ideas will be coming from them. So there's certainly a technical aspect to your work, but you also have a management aspect to your role with tensorflow, leading the project, a large number of developers and people, so what do you look for in a good team, What do you think, You know, google has been at the forefront of exploring what it takes to build a good team and tensorflow is one of the most cutting edge technologies in the world. So in this context, what do you think makes for a good team? [/INST] It's definitely something I think a favorite about, I think the uh in terms of, you know, the team being able to deliver something, well, one of the things that's important is a cohesion across the team, so being able to execute together and doing things, it's not like at this scale and individual engineer can only do so much, there's a lot more that they can do together, even though we have some amazing superstars across google and in the team, but there's, you know, often the way I see it as the product of what the team generates is way larger than the hole or, you know, the uh individual put together, and so how do we have all of them work together? The culture of the team itself, hiring good people is important, uh but part of that is, it's not just that okay, we hire a bunch of smart people and throw them together and let them do things, it's also people have to care about what they're building, people have to be motivated for the right kind of things, that's often an important factor. Um and, you know, finally, how do you put that together with uh somewhat unified vision of where we want to go? So are we all looking in the same direction or just going all over? And sometimes it's a mix of google is a very bottom up organization in some sense, uh also research even more so, uh and that's how we started, but as we've become this larger product and ecosystem, I think it's also important to combine that well with uh mix of, okay, here's the direction we want to go in, there is exploration, we'll do around that, but let's keep staying in that direction are just all over the place. </s><s>[INST] And is there a way you monitor the health of the team, sort of like, is there a way, you know, you did a good job team is good, like, I mean, you're sort of, you're saying nice things, but it's sometimes difficult to determine how aligned, because it's not binary, it's not like it's, there's tensions and complexities and so on and the other element to visit the measure superstars, you know, there's so much even a google, such a large percentage of work is done by individual superstars too. So there's a, uh and sometimes those superstars could be against the dynamic of a team and those, those tensions have, it was a, I mean, I'm sure in tensorflow it might be a little bit easier because the mission of the project is so sort of beautiful, you're at [/INST] the cutting edge is </s><s>[INST] exciting. Uh but have you had struggled with that? Has there been challenges? [/INST] There are always people challenges in different kinds of fairs that I think we've been, what's good about getting people who care and are, you know, have the same kind of culture and that's google in general to a large extent, but also, like you said, given that the project has had so many exciting things to do, there's been room for lots of people to do different kinds of things and grow, which, which does make the problem a bit easier, I guess, and it allows people depending on what they're doing, if there's room around them and that's fine. But yes, we do it, we do cared about whether superstar or not, like they need to work well with the team </s><s>[INST] across google, that's interesting to hear. So it's like uh superstar, not the productivity broadly is about the team. [/INST] Yeah, I mean, they might add a lot of value, but if they're putting the team then that's a problem. </s><s>[INST] So, in hiring engineers, it's so interesting. Right, the hiring process, what do you look for? How do you determine a good developer, a good member of the team from just a few minutes or hours together. [/INST] So </s><s>[INST] again, no magic answers, I'm [/INST] sure Yeah, and google has a Hiding process that we've refined over the last 20 years I guess and that you've probably heard and seen a lot about, so we do work with the same hiding process and that that's really helped for me in particular. I would say in addition to the core technical skills, what does matter is their motivation in what they want to do because if that doesn't align well with where you want to go, that's not going to lead to long term success for either of them or the team. Um And I think that becomes more important, the more senior the person is, but it's important at every level, like even the junior most engineer, if they're not motivated to develop what they're trying to do, how we're smart there, it's going to be hard for them to succeed. </s>

<s>[INST] Isn't that exciting? Because the wearable stuff with physiological information, it feels like it violates privacy less than the non contact face based methods. [/INST] Yeah, it's interesting. I think what people sometimes don't, you know, it's been in the early days, people say, oh, wearing something or giving blood is invasive, right? Whereas a camera's less invasive because it's not touching you. I think on the contrary, the things that are not touching you or maybe the scariest because you don't know when they're on or off and you don't know when people and you don't know who's behind it, right? A wearable uh depending upon what's happening to the data on it. If it's just stored locally or if it's streaming and what it is being attached to uh in in a sense, you have the most control over it because it's also very easy to just take it off right now, it's not sensing me. So if I'm uncomfortable with what it's sensing now, I'm free, right? If I'm comfortable with what it's sensing then and I happen to know everything about this one and what it's doing with it. So I'm quite comfortable with it. Uh then I'm, you know, I have control, I'm comfortable control is one of the biggest factors for an individual in reducing their stress. If I have control over it, if I know all there is to know about it, then my stress is a lot lower and I'm making an informed choice about whether to wear it or not or want to wear it or not. I wanna wear it sometimes, maybe not others. </s><s>[INST] Right? So that control, I'm with you, that control, even if you have the ability to turn it off is a really important [/INST] thing is huge. And we need to maybe, you know, if if there's regulations, maybe that's number one to protect his people's ability to, it's easy to opt out as to opt in. </s><s>[INST] So you've uh studied a bit of neuroscience as well, how have looking at our own minds sort of the biological stuff for the the neuro biological, the neuroscience, look at the signals in our brain, help you understand the problem and the approach of effective computing. So [/INST] originally I was a computer architect that I was building hardware and computer designs and I wanted to build ones that work like the brain. So I've been studying the brain as long as I've been studying how to build computers, </s><s>[INST] have you figured out anything yet? [/INST] Very little. It's so amazing. You know, they used to think like, oh, if you remove this chunk of the brain and you find this function goes away. Well, that's the part of the brain that did it. And then later they realized if you remove this other chunk of the brain, that function comes back and oh no, we really don't understand it, brains are so interesting and changing all the time and able to change in ways that will probably continue to surprise us when we were measuring stress. You may know the story where we found an unusual big skin conductance pattern on one wrist and one of our kids with autism and in trying to figure out how on earth you could be stressed on one wrist and not the other. Like how can you get sweaty on one wrist? Right. When you, when you get stressed with that sympathetic fight or flight response, like you kind of should like sweat more in some places than others, but not more on one wrist than the other. That didn't make any sense. We learned that what had actually happened was a part of his brain had unusual electrical activity and that caused an unusually large sweat response on one wrist and not the other. And since then we've learned that seizures caused this unusual electrical activity and depending where the seizure is, if it's in one place and it's staying there, you can have a big electrical response. We can pick up with a wearable at one part of the body. You can also have a seizure that spreads over the whole brain. Generalist grand mal seizure and that response spreads and we can pick it up pretty much anywhere as we learned this and then later built embrace that's now FDA cleared for seizure detection. We have also build relationships with some of the most amazing doctors in the world who not only helped people with unusual brain activity or epilepsy but some of them are also surgeons and they're going in and they're implanting electrodes. Not just to momentarily read the strange patterns of brain activity that we'd like to see return to normal. Um but also to read out continuously what's happening in some of these deep regions of the brain during most of life when these patients are not seizing most of the time, they're not seizing most of the time they're fine. And so we are now working on mapping those deep brain regions that you can't even usually get with E. G. Scalp electrodes because the changes deep inside don't reach the surface. Uh But interesting when some of those regions are activated we see a big skin conductance response who would have thunk it right? Like nothing here but something here. In fact right after seizures that we think are the most dangerous ones that precede what's called Sudeep sudden unexpected death in epilepsy uh there's a period where the brain waves go flat and it looks like the person's brain has stopped, but it hasn't uh the activity has has gone deep into a region that can make the cortical activity look flat like a quick shutdown signal here. Um It can unfortunately cause breathing to stop if it progresses long enough before that happens, we see a big skin conductance response in the data that we have, the longer this flattening, the bigger our response here. So we have been trying to learn, you know, initially, like why why are we getting a big response here when there's nothing here? Well, it turns out there's something much deeper so we can now go inside the brains of some of these individuals, fabulous people who usually aren't seizing and get this data and start to map it. So that's an active research that we're doing right now with with top medical partners. </s>

<s>[INST] so it's adding almost [/INST] like a </s><s>[INST] time to mention too, is trying to trying to optimize across time. So it's considering when, when you're programming, [/INST] you're not thinking. And absolutely, and so the risk era made things so risc chips, ri sc the the risk risk chips as opposed to Sisk chips. The RISC chips made things more complicated for the compiler because um what they ended up doing is ending up adding pipelines to the processor where the process can do more than one thing at a time, but this means that the order of operations matters a lot. And so one of the classical compiler techniques that you use is called scheduling and so moving the instructions around so that the processor can act like keep its pipelines full instead of stalling and getting blocked. And so there's a lot of things like that, that are kind of bread and butter compiler techniques have been studied a lot over the course of decades now, but the engineering side of making them real is also still quite hard. And you talk about machine learning, this is this is a huge opportunity for machine learning because many of these algorithms are full of these like hokey hand rolled heuristics, which worked well on specific benchmarks, don't generalize and full of magic numbers. And you know, I hear there's some techniques that are good at handling that. </s><s>[INST] So what would be the uh if you were to apply machine learning to this, what's the thing you try to optimize it? Ultimately the running time [/INST] of uh you can pick your metric and there's there's running time, there's memory use, there's there's lots of different things that you can optimize for code size is another one that some people care about in the embedded space. </s><s>[INST] Is this like the thinking into the future? Or somebody actually been crazy enough to try to have machine learning based parameter tuning for optimization of compilers. [/INST] So this is something that is, I would say research right now, um there are a lot of research systems that have been applying search in various forms and using reinforcement learning is one form, but also brute force search has been tried for quite a while and usually these are small, small problem spaces so find the optimal way to co generate a matrix multiply for a gpu right? Something like that where you say there there's a lot of uh design space of do you unroll loops a lot? Do you execute multiple things in parallel? And there's many different confounding factors here because graphics cards have different numbers of threads and registers and execution ports and memory bandwidth in many different constraints interact in nonlinear ways. And so search is very powerful for that and it gets used in in certain ways but it's not very structured. This is something that we need, we as an industry need to fix, </s><s>[INST] She said 80s. But like so have there been like big jumps in improvement and optimization? Yeah. Since then what's [/INST] Yeah, it's largely been driven by hardware. So hardware's hardware and software. So in the mid nineties, java totally changed the world. Right? And I'm still amazed by how much changes introduced by the way or in a good way. So like reflecting back java introduced things like all at once introduced things like jit compilation. None of these were novel, but it pulled it together and made it mainstream and made people invest in it. Jit compilation, garbage collection, portable code, safe code, like memory safe code. Um like a very dynamic dispatch execution model. Like many of these things which had been done in research systems and have been done in small ways in various places really came to the forefront, really changed how things worked and therefore changed the way people thought about the problem, javascript was another major world change based on the way it works um but also on the hardware side of things. Um uh multi core and in vector instructions really change the problem space and are very um they don't remove any of the problems that we face in the past, but they add new kinds of problems of how do you find enough work to keep a four wide vector busy? Right. Or if you're doing a matrix multiplication, how do you do different columns of that matrix in at the same time. And how do you maximally utilize the arithmetic, compute that one core has. And then how do you take it to multiple course? How </s><s>[INST] did the whole virtual machine things change the compilation pipeline? [/INST] Yeah. So, so what what the job of virtual machine does is it splits just like I was talking about before, where you have a front end that parses the code and then you have an intermediate representation that gets transformed what java did was they said we will parse the code and then compiled to what's known as java byte code. And that bike code is now a portable code representation that is industry standard and locked down and can't change. And then the back part of the compiler, the optimization and cogeneration can now be built by different vendors. Okay. And java byte code can be shipped around across the wire, it's memory safe and relatively trusted </s><s>[INST] because of that. It can run in the browser [/INST] and that's why it runs in the browser. Right. And so that way you can be in, you know, again back in the day you would write a java applet and you use as a, as a web developer, you build this mini app that runs on a web page. Well a user of that is running a web browser on their computer. You download that java byte code which can be trusted and then you do all the compiler stuff on your machine so that you know that you trust that </s>

<s>[INST] Can Satoshi with Bitcoin to build in uh, resistance mechanism or prevention mechanism for the land grab. [/INST] It would be hard to because, you know, if you have capital and resources, I mean if it was a proof of work chain, you just have people that would invest a ton of money and mining for instance. But most new Blockchain as cryptocurrencies are just sold basically they're issued in token offerings kind of thing. So, </s><s>[INST] so it's hard to enforce through the protocol, the gist, centralization of control. [/INST] Our would be challenging to and people have tried to do air drops, you know, where they, you know, distribute coins. A large number of people basically doesn't work. Most people don't care about the airdrop. So it's hard to have an equitable distribution. I think the conditions of Bitcoins launch, we're so lucky and favorable that they are very unlikely to be replicated. So I do think it's going to be real challenge to ever have a new competitor that's as decentralized as leaderless as dispersed, distributed as Bitcoin has, has its credibility. I don't know how you could overrule it all those important features. </s><s>[INST] What about Bitcoin's comparison to other current cryptocurrencies? So Bitcoin versus ethereum for example, why? Um is it possible that ethereum overtakes Bitcoin? [/INST] That's certainly possible. I'm not ruling it out a theory. Um leadership is sort of wise enough to understand that they shouldn't compete with Bitcoin on those most profound qualities like ethereum doesn't really aspire to be more um sound from a monetary perspective than Bitcoin. Right. The in fact the theory in leadership are sort of constantly tweaking the monetary policy. So they went for a completely different trade off, right? They also don't compete to be as decentralized from a governance perspective, right? Because there's leadership, there's any foundation, there's a charismatic leader of metallic and a theorem has this policy of hard forks. So, and Bitcoin, hard forks are extremely rare in a theory um that it's the default way to change things. So it's a much more adaptive system and it changes more frequently, but that also means that it's they're incurring more risk when they introduced those changes, there's much more complexity. So ethereum is smart because they sort of understood Bitcoin is the top dog when it comes to a sound money, a digital gold type thing and they went for all of the different trade offs. They wanted to be more of a platform. They wanted to have more complexity, the transactional air, they wanted to take on more risk in terms of changing the protocol, they wanted to change more quickly. They wanted to make the monetary policy more mutable. So ethereum takes that completely different tack of course, you know, not ruling out that it could take over Bitcoin from a marketing perspective, it's just a very different system. And I tend to think the big one is the most disruptive one because it's the most equipped to challenge sovereign currencies in the grand scheme. </s><s>[INST] Do you think they can coexist? So like in the future is do you see a world where you know uh ethereum captures some large percent of the market and but nevertheless the minority, [/INST] 100%, 100% Bitcoin has already been tokenized and put onto ethereum many units of Bitcoin. I think over a billion dollars worth. So not only did they coexist they are actually mutually stick. So they're like two creatures that have this, you know, it's like the the rhino and like the bird, the pecs, the parasites off the rhino's back or whatever, right? So I don't know which is which and the analogy, but </s><s>[INST] yeah, I don't know who the parasites [/INST] are, you know, or you know, the alligator and the teeth cleaning fish or whatever. Right? So you know I always wondered why the alligator doesn't just eat the fish. But uh I guess they're brushing its teeth basically. So um ethereum is it gives you more transactional flexibility, there is much more experimentation happening there. It has this whole decentralized finance element. There's a huge number of Bitcoins that circulate on the ethereum protocol because the theory um is open to other asset types basically. So I think that's actually a creative to both systems because ethereum gets to have this good form of collateral Bitcoin on the system which is good volatility characteristics and then it's a supply sink for Bitcoins which are sort of now they're injected into this third party protocol and that I think reduces the velocity of Bitcoin overall. It's probably good for the valuation. </s><s>[INST] So you see it's quite possibly could be a symbiotic relationship. That's really interesting. [/INST] I think so. I think so. </s><s>[INST] What are your thoughts about metallic Burin? What are your thoughts about some of the other figures in the space outside of Bitcoin? [/INST] I think metallic made some mistakes with the theory ultimately. Like I disagree with some of the decisions that were made along the way. Like there's this infamous case of this bailout where 14% of Ether was lost in the smart contract or really the smart contract that a lot of veteran leadership were sort of backing and supporting was hacked and then the foundation with metallic support chose to make a change to the underlying protocol to undo the hack right? So to me that was not the most prudent approach because you're basically violating the core protocol rules in order to undo you know to bail out a specific contract which has failed granted there was a lot of ether in there But I think that shook the credibility of the ethereum system that happened back in 2016. I think That was one reason why I I became disenchanted with ethereum. </s><s>[INST] So basically even if in that case that might fix an important problem that opens the door to centralized like manipulation of the protocol in the future. [/INST] Yeah, basically demonstrates that there's certain elites at the protocol level that can exercise specific control over the system and you know a lot of people have lost money and hacks on ethereum and a lot of contracts have gone south, a huge amount of value but they didn't get a bailout and it was just when you know this specific contract called the DAO de ao was hacked the you know the leadership intervened and and you know to their credit, they haven't had a significant intervention or bailout since then but it did normalize the practice and I think it weakened the social contract. So I would prefer that you sort of bite the bullet in that situation and you accept the failure of that contract. There </s>

<s>[INST] I wonder if there's ideas like big cloud for example, I don't know if you saw that if there's ideas built on top of this concept doesn't have to be like a theory of mine F. D. It could be just the concept of non fungible tokens whether those kinds of things can take hold and they, it's less about financial transactions and more about almost like um I don't I don't know how to put it, but like staking identity in some way, whether it's Big cloud or identity of objects, like there might be some way of connecting physical reality and digital reality in some interesting ways. So just the financial aspect is a way to like put some validity behind the identity. I wonder if there's like ideas there that are yet to be discovered or ideas that yet to take hold. Like Big Cloud seems interesting. Seems shady as how it seems a little scam me. I don't know if I like the idea that you can bet on people essentially. [/INST] Yeah, I think my market cap on bit cloud is like $90,000 and I haven't done anything there. So </s><s>[INST] did you take, did you like take like verify yourself or [/INST] whatever? If not, I think people would yell at me on twitter if I did so </s><s>[INST] and it's unclear whether it's a scam yet or not, right? It's unclear where it's coming from. [/INST] Well, there's some details about the, you know, investors. It's backed by some pretty big name investors. So I probably wouldn't use the word scam to describe it, but it's got Ponzi like dynamics, like everything in crypto. So there is very questionable and then also is using people's likeness without their permission, which is I think a legal question, you know, so there's open questions around it, but you know, is our public block chains and you know that sort of architecture, is that going to be useful for decentralized, alternative forms of social media? 100%. Yes. You know, I'm super super bullish on that idea, basically creating open protocols, open name spaces, ways to organize without the dependence on a single node effectively in silicon Valley, you know, the twitter node or the facebook node. I think it's a matter of urgency that we create digital gathering spaces where you have strong property rights, you know, you have a claim on your identity, you have a claim on your data and open architecture is our way to do that. I don't know if it will be a Blockchain, but certainly I think the the general, you know, concept introduced by block chains is a good template for how to, you know, organize these systems, </s><s>[INST] value freedom, value, decentralization of power, whatever the mechanism let me ask you about love. So there is a Bitcoin maximalist community that sometimes, so those folks in general have a strong belief that Bitcoin is good for the world and it's almost an ethical imperative to, to sort of help Bitcoin succeed, which I think is uh, as a member of any community, I think it's beautiful to believe in the vision of the community, right? There does seem to be some properties of what some may call like toxicity or derision and mockery and those kinds of things. Um you know uh some folks have criticized this right that Bitcoin maximal is um is not necessarily good for the world, even if Bitcoin is good for the world. Well what are your thoughts about this kind of approach philosophically or practically to uh the spread of Bitcoin? And is there a way that we can add more love to the world while we add more Bitcoin to the world? [/INST] That's a great question. I mean, you know, Bitcoin is sort of what you make of it, so you can define your own path as you advocate for Bitcoin or don't for that matter. So my chosen approach is the approach you see here, which I try to minimize the amount of sort of harshness or mockery, although I've been known to be mean on twitter too, you know what </s><s>[INST] twitter is a specific, sorry to interrupt is a specific media where this takes its worst form. So I'm learning, listen, I'm actually because of this podcast, but in general and part of different communities and some are full of like on the bash it love and some are like when I experienced on twitter the Bitcoin community at first I was off put in terms of the intensity, the mockery, I bet layers of lull, like the layers of not taking anything seriously and I think there's power to that there's freedom to that. I appreciate it. I have respect for it but it's not my thing on twitter. It's just not, not the way I enjoy communicating on twitter. [/INST] I retired from twitter. I had I had 100,000 followers and then I retired so I'm free now. I don't have to tweet anymore. It's great. But I totally can see the point. I wish that corners were gentle or in their approach. Not all big corners are like that of course there's you know 50-100 million of them worldwide and a few tens of thousands on twitter. So I'm not going to claim that they're necessarily representative. The toxicity though is kind of a learned habit because Bitcoin has had so many episodes where strong willed institutions today's billionaires. The dice are pretty toxic. You could say right? I'm basically mocking critics of Bitcoin </s><s>[INST] but at the same time you're saying that the criticism has been predictable and repeatable and it's been the same throughout. [/INST] Yeah. And that's a that's a pretty you know dismissive thing to say right. That I can reduce you to uh an algorithm of you know with 11 uh right and yeah permutations. Um but you know the thing to remember I guess is that some of the best funded companies in the Bitcoin space, the most powerful miners billionaires have tried to change and co opt and alter Bitcoin um to shape it to their liking and without these incredibly hard core high priests of the Bitcoin protocol it would have been hopeful is hopelessly malley ated in all number of ways. And so there is a reason why someone would be incredibly protective of Bitcoin. Does that justify events toxicity? You know, on social media? Probably not, but it's a leaderless protocol. So the whole point is that it's money for enemies and you know some of the Bitcoin maximums came for me to when I made suggestions that they didn't like but I'm happy to use it the protocol because I know that that transaction will be final regardless of how odious my counter party is or how how you know politically disfavored their opinions are. </s>

<s>[INST] I should know this. That's very interesting. It's very interesting the connection [/INST] there and that's why research is very, it's very hard to predict the outcomes. And again, the federal government spent a ton of money on things that they thought were useful for resonate Urz, but it ended up enabling these small um, V. S, uh which is great because I could have never raised that much money and told you sold this project, hey, we want to build these small U. V. S. Can you, can you actually fund the development of low cost I am us, </s><s>[INST] so why do you need a name? You are. [/INST] So So I was, I'll come back to that, but but so in 2007, were able to build these and then the question you're asking was a good one. How do you coordinate the motors to, to, to develop this? Uh, but over the last 10 years everything is commoditized, high school kid today can pick up a raspberry pi kit and build us. All the low level functionality is all automated, but basically at some level you have to drive the motors at the right RPMs, the right velocity in order to generate the right amount of trust in order to position it and oriented in a way that you need to in order to fly. Um The feedback that you get is from on board sensors and the I. M. U. Is an important part of it. The I. M. You tells you what the acceleration is as well as where the angular velocity is. Uh And those are important pieces of information. In addition to that you need some kind of local position or velocity information. For example when we walk we implicitly have this information because we kind of know how how what our stride length is. Uh We also are looking at images fly past our retina if you will. And so we can estimate velocity. We also have accelerometers in our head and we're able to integrate all these pieces of information to determine where we are as we walk. And so robots have to do something very similar. You need an I am you you need some kind of a camera or other sensor that's measuring velocity. Um And then you need some kind of a global reference frame if you really want to think about uh doing something in a world coordinate system. And so how do you estimate your position with respect to the global reference frame? That's important as well. </s><s>[INST] So coordinating the RPMs of the four motors is what allows you to first of all fly and hover and you can change the orientation and the velocity of the and so on [/INST] exactly how much of </s><s>[INST] degrees of freedom that you're [/INST] Playing 6° of freedom, but you only have four inputs the four motors. And uh and it turns out to be a remarkably versatile configuration. You think at first world I only have four motors, how do I go sideways? But it's not too hard to say, well, if I tell myself I can go sideways, um and then you have four motors pointing up, how do I, how do I rotate in place about a vertical axis? Well, you rotate them at different speeds and that generation reaction moments and that allows you to turn. So it's actually a pretty, it's an optimal configuration from from an engineering standpoint. Um It's very simple um very cleverly done and and very versatile. </s><s>[INST] So if you could step back to a time. So I've always known flying robots as uh to me it was natural that the quadcopter should fly, but when you first started working with it, how surprised are you that you can make do so much with the four motors, How surprising is that you can make this thing fly first of all, you can make it hover, then you can add control to it [/INST] firstly. This is not the four motor configuration is not ours, you can it has at least 100 year history. Um and various people, various people try to get quad rotors to fly without much success. As I said, we've been working on this since 2000. Our first designs were, well, this is way too complicated. Why not? We try to get an Omni directional flying robots are so our early designs, we had eight voters. And so these eight rotors were arranged uniformly on a sphere if you will. So you can imagine a symmetric configuration and so you should be able to fly anywhere. But the real challenge we had is the strength to weight ratio is not enough. And of course we didn't have the sensors and so on. Um, so everybody knew, or at least the people who worked with Rhoda crafts new four rotors, we get it done. Um, so that was not our idea, but it took a while before we could actually do the onboard sensing and the computation that was needed for the kinds of agile maneuvering that we wanted to do in our little aerial robots. And that only happened Between 2007 and 2009 in our lives. </s><s>[INST] Yeah. And you have to send the signal 500 times a second. So the compute there everything has to come down in price and what are the steps of getting from point A to point B. So you we just talked about like local control. But if all the kind of cool uh, dancing in the air that I've seen, you show, how do you make it happen? It would trajectory make a trajectory first of all? Okay, figure out a trajectory. So planning trajectory and then how do you make that trajectory happen? [/INST] I think planning is a very fundamental problem in robotics, I think, you know, 10 years ago it was an esoteric thing. But today with self driving cars, you know, everybody can understand this basic idea that a car sees a whole bunch of things and it has to keep a lane or maybe make a right turn or switch lanes. It has to plan a trajectory, it has to be safe, it has to be efficient. So everybody is familiar with that. That's kind of the first step that that you have to think about when you when you when you when you say autonomy. Um And so for us it's about finding smooth motions, motions that are safe. Um So we think about these two things. One is optimal. Itty one of safety. Clearly you cannot compromise safety. Uh So you're looking for safe, optimal emotions. Uh The other thing you have to think about is can you actually compute a reasonable trajectory in a fast, in a in a small amount of time because you have a time budget. So the optimal becomes sub optimal but uh in our lab we we focus on synthesizing smooth trajectory that satisfy all the constraints. In other words don't violate any safety constraints. Um and is as efficient as possible. And when I say efficient, it could mean I want to get from point A to point B as quickly as possible or I want to get to it as gracefully as possible. Um Or I want to consume as little energy as possible, </s>

<s>[INST] And that's a kind of immortality is ideas that you've invented living on long after you in the minds of others and humans are ultimately not, are like meat vehicles that carry ideas for a brief for just a few years may not be the important thing. It might be the ideas that we carry with us and invent new ones. Like we get a bunch of baby ideas in our head, We borrow them from others and then maybe we invent a new one and then you one might have a life of its own. And it's fun. It's fun to think about that idea of living from many centuries to come unless we destroy ourselves, but maybe a, I will borrow it and we'll remember Poe as like that one human that, that that helped us out before we of course killed him and the rest of human civilization On that note. Well this is a huge honor. You're one of the great educators I've ever gotten a chance to interact with. So it's it's truly an honor that you would talk with me today. It means especially a lot that you would travel out to Austin to talk to me. It really means a lot. So thank you so much. Keep on inspiring. And I'm one of your many, many students. Thank you so much for talking today. [/INST] Thank you. Thank you. It's actually a real honor for me to talk to you and to get this chance to have this really intellectual conversation through all of these topics. Thanks bro, </s><s>[INST] Thanks for listening to this conversation with pasta and lo and thank you to Jordan's harbored the show on it. Better help eight sleep and element check them out in the description to support this podcast. And now let me leave you with some words from Isaac Newton. I can calculate the motion of heavenly bodies but not the madness of people. Thank you for listening. I hope to see you next time [/INST] </s>

<s>[INST] That's magic spoon dot com slash lex and use gold lex. This is a lex Friedman podcast And here is my conversation with Kathryn declare why is Pluto not a planet anymore. Does this upset you or has justice finally been served? [/INST] So I get asked this all the time. I think all, all planetary scientists get asked about Pluto, especially by kids who we just love for Pluto to still be a planet. Um, but the reality is um when we first discovered Pluto, it was, it was a unique object in the outer solar system. And we thought we were adding a planet to the inventory of planets that we had. And then over time it became clear that Pluto was not a unique large objects in the outer solar system, that there were actually many of these. Um, and as we started discovering more and more of them, we realized that the concept of Pluto being a planet um didn't make sense unless maybe we added all the rest of them as planets. So you know, you could have imagined actually a different direction that this could have gone where all the other objects that were discovered in that belt or at least all the ones, let's say above, above a certain size became planets instead of Pluto being declassified. Um, but we, we're now aware of many objects out there in the outer solar system and what's called the kuiper belt that are of the same size or in some cases even larger than Pluto. Um so the declassification was really just a realization that it was not in the same category as the other planets in the solar system. And we basically needed to refine our definition in such a way that took into account that there's this belt of debris out there in the outer solar system of things with a range of sizes. </s><s>[INST] Um Is there a hope for clear categorization of what is a planet or not? Or is it all just gray area when you study planets? When you study moon, satellites of those planets? Is there are lines that are that could be cleanly drawn or is it just a giant mess? This is all like a fluid. Let's not mess. But it's like fluid of what is the planet? What is a moon of a planet? What is debris? What is asteroids? All that kinda? [/INST] So there are technically clear definitions that were set down by the IU the international astronomy union. Um </s><s>[INST] is it size related? Like what are the parameters business? [/INST] So the parameters are that it has to orbit the Sun which was essentially to rule out satellites. Of course this was a not very forward thinking definition because it technically means that all extrasolar planets according to that definition are not planets. Um So it has to order orbit the sun. It has to be large enough that its gravity has caused it to become spherical in shape. Um which also applies to satellites and also applies to Pluto. The third part of the definition is the thing that really rules out everything else, which is that it has to have cleared out its orbital path. Um And because Pluto orbits in a belt of material, it doesn't satisfy that stipulation, </s><s>[INST] why didn't you clear out the path is not big enough to knock everybody out of the way. [/INST] Um and this actually is not the first time it has happened so series when it was discovered, ceres is the largest asteroid in the asteroid belt. And it was originally considered a planet when it was first discovered. And it went through exactly the same story history where people actually realized that it was just one of many asteroids in the asteroid belt region and then it got declassified to an asteroid and now it's back to a dwarf planet. So there is a lot of reclassification. So to me as somebody who studies solar system objects, um I I just personally don't care my level of interest and something has nothing to do with what is classified as. So my favorite objects in the solar system are all moons. And frequently when I talk about them, I refer to them as planets because to me they are planets. They have volcanoes, they have geology, they have atmospheres, their planet like worlds. And so the distinction is not super meaningful to me, but i it is important just for having a general framework for for understanding and talking about things to have a precise definition. </s><s>[INST] So you don't have a special romantic appreciation of a moon versus the planet versus an asteroid. It's just an object that flies out there and doesn't really matter what the categorization is because there's movies about asteroids and stuff. And then there's like and then there's movies about, you know, the moon, whatever. It's a really good movie. You know, there's something about moons that that's almost like an outlier. Like you think of the moon as a thing that's the secret part and the planet is like the more like vanilla, regular part. None of that. You don't have any of that. [/INST] No, I actually do. I really satellites are, the moons are my favorite things in the solar system and I think part of what you're saying, I agree from maybe a slightly different perspective, which is from the perspective of exploration, we've spent a lot of time sending spacecraft missions to planets. We had a mission to jupiter, we had a mission to Saturn. We have plenty of missions to mars and missions to venus. I think the exploration of the moons in the outer solar system is the next frontier of solar system exploration </s>

<s>[INST] no, no. I I think about it a lot. I mean that's a really good thing to meditate on. We can talk about maybe how while you bring that up, what if it is or not a heavy question, but I get uh I think about mortality a lot and uh for me it feels like a really good way to focus in on is what you're doing today, The people you have around you, the family you have, is it uh does it bring you joy? Does it bring you fulfillment? And basically uh for me of long ago, okay, try to be ready to die any day. So like today, uh you know, I, I kind of woke up, look if I was nervous about talking to um have I really admire your work. And the book is very good and super exciting topic. Uh but then you know, there's this also feeling like if this is the last conversation I have in my life, you know if I die today, let's be let's be uh the right like my glad today happened and it is and I am glad today uh happened. So that that's the way [/INST] that's so unique. I never got that answer from a single person. </s><s>[INST] The business of life. There's goals, their dreams, there's like planning [/INST] very few people make it happen. That's what I learned. And so a lot of these people </s><s>[INST] like you run out of time, [/INST] it's not so much amount of time, but I'd come back later and be well okay, why don't you do that? If that's what you would do if you're going to die a year from now, why don't you, why don't you make it real simple things spend more time with family. Like why why don't you do that? And no one had an answer. It turns out unless you usually, unless you have, you really do have a pressing end of life. People don't do their bucket list or try to change their career. And some people can't so we can't like for a lot of people they can't do anything about it and that's that's fine. But the ones who can take action for some reason never do. And that was one of the ways that mike's death at the time, his impending death really really affected me because you know for these sick people what I learned, he had a bucket list and he was able to do some of the bucket list. It was awesome. But he got sick pretty quickly. So if you do only have a year to live, it's ironic because you can't do, you can't do the things you wanted to do because you get too sick too fast. </s><s>[INST] What were the bucket list things for you that you realize like what am I doing with my life? [/INST] That was the major concept of him after he died. I didn't know like I I was just lost because when something that profound happens all the things I was doing, most of the things I was doing were just meaningless was so tough to, to find an answer for that. And that's when I settled on I'm going to devote the rest of my life to trying to find another earth and to find out too find that we're not alone. </s><s>[INST] What is that longing for connection with others? Um What's that about? What do you think? Why is that so full of meaning? [/INST] I don't know why. I mean I think it's how we're hardwired. Like one of my friends some time ago actually my dad died. He never heard someone say this before but he's like Sarah you know, why are we evolved to take death so harshly what kind of society would we be if we just didn't care people died. That would be a very different type of world. How would we as a species have got to where we are? So I think that is tied hand in hand with, Why do we, why do we see connection? It's just that what we were talking about before, that sub consciousness that we don't understand. </s><s>[INST] Yeah coupled, you know, the other side the flipside of the coin of connection and, and love is a fear of loss. It's like that was again, I don't know. It's so, it makes you appreciate the moment Is that the thing ends? [/INST] Yeah, it's definitely a hard won, the thing ends, but it's hard to not, you, you wouldn't want to limit like it's like my dog who I love so much. I'll start to cry like I can't think about the end. I know he'll age much faster than I will and some day will end, right? But it's too sad to think of. But should I not have got a dog, I should have not brought this sort of joy into my life because I know it won't be forever. It's </s><s>[INST] well, there's uh, there's a philosopher Ernest Becker who wrote a book, denial of death. I just, uh, and warm with the cores. And there's another book talks about terror management theory, Sheldon Solomon. I just talked to him a few weeks ago, uh, the brilliant philosopher psychologist, that their theory, whatever you make of it, is that the fear of death is at the core of everything, everything we do. So like you're that you think you don't think about the mortality of your dog, but you do and that's what makes the experience rich. Like there's this kind of like in the shadows looks the knowledge that this won't last forever and that makes every moment just special in some kind of weird way that the moments are special for us humans? Uh I mean sorry to use romantic terms like love, but what do you make, what did you learn about love from, from losing it from losing your husband? [/INST] Well, I learned to love the things I have more. I learned to love the people that I have more and to not let the little things bother me as much. </s>

<s>[INST] Well so the limit and by the way I didn't know about Argentina, why'd you end up in Argentina? [/INST] Okay so I was basically um you know I was living in Houston texas after college where I went to school and I had never studied abroad um I kind of like I don't know, I felt like I needed some adventure or something in my life and I was like I was running this other startup that I was trying at the time a tutoring company and I could, I could work from anywhere. So my plan was you know what I'm just gonna go do like a month in every city around south America just like almost like to force myself out of my comfort zone because I I had never traveled by myself to a foreign country or whatever and um I didn't really speak the language and anyway I I landed in Buenos Aires thinking I'd go around south America where I had never been there but I basically once I was set up in South in in Buenos Aires with an apartment and a cell phone and stuff then I was like I don't want to do it all again next month so I just stayed there for most of the time and took some day trips but but it was kind of a formative experience in that regard you </s><s>[INST] got a chance sort of unexpectedly to experience the social effects of hyperinflation which is interesting But also I've never been I really really want to go as a person who likes tango as a person who likes the Argentinian national team and soccer and uh and steak. All right uh and all the other things that Argentina is known for. Okay so um economic freedom, one of the limits on economic freedom comes from government and government regulations and all those kinds of things throughout the world. Um So how does Cryptocurrency help resist that? Can you sort of elaborate a little bit further? What are the things that limit economic freedom? And how does crypto help? You [/INST] know today the world like traditional financial system is basically every country of the world for the most part has their own currency. And so there's a group of people or institutions in each of those countries that's controlling that economic policy of that money supply. And you know it can be it can be manipulated right? So it's not like many of these currencies are not linked to gold standard. You know the U. S. Kind of famously came off that in the 19 seventies for instance. But if you read Ray Dalio and all the stuff that he talks about, there's thousands of fiat currencies that have been in existence over time and basically all of them eventually get disconnected from um backing of like hard commodities and then they get over inflated and and and printed and so in times of stress um you know with Nixon I guess it was like in the US it was the Vietnam war or something like that. It kind of drove government spending and so under times of stress they say, hey it's a temporary measure. We need to break. The peg um temporary was like, you know, famous words that he used and and they go they go like they go print. And so the bad thing about that of course is that it sort of erodes people's like wealth if they can only hold their assets in cash, which basically like poor people tend to do that. You know, if you're wealthy you can hold um stocks or like real estate and things like that, but it's so it's really a tax on the poorest people in society inflation. So anyway, crypto in a way Is a little bit of like um return to the gold standard in this digital era, right? Bitcoin, there's guaranteed scarcity of it. It's deflationary, there's never going to be more than 21 million. Bitcoin. And so that's that's a really important principle. Um I also think, you know, not just Bitcoin, but like Cryptocurrency generally it's really important in terms of this um you asked about regulation, right? So think about like if you wanted to make a global um borrowing and lending marketplace or a global Exchange, you would have to go to all 200 countries in the world sometimes like maybe 50 states in the us and get lending licenses or operating exchange or whatever and you know, they're not gonna, that's just an incredible amount of work and you can't even do business in many of these countries because like, you know, you have to bribe somebody or it's corrupt or whatever. And so, but with defi decentralized finance people have published um you know, like, you know, swap is a decentralized exchange, everybody in the world, no matter what country you're in, what jurisdiction can can interface with that decentralized exchange because it and there's no central um company operating it, it's it's a it's a it's a smart contract on the ethereum Blockchain, which is globally decentralized. So there's no throat to choke. There's no one person you can, or a company you can go to like, hey, shut this thing down. Even if everybody who's working on, you know, swap today stopped the, you know, the smart contract would continue to operate on the ethereum Blockchain. Um Similarly for like a borrowing lending marketplace, you know, like </s><s>[INST] it's, [/INST] you know, if you want somebody in India wants to borrow from somebody in the U. S. Or whatever, like there's very difficult to do that in the traditional financial world, but in a smart contract that's decentralized, you can enable anybody to access it. So it's it's really kind of this great democratizing force that is creating a new financial system that is more fair and more free um Yeah, and in some ways it is um it's a it's a clever, it's a clever way that's not, you know, it's it's enabling people to do that in a novel way. </s>

<s>[INST] And so we see this as ultimately a benefit to them as well. [/INST] Can you say how much stretch costs? Not yet. But I will say that when we engage with our customers, they'll be able to see a return on investment </s><s>[INST] in typically two years. Okay, so that's something that you're constantly thinking about how, and I suppose you have to do the same kind of thinking with Spot. So it seems like with Stretch, the application is like directly obvious. Yeah, it's a slam dunk. [/INST] Yeah, and so you have a little more flexibility. Well, I think we know the target. We know what we're going after. And with Spot, it took us a while to figure out </s><s>[INST] what we were going after. Well, let me return to that question about maybe the conversation you were having a while ago with Larry Page, maybe looking to the longer future. Of social robotics, of using Spot to connect with human beings, perhaps in the home. Do you see a future there? If we were to sort of hypothesize or dream about a future where a Spot-like robots are in the home [/INST] as pets, a social robot? We definitely think about it. And we would like to get there. We think the pathway to getting there is likely through these industrial applications and then mass manufacturing. Let's figure out how to build the robots, how to make the software so that they can really do a broad set of skills. That's gonna take real investment to get there. Performance first, right? The principle of the company has always been, really make the robots do useful stuff. And so the social robot companies that tried to start someplace else by just making a cute interaction, mostly they haven't survived. And so we think the utility really needs to come first. And that means you have to solve some of these hard problems. And so to get there really needs to come first. We're gonna go through the design and software development in industrial. And then that's eventually gonna let you reach a scale that could then be addressed to a consumer-level market. And so, yeah, maybe we'll be able to build a smaller Spot with an arm that could really go get your beer for you. But there's things we need to figure out still. How to safely, really safely. And if you're gonna be interacting with children, you better be safe. And right now we count on a little bit of standoff distance between the robot and people so that you don't pinch a finger in the robot. So you've got a lot of things you need to go solve </s><s>[INST] before you jump to that consumer-level product. Well, there's a kind of trade-off in safety because it feels like in the home, you can fall. Like you don't have to be as good at, like you're allowed to fail in different ways, in more ways, as long as it's safe for the humans. So it just feels like an easier problem to solve because it feels like in the factory [/INST] you're not allowed to fail. That may be true, but I also think the variety of things a consumer-level robot would be expected to do will also be quite broad. They're gonna want to get the beer and know the difference between the beer and a Coca-Cola or my snack. And they're all gonna want you to clean up the dishes from the table without breaking them. Those are pretty complex tasks. And so there's still work to be done there. </s><s>[INST] Yeah, so to push back on that, here's where application, I think they'll be very interesting. I think the application of being a pet, a friend. So like, no tasks, just be cute. Because not cute, not cute. The dog is more, a dog is more than just cute. A dog is a friend, is a companion. There's something about just having interacted with them and maybe because I'm hanging out alone with robot dogs a little too much. But like there's a connection there and it feels like that connection is not, should not be disregarded. [/INST] No, it should not be disregarded. Robots that can somehow communicate through their physical gestures are you're gonna be more attached to in the long run. Do you remember Ibo, the Sony Ibo? They sold over 100,000 of those, maybe 150,000. Probably wasn't considered a successful product for them. They suspended that eventually and then they brought it back, Sony brought it back. And people definitely treated this as a pet, as a companion. And I think that will come around again. Will you get away without having any other utility? Maybe in a world where we can really talk to our simple little pet because chat GPT or some other generative AI has made it possible for you to really talk in what seems like a meaningful way. Maybe that'll open the social robot up again. That's probably not a path we're gonna go down because again, we're so focused on performance and utility. We can add those other things also, but we really wanna start </s><s>[INST] from that foundation of utility, I think. Yeah, but I also wanna predict that you're wrong on that. So which is that the very path you're taking which is creating a great robot platform will very easily take a leap to adding a chat GPT like capability, maybe GPT-5. And there's just so many open source alternatives that you could just plop that on top of spot. And because you have this robust platform and you're figuring out how to mass manufacture it and how to drive the cost down and how to make it reliable, all those kinds of things, it'll be a natural transition to where just adding chat GPT on top of it will create- [/INST] I do think that being able to verbally converse or even converse through gestures, part of these learning models is that you can now look at video and imagery and associate intent with that. Those will all help in the communication between robots and people, for sure. And that's gonna happen obviously more quickly </s>

<s>[INST] It could be just a gigantic background model, language model. And then you [/INST] do </s><s>[INST] um the training becomes like prompting that model to um [/INST] to </s><s>[INST] essentially do a kind of quarrying a search into the space of the things that's learned to customize that to whatever problem you're trying to solve. So maybe if you form a large enough language model, you can go quite quite a long way that, you [/INST] know, I think there's some truth to that. I mean you look at the way humans approach um a game like poker, they're not coming at it from scratch, They're coming at it with a huge amount of background knowledge about, you know, how humans work, how the world works. Um the idea of money, so they're able to leverage that kind of information to to pick up the game faster. Uh So it's not really a fair comparison to then compare it to an ai that's like learning from scratch. And maybe one of the ways that we address this uh sample complexity problem is by allowing ai is to leverage that general knowledge across a ton of different domains. </s><s>[INST] So like I said you did uh a lot of incredible work in the space of research and actually building systems, what advice would you give to? Uh let's start with beginners, What advice would you give to beginners interested in machine learning just there at the very start of their journey and they're in high school and college thinking like this seems like a fascinating world. What advice would you give them? [/INST] Um I would say that there are a lot of people working on similar aspects of machine learning and to not be afraid to try something a bit different. My my own path in ai is pretty atypical for machine learning researcher today. I mean, I started out working on game theory and um and then shifting more towards reinforcement learning as time went on and that actually had a lot of benefits, I think because it allowed me to look at these problems in a very different way from the way a lot of um machine learning researchers view it. Um and that comes with drawbacks in some respects. Like I think there's definitely aspects of machine learning where, you know, I'm weaker than most of the researchers out there, but I think that diversity of perspective, um you know, when I'm working with my teammates, um there's something that I'm bringing to the table, there's something that they're bringing to the table and that kind of collaboration becomes very fruitful for that reason. </s><s>[INST] So there could be problems like, like poker, like you've chosen diplomacy, that could be problems like that still out there that you can just tackle even if it seems extremely difficult. [/INST] Um I think that there's a lot of challenges, challenges left? And I think having a diversity of viewpoints and backgrounds is really helpful for working together to figure out how to tackle those kinds of challenges </s><s>[INST] as a beginner. So that I would say that's that's more for like a grad student, but already built up a base like a complete beginner, What's a good journey? So for you that was doing some more on the math side of things, doing game theory, all that. So it's basically built up a foundation in something, so programming mathematics, it could even be physics, but build that foundation. [/INST] Yeah, I would say build a strong foundation in math and computer science and statistics in these kinds of areas, but but don't be afraid to try something that's different and learn something that's different from, you know, the thing that everybody else is doing to get into machine learning, um you know, there's there's value in having a different background than everybody else. Um Yeah, so, but certainly having a strong math background, especially in things like linear algebra and statistics and probability um are incredibly helpful today for for learning about and understanding machine learning. </s><s>[INST] Do you think one day we'll be able to, since you're taking steps from poker to diplomacy one day we'll be able to uh figure out how to live life optimally. [/INST] Well, what is it like in in poker and diplomacy? You need a value function, You need to have a reward system and so what does it mean to live a life that's optimal. </s><s>[INST] So, okay, so then you can exactly like lay down a reward function being like I want to be rich or I want to be um I want to be in a happy relationship and then you'll say, well do x, [/INST] you know, there's there's a lot of talk today about in ai safety circles about like miss specification of, you know, reward function. So you you say like, okay, my objective is to be rich and maybe the ai tells you like, okay, well if you want to maximize the probability that you're rich, go rob a bank. And so you wanna, is that is that really what you want, is your objective really to be rich at all costs? Or is it more nuanced than that </s><s>[INST] understand the consequences? Yeah. Yeah. So yeah, that that that's maybe life is more about defining the reward function that minimizes the unintended consequences than it is about the actual policy that gets you to the reward function. Maybe life is just about constantly updating the reward function. [/INST] I think one of the challenges in life is figuring out exactly what that reward function is. Sometimes it's pretty hard to specify the same way that, you know, trying to handcraft the ultimate policy in a game, like chess is really difficult. Um It's not so clear cut what the reward function is for for life, </s><s>[INST] I think one day I will figure it out and I wonder what that would [/INST] be. Until </s>

<s>[INST] that I shouldn't be too happy with. But I'm, I'm okay with all of that. And I still got love in [/INST] your heart. I mean, I just talked with him like last year, or I guess it was even this year about mentioning that I'm going off doing this AI stuff. I'm, I'm going big into artificial intelligence. And he had a bunch of ideas for how AI is going to play into gaming. And, you know, asked if I was interested in collaborating. And it's not in line with what, with what I'm doing. But I do, you know, I wish almost everyone the best. I mean, I, I know I may not have parted on the best of terms with, I, you know, with some people, but I was thrilled to see Tom Hall, I writing VR games now he wrote, working on a game called Demio, which is really an awesome VR game. It's like Dungeons and Dragons. We all used to play Dungeons and Dragons together. That was one of the things that was what we did on Sundays. In the early days, I would dungeon master and they'd all play. And I, you know, </s><s>[INST] so it really made me smile seeing Tom involved with an RPG game in virtual reality. You were the CTO of Oculus VR since 2013, and maybe less than your involvement a bit. In 2019, Oculus was acquired by Facebook, now meta in 2014. You've spoken brilliantly about both the low level details, the experimental design and the big picture vision of virtual reality. Let me ask you about the, the metaverse, the big question here, both philosophically and technically. How hard is it to build the metaverse? What is the metaverse in your view? You started with discussing and thinking about Quake as a kind of metaverse. As you think about it today, what is the metaverse, the thing that could create this compelling user value, this experience that will change the world, and how hard is it to build it? [/INST] So the term comes from Neil Stevenson's book Snow Crash, which many of us had read back in the 90s. It was one of those kind of formative books. And there was this sense that the, the possibilities and kind of the, the freedom and unlimited capabilities to build a virtual world that, that does whatever you want, whatever you ask of it, has been a powerful draw for generations of developers, you know, game developers specifically, and people that are thinking about more, more general purpose applications. So we were talking about that back in the Doom and Quake days about how do you wind up with an interconnected set of worlds that you kind of visit from one to another. And as web pages were becoming a thing, you start thinking about how, what is the interactive kind of 3D based equivalent of this? And there were a lot of really bad takes you had, you know, like Vermal then virtual reality markup languages. And there's aspects like that, that, that came from people saying, well, what would, what kind of capabilities should we develop to, to enable this? And that kind of capability first work has usually not panned out very well. On the other hand, we have successful games that started with things like Doom and Quake and communities that formed around those and whether it was server lists in the early days or literal portalling between different games. And then modern things that are on completely different order of magnitude like Minecraft and Fortnite that have a hundred million plus users. You know, I still think that that's the right way to go to build the metaverse is you build something that's amazing that people love and people wind up spending all </s><s>[INST] their time in because it's awesome. And you expand the capabilities of that. So even if [/INST] it's a very basic experience, Minecraft is Minecraft is amazing case study. And so many things where what's been able to be done with that is really enlightening. And there are other cases where like right now Roblox is basically a game construction kit aimed at kids and that was capability first play and it's achieving scale that's on the same order of those things. So it's not impossible, but my preferred bet would be you make something amazing that people love and you make it better and better. And that's where I could say we could have gone back and followed a path kind of like that in the early days. If you just kind of take the same game, whether it's when Activision demonstrated that you could make call of duty every year. And not only is it not bad, people kind of love it. And it's a, it's very profitable. The idea that you could have taken something like that, it take a great game, release a new version every year that lets the capabilities grow and expand to start saying it's like, okay, it's a game about running around and shooting things, but now you can have, I bring your media into it. You can add persistence of social sense signs of life or whatever you want to add to it. I still think that's quite a good position to take. And I think that while META is doing a bottoms up capability approach with Horizon Worlds, where it's a fairly general purpose, creators can build whatever they want in their sort of thing. It's hard to compare and compete with something like Fortnite, which also has enormous amounts of creativity, even though it was not designed originally as a general purpose sort of thing. So there's, we have examples on both sides. Me personally, I would have bet on trying to do entertainment, valuable destination first and expanding from </s>

<s>[INST] Can they pull themselves into something that looks more like a symmetric war as opposed to a symmetric one? So accidentally engaging on the lone individual than getting themselves into trouble. Are they more aggressive in avoiding these kinds of battles? [/INST] They're very, very keen to avoid those kinds of battles, but occasionally, uh, they can make a mistake. Um, but so far there have been no observations, anything like a battle in which both sides maintain themselves. And I think you can very confidently say that overwhelmingly. What happens is that if they discover that there's several individuals on the other side, then both sides retreat. Nobody wants to get hurt. What they want to do is to hurt others. </s><s>[INST] Yes, So you mentioned jane Goodall, you've worked with her. What was it like working with her? What have you learned from her? [/INST] Well, she's wonderfully independent, courageous person, you know who she famously began her studies not as a qualified person um in terms of education, But qualified only by enthusiasm and a considerable experience even in her early 20s with nature. So she's courageous in the sense of being able to take on challenges. The thing that is very impressive about her is her total fidelity to the observations. Very unwilling to extend beyond the observations, waiting until they mount up. And you've really got a confident picture and tremendous attention to individuals. So that was an interesting problem from her point of view because When she got to know the chimpanzees of Gumby, this particular community of Kazakh alot about 60 individuals. So Gandhi was in Tanzania on Lake Tanganyika. She was there initially with her mother and then alone for uh two or three years of really intense observation um and then slowly joined by other people. What what she discovered was that there were obvious differences in individual personality. And the difficulty about that was that when she reported this to the largest scientific world. Initially, her advisers at Cambridge uh they said well you know we don't know how to handle that because you've got to treat all these animals as the same basically. Because um there is no research tradition of thinking about personalities. Well. Now Whatever it is 60 years later uh the study of personalities is a very rich part of the study animal behavior. Um It anyway the important point in terms of, you know what she like is that she stuck to her guns and she absolutely insisted that you know, we have to show, describe in great detail the differences in personality among these individuals and then you can leave it to the evolutionary biologists think about what it means. </s><s>[INST] So what is the process of observation like this? Uh Like observing the personality but also observing in a way that's not projecting your beliefs about human nature or animal nature onto chimps which is probably really tempting to project. So your understanding of the way the human world works, projecting that onto the chimp world. [/INST] Yes. I mean it's particularly difficult with chimps because chimps are so similar to humans in their behavior that it's very easy uh to to make those projections as you say. The process involves um making very clear definitions of order behavior is you know aggression can be defined in terms of a forceful hit a bite and so on. Um And writing down every time these things happen and then slowly totting up the numbers of times that they happen. You know from individual A. Towards individuals B. C. D. And E. Uh so that you build up a very concrete picture rather than interpreting at any point and stopping and saying well you know they seem to be rather aggressive. So the uh the formal system is that you build up a pattern of the relationships based on a description of the different types of interactions. The aggressive and the friendly interactions and all of these are defined in concrete and so that from that you extract a pattern of relationships and the relationships can be defined as um relatively friendly, relatively uh aggressive competitive based on the frequency of these types of interactions. And so one can talk in terms of individuals having relationship, which on the scores of friendliness is two standard deviations outside the mean. I mean, you know, it's weird </s><s>[INST] which direction? Sorry? Uh Both directions. [/INST] Well, I mean, you know that that there will be obviously the friendly ones would be the ones who have exceptionally high rates of spending time close to each other, of touching each other in a gentle way of grooming each other. Uh And by the way, finding that those things are correlated with each other. So it's possible to define a friendship with a capital F in a very systematic way. And and to compare that between individuals, but also between um communities of chimpanzees uh between different species. So that, you know, we can say that in some species, individuals have friends and others they don't at all. What about just </s><s>[INST] because there's different personalities and because they're so fascinating. What about sort of falling in love or forming friendships with chimps? You know, like really, you know, um connecting with them as an observer. What what role does that play? Because you're tracking these individuals that are full of life and intelligence form for long periods of time. Plus as a human, especially in those days for jane she's alone observing it gets lonely as a human. I mean, probably deeply lonely as a human being observed these other intelligent species. [/INST] It's a very reasonable question. And of course jane, in as early years, I think she's willing now to talk about the fact that she regrets to some extent how close she became. Um And the problem is not just from the humans. The problem is from the chimpanzees as well because uh they do things that are extremely affectionate if you like. You know, at one. offered a ripe fruit to a chimpanzee called david graybeard, David graybeard took it and squeezed her hand as if to say thank you. And then I think you gave it back if I remember rightly. </s>

<s>[INST] so I clear is the new conference, new ish. [/INST] Uh Yeah, I guess so. </s><s>[INST] And I cmos around in two [/INST] 1000 I see Ml predates that </s><s>[INST] I think my most cited. I see ml papers from 94, [/INST] Michael knows this better than me because of course he's significantly older than I. But the point is, what is the difference? The difference between ice ml in Europe in the late 90s, early 2000s. </s><s>[INST] I don't know what everyone else's perspective would be, but I had a particular perspective at that time which is I felt like I see mel was more of a of a computer science place and the nips europe's was more of an engineering place. Like the kind of math that happened at the two places as a computer scientist I felt more comfortable with the I. C. M. L. Math and the number of people would say that that's because I'm dumb and that's such an engineering thing to say. So [/INST] I agree with that. Part of what I do a little differently. Actually had a nice conversation with tom Dietrich about this in twitter on twitter just a couple of days ago I put a little differently which is that I see Ml was machine learning done by computer scientists and nerves. Was machine learning done by computer scientists trying to impress statisticians. Which was weird because it's the same people at least by the time I started paying attention. But it just felt very very different. And I think that that perspective of whether you're trying to impress the statisticians or you're trying to impress the program which is actually very different and has real impact on what you choose to worry about and what kind of outcomes you come too. So I think it really matters. Think competition, statistics as a means to an end. It is not an end in some sense. Um And I think that really matters here in the same way that I don't think computer science is just engineering or just science or math or whatever but okay </s><s>[INST] so I'd have to now agree that now we agree on everything. [/INST] Yes. Yes. The important thing here is that my opinions may have changed. But not the fact that I'm right I think is what what we just came through. </s><s>[INST] My opinions may have changed and not the fact that I'm [/INST] wrong. That's right. </s><s>[INST] Lost me. I'm not [/INST] even think I lost myself there to wear. Well </s><s>[INST] this happens to us sometimes we're Sorry, how does neural networks change this? Just leaving linger on this topic? Change this idea of statistics. How big of a pie statistics is one of the machine learning thing, like because it sounds like hyper parameters and also just the role of data. You know that people are starting to use this terminology software two point oh, which is like the act of programming as a as a like you're a designer in the hyper parameter space of neural networks and you're also the collector and the organizer and the cleaner uh of the data and that's part of the programming. Uh So how did on in europe's versus I CML topic? What's the role of neural networks and redefining the size and the role of machine learning. [/INST] I can't wait to hear what Michael thinks about this. But I would add one. But that's true. I will force myself. I think the there's one thing I would add to your description which is the kind of software engineer parties. What does it mean to debug for example. But this is a difference between the kind of computational statistics view of machine learning and computational view of machine learning which is I think one is worried about the equation as it were and by the way this is not a value judgment, I just think it's about perspective. But the kind of questions you would ask, you start asking yourself what does it mean to program and develop and build the system is a very computer science view of the problem. When if you get on data science, twitter and check on twitter, you actually hear this a lot with the the economist and the data scientist complaining about the machine learning people was with it's just statistics and I don't know why they don't don't see this, but they're not even asking the same questions, they're not thinking about it as a kind of programming problem. And I think that that really matters just asking this question I actually think is a little different from uh programming and hyper parameter space and sort of collecting the data. But I do think that that immersion really matters. So I'll give you a quick a quick example the way I think about this. So I teach machine learning Michael and I have Qatada Machine Learning class which has now reached out of 10,000 people at least for the last several years for somewhere there's about and my machine learning assignments are of this form. So the super the first one is something like implement these five algorithms, K and N and S. S. VMS and boosting and decision trees and neural networks and maybe that's it, I can't remember and when I say implement, I mean still the code. I am completely uninterested. You get zero points for getting the thing to work and I want you spending your time worrying about getting the corner case right of you know what happens when you are trying to normalize distances and the points on the thing and so you divide by. I'm not interested in that right steal the code. However you're going to run those algorithms on two datasets. The data sets have to be interesting. What does it mean to be interesting? Well datasets interesting if it reveals differences between algorithms which presumably are all the same because they can represent whatever they can represent. And two data sets are interesting together if they show different differences as it were And you have to analyze them, you have to justify their interesting this and you have to analyze the whole bunch of ways. But all I care about is the data in your analysis, not the programming and I occasionally end up in these long discussions with students. Well I don't really, I copy and paste the things that I said the other 15,000 times it's come up which is good but the only way to learn really understand it's to code them up which is a very program or software engineering view of the world. If you don't program that you don't understand it which is by the way, I think is wrong in a very specific way, but it is a way that you come to understand because then you have to wrestle with the algorithm. But the thing about machine learning is not just sorting numbers were, in some sense, the data doesn't matter what matters. Is that well, this algorithm working these abstract things upon the less the other machine learning the data matters, it matters more than almost anything and not everything, but almost anything. And so as a result, you have to live with the data and don't get distracted by the algorithm per se. And I think that that focus on the data and what it can tell you and what question is actually answering for you as opposed to the question you thought you were asking is a key and important thing about machine learning and is away the computational ist as opposed to statisticians bring a particular view about how to think about the process. The statisticians, by contrast being I think I'd be willing to say a better view about the kind of formal math that's behind it and what an actual number ultimately is saying about the data and those are both important, but they're also different. </s>

<s>[INST] you're almost making an argument against uh earlier statement, you've made your convincing me at first you said there's no way a computer could raise a child essentially. But now you've described making me think of it, if you have that kind of idea meritocracy, you have this rigorous approach to Bridgewater takes an investment and apply to raising a child. It feels like through the process you just described, we could as a society, arrive at a set of principles for raising a child and encoded into a computer. [/INST] That originality will not come from machine learning the first </s><s>[INST] time you do so that the original Yes, that's [/INST] what I'm referring to. </s><s>[INST] But eventually, as we could together develop it and then we can automate it. [/INST] That's why I'm saying the processing can be done by the computer. So we're saying the same thing. We're not inconsistent. We're saying the same thing that the processing of that information in those algorithms can be done by the computer in a very, very effective way. You don't need to sit there and process and try to weigh all those things in your equation and all those things. But that notion of okay, how do I get at that principle? </s><s>[INST] And you're saying you'd be surprised uh how much [/INST] you can express? That's right. You can do that. So this is where uh I think you're going to see the future. And right now we, you know, go to our devices and we get information to a large extent and then we get some guidance. We have our GPS and the like in my opinion. Principles. Principles. Principles. Principles. I want to emphasize that you write them down. You've got those principles, they will be converted into algorithms for decision making. And they're going to also have the benefit of collective decision making because right now individuals based on what's stuck in their heads are making their decisions in very ignorant ways. They're not the best decision makers. They're not the best criteria and their operating when those principles are written down and converted into algorithms, it's almost like you look at that and follow the instructions and it'll give you better results. Medic medicine will be much more like this. You can go to your local doctor and you could ask his point of view and whatever and he's rushed and he may not be the best doctor around and you're going to go to this thing and get that same information or just automatically have an input in that and it's going to tell you okay here's what you should go do and it's going to be much better than your local doctor and that the converting of information into intelligence. Okay. Intelligence Is the thing we're coming out with um again I'm 70 and I want to pass all these things along so um all these tools that I've found the need to develop all over these periods of time, all those things I want to make them available and what's gonna happen as you there, they're going to see this. They're going to see these tools operating much more that way? The idea of converting data into intelligence Intelligence for example on what they are like what are your strengths and weaknesses? Intelligence on who do I work well with under what circumstances? Intelligence. We're going to go from what are called systems of record which are a lot of okay information organized in the right way to intelligence. And we're gonna that that'll be the next big move in my opinion. And so you will get intelligence back </s><s>[INST] And that that intelligence comes from reducing things down two principles into [/INST] that's how it happens. </s><s>[INST] So what's your intuition if you look at the future societies do you think well be able to reduce a lot of the The details of our lives down two principles that would be further and further automated? [/INST] I think the real question hinges on people's emotional emotions and irrational behaviors. I think that there's subliminal things that we want. Okay and then there's cerebral, you know, conscious logic and the too often are at odds. So there's almost like to use in you, right? And so let's say what do you want? Um your mind will answer one thing, your emotions will answer something else. So when I think about it I think emotions are I want inspiration, I want, love is a good thing, being able to have a good impact, but it is in the reconciliation of your subliminal wants and your intellectual wants so that you really say they're aligned and so to do that in a way to get what you want. So irrationality is a bad thing if if it means that it doesn't make sense in getting you what you want, but you better decide what you you're satisfying is at the lower level you emotional subliminal one or is it the other. But if you can align them. So what I find is that by going from my you experience the decision, do this thing subliminally and that's the thing I want, it comes to the surface, I find that if I can align that with what my logical me wants and do the double double check between them and I get the same sort of thing that that helps me a lot I find for example meditation is one of the things that helps to achieve that alignment? It's fantastic for achieving that alignment. And often then I also want to not just do it in my head. I want to say does that make sense? Help you. So I do it with other people and I say, okay well let's say I want this thing and whatever does that make sense? And when you do that kind of triangulation, your to use and you do that with also the other way, then you certainly want to be rational, right? But rationality has to be defined by those things </s>

<s>[INST] was the warehouse robots and transport little things. [/INST] So basically instead of a person going and walking around and picking the seven items in your order. Um These robots go and pick up a shelf and move it over in a row where like the seven shelves that contain the seven items are lined up and uh you know, laser or whatever points to what you need to get and you go and pick it and you place it to fill the order and so that people are fulfilling the final orders. What was interesting about that is that when I was asking him about like kind of the impact on labor when they transition that warehouse, the throughput increased so much that the jobs shifted towards the final fulfillment, even though that robots took over entirely the search of the items themselves and the labor, the job stayed like nobody like that was that was actually the same amount of jobs uh roughly they were necessary but the throughput increased by I think over two X or some some amount, right? Like so um you have these situations that are not zero sum games in this really interesting way and the optimist in me thinks that there's these types of solutions in almost any industry where the growth that's enabled creates opportunities that you can then leverage, but you've got to be intentional about finding those and really helping make those links because and even if you make the argument that like there's a net positive locally, there's always tough hits that you got to be very careful about. </s><s>[INST] That's right. You have to have an understanding of that link because there's a short period of time whether training is required or just mental transition or physical or whatever is required. That's still gonna be short term pain. The uncertainty of it. There's families involved, you know, it it's, I mean it's exceptionally is difficult on a human level and you have to really think about that. Even you can't just look at the economic metrics always as human beings. [/INST] That's right. And you can't even just take it as like, okay, well, we need to like subsidize or whatever because like there is an element of just personal pride where majority of people, like people don't want to just be okay, but like they want to actually like have a craft like you said and have a mission and uh I feel like they're having a really positive impact. And so, um, my personal belief is that there's a lot of transferability and skill set um that is possible, especially if you create a bridge and an investment, um, to enable it. Um and to some degree, that's our responsibility as </s><s>[INST] well. Um, this process, uh you mentioned Kiva robots amazon, let me ask you about the astro robot, which is, I don't know if you've seen it, amazon has announced that uh it's a home robot that they have, the screen looks awfully a lot like uh Cosmo has uh I think different vision probably. What are your thoughts about like home robotics in this kind of space. There's, there's been uh quite a bunch of home robots, social robots that very unfortunately have closed their doors. That for various reasons perhaps they were too expensive manufacturing challenges, all that kind of stuff. What are your thoughts about amazon getting into the space? [/INST] Yeah, we had some signs that they were getting into like long, long, long ago, maybe a little bit too interested in Cosmo and that during our conversations, but they're also very good partners actually for us as we kind of just integrated a lot of shared technology. But </s><s>[INST] if I could also get your thoughts on you could think of Alexa as a robot as well. Echo, do you see those as fundamentally different just because you can move and look around is that fundamentally different than the thing that just sits in place? [/INST] Uh It opens up options. Um But uh you know my first reaction is I think like uh I have my doubts that this was going to hit the mark because I think for the price point that it's at and the like kind of functionality and value propositions that they're trying to put out. It's uh it's still searching for like tequila application that like justifies, I think it was like a $1500 price point or kind of somewhere on there that's a really high bar. So there's enthusiast, an early adopters will obviously kind of pursue it, but you have to like really really hit a high mark at that price point which we always tried to, we were always very cautious about jumping too quickly to the more advanced systems that we really wanted to make but would have raised the bar so much you have to be able to hit it uh in today's cost structures and technologies, the mobility is an angle that hasn't been utilized but it has to be utilized in the right way. Um And so that's gonna be the biggest challenges. Like can you meet the bar of what, what the mass market consumer like, you know, think like you know our our neighbors, our friends, parents like would they find a deep devalue like and you know in this at a mass scale that you know that justifies the price point. I think that's in the end of one of the biggest challenges for robotics, especially consumer robotics um where you have to kind of meet that bar. Uh it becomes very very hard. Um </s>

<s>[INST] there's also ethical rules. Um we have to contend with the idea and we have to contend with that idea with ai [/INST] But what's gonna motivate it is, I mean, I talked to people who really miss people who are gone and they would love to get something back even if it isn't perfect. Uh and that's what's gonna motivate this </s><s>[INST] and that person lives on in some form. And the more data we have, the more we're able to reconstruct that person and allow them to live [/INST] on. And eventually as we go forward, we're going to have more and more of this data because we're going to have none of us that are inside our neocortex and we're gonna collect a lot of data, in fact anything that's data is always collected, </s><s>[INST] there's something a little bit sad which is becoming or maybe it's hopeful which is more and more common these days, which when a person passes away you have their twitter account. You know when you have the last tweet they tweeted like something that [/INST] you can recreate them now with large language models and so on. I mean you can create somebody that's just like them and can actually continue to uh communicate. </s><s>[INST] I think that's really exciting because I think in some sense like if I were to die today in some sense I would continue on if I continue tweeting I tweet therefore I am. [/INST] Yeah. Well I mean that's one of the advantages of a replicants that can recreate the communications of that, that person. </s><s>[INST] Do you hope? Do you think? Do you hope humans will become a multi planetary species? You've talked about the phases, the six epics and one of them is reaching out into the stars in part, [/INST] yes, but the kind of attempts for making now to go to other planetary objects doesn't excite me that much because it's not really advancing anything. It's </s><s>[INST] not efficient enough. [/INST] Yeah. We're also putting out other uh human beings which is a very inefficient way to explore these other objects. </s><s>[INST] What [/INST] I'm really talking about in the in the sixth epic universe, wakes up. It's where we can spread our super intelligence throughout the universe and that doesn't mean sending very soft squishy creatures like humans. </s><s>[INST] The universe wakes up. I [/INST] mean we we would send intelligence masses of nanobots, which can then go out and uh colonize these, these other parts of the universe. </s><s>[INST] Do you think there's intelligent alien civilizations out there that are bots might meet [/INST] my hunch is no, most people say yes, absolutely. I mean, and they'll cite that drake equation and I think in the singularity is near um I have two analyses of the drake equation, both with very reasonable assumptions and one gives you thousands of advanced civilizations in each galaxy and another one gives you one civilization and we know of one. </s><s>[INST] A lot of the [/INST] analyses are forgetting the exponential growth of of computation because we've gone from where the fastest way I could send a message to somebody was with a pony, which was what like a century and a half ago, </s><s>[INST] uh [/INST] to the advanced civilization we have today. And and and if you accept what I've said, go forward a few decades, you can have absolutely fantastic amount of civilization compared to a pony. And that's in a couple of 100 years. Yeah. </s><s>[INST] The speed and the scale of information transfer is just, is growing exponentially in a blink of an eye [/INST] now. Think about these other civilizations, they're gonna be spread out at cosmic times. So if something is like ahead of us or behind us, it could be ahead of us or behind us by maybe millions of years, which isn't that much, I mean, the world is billions of years old, 14 billion or something. So even 1000 years, if two or 300 years is enough to go from a pony to a fantastic amount of civilization, we would see that so of other civilizations that have occurred. Okay. Some might be behind us, but some might be ahead of us. If they're ahead of us, they're ahead of us by thousands, millions of years and they would be so far beyond us. They would be doing galaxy wide engineering, but we don't see anything doing galaxy wide engineering. </s><s>[INST] So either they don't exist or this very universe is a construction of an alien species. We're living inside a video game. [/INST] Well, that's another explanation that yes, you've got some teenage kids and another civilization. </s><s>[INST] Do you, do you find compelling? The simulation hypothesis as a thought experiment that we're living in a simulation [/INST] of the universe is computational, so we are an example in a computational world, therefore uh it is a simulation doesn't necessarily mean an experiment by some high school kid in another world, but it's nonetheless is taking place in a computational world and everything that's going on is is basically a form of, of computation. Um so you really have to define what you mean by uh this whole world being a simulation. </s><s>[INST] Well, then it's the, it's the teenager that that makes the video game, you know, us, humans with our current limited cognitive capability have strive to understand ourselves and we have created religions, we think of God, whatever that is, Do you think God exists? And if so, who is God? I [/INST] alluded to this before we started out with lots of particles going around and there's nothing that represents love and creativity. Um, and somehow we've gotten into a world where love actually exists and that has to do actually with consciousness because you can't have love without consciousness. So to me, that's God, the fact that we have something where love, where you can be devoted to someone else and and really feel that love, um, that's that's God. And if you look at the old testament, it was actually created by several different Robin? It's in there and then I think they've identified three of them, One of them dealt with God as a person that you can make deals with and he gets angry and he wrecks vengeance on various people, but two of them actually talk about God as a symbol of love and peace and harmony and and so forth. That's how they describe God. So that's my view of God, </s>

<s>[INST] I know how that goes. But she [/INST] tried but she said to him, but even the dog is allowed to eat the crumbs that fall from the master's table and he went, helped, helped. Now let's go back to just said about Putin or if you don't quote alone and myself and my family, of course the family is first, but once the family is good, it has to then spread to the community dan to the ST country world. The problem we have sometimes is that, and this is the reason why a lot of powerful families was overthrown. Like why do they had their own king with the guillotine? Because that once the family was strong, they didn't let the the wealth, the opportunity expand out. You know, look at the wu tang, Yes, our family was made strong first. But then all the woman was able to form their own corporations and they had their own sub families, it has to grow </s><s>[INST] out and they took over the world. You've talked about being vegan. I don't think I heard you explain this because it connects somehow about how you think about life. So you talk about when your family is good, you grow that like circle of empathy. You grow the community. Is that how you think about being vegan? That just the capacity of living beings on earth to suffer, that you just don't want to add suffering to them? [/INST] Yeah, I mean you said it clear, it's like nothing in all reality. I came to a realization and nothing really has to die for me to live. No animal, the plants themselves. Right? So this is say, you know, you want a steak, which is probably the most, you know, you know, I don't know the most expensive piece of me, but let's just say the steak is, you know, top of the line, nice steak and you're eating a steak for the protein to help build your muscle And I don't know if you got it from a cow or bull, but whether it's a cow or the ball, they grow to about £1,500 and it was a board, it's all mostly muscle and it's only eating grass. Uh huh. </s><s>[INST] Yeah. Yeah, there's there's yeah, it's possible to uh both as an athlete and just as a human being to perform well without meeting me. There's something, especially in the way we're treating animals to deliver that meat to the plate. I think about that a lot. So I I do I'm a robotics person ai person and I think a lot about, I don't know if you think about this kind of stuff but building A. I. Systems as they become more and more human. Like you start to ask the question of are we okay? Uh if we give the capacity for AI systems to suffer first to feel, but then to suffer um to hate and to love to feel emotion, how do we deal with that? It starts asking the same questions you ask of animals. Are we, are we okay adding that suffering to the world? [/INST] And I don't think we should have the suffering because it's not necessary? Like look if it's necessary, right? Because we're, you know, survival of the first law of nature of self preservation if you are in the desert and there's nothing else to eat but that lizard. Yeah. Okay. You gotta do what you gotta do. This has gotta go, you gotta go, you gotta do what you gotta do because at the end of the day, man, it's when they say man has Dominion over these things. His Dominion is almost like a caretaker how the way we do our Dominion, we dominated eat it cook it. Like who's the first guy that looked at the lobster. It was like I'm gonna eat this thing like it's supposed to, it's hard to eat it, you gotta go through a process to get that crab. Remember we see crabs when we was kids and I didn't know why I was always getting itchy throats all that. You know, you care, you don't know just eat. But at the end of the day, a crab didn't provide no more than a finger work for meat, mate. And it was hell, getting that, Getting it out. It's not worth it. In all reality, you could have gave me a, you could have gave me a banana and did better for my body and my appetite. Am I being fulfilled as full? Look, look at the blessings of of of of life, right? If you take a seed, oh, you get an apple and you eat it in that apple is multiple. Season it if you plant that seed, it will give you a whole tree with a whole bunch of apples with all multiple seeds. But if you kill a fish, it can't be produced. It's done. If you kill a cat is done, it's not, it's not coming back. But when you deal with the plants, even after you eat the apple and then you defecate, you're defecation is what feeds the ground, the cause of apple to grow </s><s>[INST] more. Yeah. Circle of life. And especially there's a there's a guy named David Foster Wallace, he wrote a short story called Consider the lobster. If you actually think philosophically about what from a perspective the lobster that's like symbolic or something because you're basically put in the water like cold water and then it heats up slowly until until it's no [/INST] more. Then they started eating lobsters in the acquisition. Yeah. </s><s>[INST] Just enjoy. So they were probably enjoyed torturing animals and they realized they're also delicious after the torture is finished. That's probably how they discovered it. [/INST] Let me ask you a question. I know you're asking me the questions but I want to talk a little bit about the ai and you said something about trying to put the emotion in it. Yeah. Right. Um So do we so are you thinking there's an algorithm for emotion? Yes. </s>

<s>[INST] prisoners dilemma. And so he needed to set up a little prisoner's dilemma grid. So he drew it and I could see what he was drawing and the prisoner's dilemma consists of two players to parties. So he decided he would make little cartoons of the two of us. And so there was uh two criminals right that we're deciding whether or not to rat each other out. Um One of them he drew as a circle with a smiley face and a kind of goatee thing, smooth head and the other one with all sorts of curly hair. And he said this is smooth and curly. I said smooth and curly. He said you know smooth with a V. It's very important that it has the and and then the students really that the students really took to that like they they found that relatable. [/INST] He started singing Smooth Criminal by Michael Jackson </s><s>[INST] and that those those names so that so we now have a video series that an episode are kind of first actual episode should be coming out today. Um Smooth and clearly on video where the two of us discuss uh episodes of Westworld, we watch Westworld and we're like what does this say about computer science [/INST] and we never we did not watch it. No, it's on season three or whatever we have as of this recording is on season three and now two episodes total. Think about three. What </s><s>[INST] Do you think about Westworld two episodes in so I can tell you so far. I'm just guessing what's going to happen next. It seems like bad things are going to happen with the robots uprising. [/INST] So I have not. I mean, you know, I vaguely remember a movie existing. So I assume it's related to that. But </s><s>[INST] that was more of my time than your time Charles. [/INST] That's right. Because you're much older than I think the important thing here is that uh, its narrative, right? It's all about telling a story. That's the whole driving thing. But the idea that they would give these referees that they would make people, they would remember remember the </s><s>[INST] awful things that happened happened. Who [/INST] could possibly think that was going? I got to I mean, I don't know. I've only seen the first two episodes or maybe the third one. I think I believe you know </s><s>[INST] what it was. You know what the problem is that the robots were actually designed by Hannibal Lecter. That's true. [/INST] So </s><s>[INST] like what do you think is going to happen? Bad things. [/INST] It's clear that things are happening and characters being introduced and we don't yet know anything. But still, I was just struck by how it's all driven by narrative and story. And there's all these implied things like programming. Half the programming interface is talking to them about what's going on in their heads, which is both. I mean artistically it's probably useful to film it that way, but think about how it would work in real life. It just seems very creative. But there was we saw in the second episode, there's a screen, you can see things they were wearing like state in the world. It was quite interesting to just kind of asked this question so far. I mean I assume it veers off into never never land at some point, but uh </s><s>[INST] I'm also a fan, a guy named Alex Garland, he's a director of ex Machina and he is the first I wonder if Kubrick was like this actually, is he like studies what would it take to program an ai systems? Like he's he's curious enough to go into that direction. On the West Wall side. I felt there was more emphasis on the narratives than like actually asking like computer science questions like how would you build this? How would you uh [/INST] how would you debug it? I still think to me that's the key issue if they were terrible to buggers. Yeah, </s><s>[INST] well they said specifically, so we make a change and we put it out in the world and that's bad because something terrible could happen. Like if [/INST] you're putting things out of the </s><s>[INST] world and you're not sure whether something terrible is going to happen, you're probably your process [/INST] is probably I just feel like there should have been someone whose sole job it was to walk around poke is headed and say what could possibly go wrong over and over again. </s><s>[INST] I would have loved if there was an I did watch a lot more and I'm not giving anything away. I would have loved it if there was like an episode where like, like the new intern is like debugging a new model or something and like it just keeps failing and they're like all right. And then more turns into like an episode of Silicon Valley or something like that versus like this ominous ai systems that are constantly like threatening the fabric of this world that's been created. Yeah. And you know, [/INST] the other, this reminds me of something that so I agree with that, that actually very cool. At least for the small percentage of people who care about the bugging systems. But the other thing is debugging the serie it falls into the thing of the Sequels. Fear of the book. Oh my gosh, anyway, so a </s><s>[INST] nightmare show. It's a [/INST] horror movie. I think that's where we lose people. By the way early on is the people who either decide either figure out debugging or think debugging is terrible. This is where we lose people in computer. This is part of the struggle versus suffering, right? You get through it and you kind of get the skills of it or you just like this is dumb and this is a dumb way to do anything. And I think that's when we lose people. But um uh well I'll leave it at that. But I think that I think that that there's something really, really neat about framing it that way. But what I don't like about all of these all of these things and I love text marketing by the way, the ending was very depressing. </s>

<s>[INST] Yeah, there's a couple. So, um Chomsky would have been one, of course, because he's, you know, he's probably the biggest standard because one of the biggest anarchist thinkers in contemporary times. Um, I was on the fence about Herbert Spencer because he's not an anarchist. Chris. Williamson is reading the chapter for the book. He coined the term survival of the fittest, and chapter is called the Right to ignore the state from his book, social statics. It was deleted from later editions, well founded and reprinted it. Um, and uh Randolph bourne. He was an early progressive, he was the only one where one of the very few fighting against entering the Great War. And he had an essay called War Is the Health of the State, which is basically about how states love war, because it gives them an excuse to increase their power. And it's very hard to argue against increasing state power in a time of war. But since he was not himself an anarchist and there was plenty anti war and they're already I didn't include him, but those would be the ones. [/INST] Is there some people that you think, uh, the public would be surprised to learn that they are at least in part anarchist. Like I saw that Howard Zinn is supposed to be an anarchist. I mean, is there, um, like, just like Tolstoy's anarchist? Is there some people like that that you think in our modern life that would be surprised to learn? They're anarchists? I </s><s>[INST] Can't think of any off the top of my head. I mean, you could say Karl Hess, who was like Barry Goldwater speech writer from the 1964 campaign, but he's hardly a household name. Um, I mean, I think a lot of people would not ascribed to that term, but it certainly informed with this complete distrust of all authority Murray rothbard had an essay If I didn't include Anatomy of the State, I was going to include this one. It's much much shorter. And his question was, who are allies and who our enemies? And the point he makes is there's lots of people who would call themselves anarchists who are of little use, whereas someone who is still like a monarchist or for government, but genuinely hates the question Rothbart had is if there's a button and you can compress that you would end the state, Would you press it so fast? Your finger to get a blister? Those are allies, even if they're, uh, somewhat of a monarchist. So I think that is a kind of a better lens of looking at it. And I don't think anyone needs to really ascribe to anarchism as a whole ideology insofar as you're seeing right now, many people in certain fringe elements are just essentially or, or decreasing the fringe and increasing mainstream elements are realizing that this idea that whatever the state does is somehow morally binding or legitimate is something that at least bear strong questioning. [/INST] Sure. And I mean, I guess there's a lot of groups like the libertarians for example, have some element of that. Sure, for sure, harsh questioning of the ways of government </s><s>[INST] and also, I think what I love, I mean, if there's one issue where I would want people to have this kind of analysis, it is war and it is like, okay, are you really sure? Because this is 100% going to result in a lot of people being killed, a lot of people being traumatized, a lot of people who are never going to recover Children, innocent people. Are you really sure this is the right thing to do? And I think a lot of times, if the answer is, well, it's the profitable thing to do, and that is, I think, again, government and its absolute most venal and worst [/INST] you, Michael malice in many ways are a new yorker. </s><s>[INST] Oh, yes, I'll give you one example, I don't know where Austin is on the map? No idea. Not even [/INST] kidding. But does it even </s><s>[INST] matter? It doesn't matter. [/INST] But nevertheless, you've decided to move to Austin? Why do you think you're moving to Austin or why do you moving both to Austin and away from new </s>

<s>[INST] so people who are not familiar, bird crashes a comedian who takes off his shirt often has, he's a elegant layer of fat around him, he's also a party animal, so he's a weird balance of, like, healthy and unhealthy, so he drinks a lot during, I think there's some debate about that, but certainly after his, his performances, but at the same time, he's into kind of the running thing, he does quite a bit of treadmill running. I think So, and like I said, has this challenge of running 2000 miles this year, so it's fascinating to have somebody who so fully embraces life and the full joys of life as represented by the huge amounts of drinking and partying and just being a wild man, but also at the same time, like being at least curious about this, challenging yourself in the physical realm. It's kind of fascinating and it reminds me of one of my favorite comedians, like Eddie Izzard who has been doing those challenges basically off the couch, just running a marathon a day kind of thing is fascinating to see the purity of those challenges when like, exercise hasn't necessarily been deeply great ingrained in your life and you kind of just embrace the challenge anyway and take it on. And that's another way of looking at, because we've been talking about running as a a performance like optimization thing where training is such a huge part of this process, like race day is just the cherry on top. But there's for some people where the race is the cake, it's like they just take it on as a pure challenge. As the, as the, as the thing you haven't really trained for is the thing you haven't, you don't understand the intricacies of, but you take it on anyway. And that, that reveals something about the human spirit as well. [/INST] Yeah. And there's definitely like a switch that flips when you in your mind decide I'm gonna do this where then all of a sudden it goes from, like you stop thinking about all that's not possible to like, I'm just gonna do it and I think bert highlights that perfectly in a lot of cases where like he's he's maybe not even thinking it through enough to get to the point where it's like, he gets the point where he thinks this is not possible where most people would look at it and think, huh? I don't know if I can actually physically accomplish that task. Birds, just like, oh yeah, I'm gonna do it. </s><s>[INST] And my my [/INST] Thing thought with Burt was the 2000 mile thing is that we're going to find him at the end of the year with like 36 hours to go on 100 miles and that's right, </s><s>[INST] that's right. That's what's going to happen. It's going to be hilarious. Uh speaking of things that are insane and like taking on challenges that don't seem like you didn't, you didn't think through uh you're thinking about running across the country in in a challenge. You called the transcontinental run. Can you describe this challenge and what the heck you're thinking? [/INST] Yeah. Yeah. So this is uh you know, one thing that is exciting about ultramarathons, I think in a lot of places, especially early in someone's ultramarathon adventure, if they decide to do that as a part of their life is you have like these early years where you're doing things for the first time and it's like so cool and scary at the same time to think Today, I'm gonna run 100 miles and for this ever run before. It's 50 or something like that and you just know you're going to do something that you've never done before, you can experience things you would have never been able to predict. Uh and it's like this really interesting, unique human experience I think. So for me, I spent most of my career at this point like doing, I got through that phase and a lot of the events I'm really interested in and then it was like now let's repeat it and see if I can do it better and you get to that mindset for a while which is also a fun mindset. But there is that kind of like uh um desire to kind of have that human experience again of like you know, not knowing what could happen or is this doable type of a thing but still doing it and figuring it out along the way. So I would describe the transcontinental project is something like that. It's not anything unique to me or anything new. There's been a lot of people have done it before, but essentially it's a route, there's different routes, there's one kind of main one that's done for like the, that is used as the record route more or less that you go from san Francisco to new york and essentially you live out of an RV while you're running. So you run as much as you can during the day when you go to bed at night and then you get up and do it again and you're, you're handling all the logistics and the process of trying to make sure you can get up the next day and do again what you did the day before, which is gonna be the biggest difference. So for me, I've done all single day ultramarathons where you're gonna wring yourself dry at knowing the next day or week or however long you need, you're gonna be able to just kind of like shut everything down 11 and catch back up. Whereas with this, like, you know, you're doing it again and again again and you know, the record is by a guy named Pete kostelic who averaged just over 72 miles a day, finished in 42 days, six hours and 30 minutes. And I mean just like 72 miles, 73 miles and then like next day again next day again, just knowing every day when you finish, you spent a whole day running and then, okay, I'm gonna go to bed, I'm gonna wake up in the morning, I'm gonna have to do this again and then, you know, have that happen for six weeks and that's if it goes very well. </s>

<s>[INST] You broke that riot line with an avocado, maybe. [/INST] That could have gone wrong in so many ways. But it didn't, I don't know. To me, it's like, there's small lessons there. There is a case to be made about social engineering, about learning about behavior, about learning how to lie, and how to kind of move your way or navigate your way around situations like that. Small things like bartering, knowing how to bribe people in conflict zones is the thing that I show when I talk about or train people to work in hostile environments. De-escalation, you know? Specifically kind of figuring out what is of value in the environment, what things you shouldn't be doing in an environment that might be considered disrespectful or out of place. People have a tendency that didn't grow up in places that are violent to make continuous eye contact with somebody that might be an issue. Or smiling, when there's nothing to smile about. I think, you know, there's a picture, I saw somewhere of Russians taking a portrait and there's Americans there and the Americans are smiling, but the Russians aren't. Yeah. Because what is there to smile about? </s><s>[INST] Which is true. And of course it's not as simple as smile or not smile. There's subtlety to it, like you said. Eye contact is a super interesting one because I found in my own life, like not making eye contact is, other people would be joking, but it's a really powerful way to de-escalate. And there's such a fascinating thing though, because you could talk about drunk fights that are just, that are harmless, but I feel like the same dynamic applies to the most violent conflict, including wars. I feel like ego is part of this. So to me, the question of conflict, whether it's a street fight or anything else, is the calculus of, are you willing to take an L in terms of psychology? Somebody grabs your wife's ass, you mentioned. Boy, if you let that happen, you go home, you're gonna have to pay the price of you were the person who didn't define you, like in your relationship, you didn't defend your wife's honor. Yeah. You're gonna psychologically pay that price yourself and depending on your wife, she might secretly also lose a little bit of respect for you. Yeah. Now, how you play that calculus, because now we see the war in Ukraine, I would say there is elements of similar posturing in the United States, in Europe, in Ukraine, Russia, China, leadership. At a macro level. At a geopolitics, it's still somebody grabs somebody's ass and you're not backing down. [/INST] So to take those losses and basically just posture, lower your head and live to fight another day type of situation. The thing with modern violence is the access to weaponry. I mean, again, nobody owns life, but anybody that can hold a frying pan can own death. I've seen people get double leg takedown, somebody on the ground. It's a different thing doing in the mats versus concrete. That's a good way to kill somebody. The most prolific impact weapon on the planet is the planet itself. You can see various videos of people online where they fall and they hit their head or somebody hits their head, and they go into the stretched out fit basically. And that might not kill you then, but it'll kill you that night or the second night if you don't get checked out. People bleed out internally, get an edema. Again, the whole aspect of me showing how some of these things, not only some of these methodologies and somehow people prepare for violence and how people experience violence, how they make their weapons, how the people fight in the streets and stuff like that. It's to recognize that behavior from the inception. There's a video I show where there's a bunch of street kids in Rio de Janeiro. I think it's during the Olympics where they're snatching chains and cell phones from people. And it's a fun video, see it. The first thing you learn about it is how they target people. Now, who are they going after? There's a bunch of people there. Why are they going after that specific person? And you start learning about profiling and how they identify victim mentality or the perfect victim, lack of awareness. They keep on a straight line avoidance, avoidance of eye contact if they're doing something nefarious or wrong and how they pick who they're gonna go after. The small people, the women, even some of the men, and they separate the men that they're perfect victims versus the men that it's gonna turn around and punch them in the face What are they looking for? Well, first off, you notice that the men that are in that environment that look at them and are aware of their presence, their hyperaware, or the ones that are not good to target. So that's the first lesson there. So it's probably a good idea, not only to be hyperaware, but to recognize that hyperawares in others. If I wanna separate myself from the victim crowd. Another thing you notice is these are kids going after some going adults. And some of these grown adult men are with women. And you see them, you know, kind of getting outside of the grasp of this kids that are trying to rip their chains off their neck or their cell phones and they have no consideration for the women around them. You see other men that are with women and you see them grab the women and put them behind them. And immediately they'll say this is the wrong one. Let me move off to the next one. So that small little lesson in those videos will show you first how these kids are growing up to profile and target who the perfect victims are. That's a school for them. And that is an adversarial school. </s>

<s>[INST] If somehow you like this podcast, please subscribe on Youtube. Review it with five stars on apple podcast. Follow on Spotify. Support on Patreon connect with me on twitter at lex Friedman. As usual, I'll do a few minutes of ads now and no ads in the middle. I try to make these interesting but I give you time stamps So if you skip, please still check out the sponsors by clicking the links in the description. It's the best way to support this podcast. This show is sponsored by Graham Early, a service I use in my writing to check spelling, grammar, sentence structure and readability. It's like a second pair of eyes that helps make sure that my writing sounds more like Ernest Hemingway's The Old Man and the Sea, one of my favorite books. [/INST] And less like </s><s>[INST] James Joyce's Finnegans Wake, which is a novel of attempted to read several times in college and failed miserably on every occasion. Graham early is doing a couple of tests, reads with this podcast and they will judge whether they want to sign a long term contract based on how many people sign up to Graham early premium, which I've been using and enjoying for many years now. So I think you know what to do, especially if you enjoy this podcast. I want your writing to not be a chaotic stream of consciousness like the precursor to the GPT three language model. That is the post ulysses writing of Mr James Joyce. Get 20% off graham early premium by signing up at graham early dot com slash lex. That's 20% off grandma lee premium at Graham early dot com slash flex. Now here's the tricky part. You have to sign up on a desktop, not your smartphone, otherwise it doesn't count as support for this podcast, so please use your desktop or laptop. I hope you'll be up for this challenge. This show is also sponsored by sun basket. Some basket delivers fresh, healthy delicious meals straight to your door. As you may know. My dad is pretty minimalist, so it's nice to get some healthy variety into the mix. They make it easy and convenient with everything pre portioned and ready to prep and cook. You can enjoy delicious healthy dinner in as little as 15 minutes. I've enjoyed a lot of meals from their menus that fit my, what they call carb conscious ways the items on the menu are always delicious. Like today it's black angus rib eye steak with broccoli and radishes, italian sausages and vegetable skewers with two romesco sauce, mediterranean lemon chicken with baby broccoli, artichokes and olives. What the heck is baby broccoli, I don't know right now some basket is offering 35 bucks off your order when you go right now to some basket dot com slash lex and enter promo code lex at checkout again, visit some basket dot com slash lex and use code lex to get $35 off your order. This show is sponsored by ECM Rush, which if you look around seems to be one of if not the most respected digital marketing tool out there. It does a lot of stuff including S SEO optimization of keywords, back links, content creation, social media posts and so on. They have over 45 tools and a trusted by over six million marketers worldwide. I don't like numbers but that's because I'm an idiot with that stuff and in general I speak from the heart and data be damned. But somebody needs to pay attention to numbers because otherwise you can't make optimal decisions. I believe heart comes first data second, but both are necessary. I started using them just for fun to explore. Not numeric things like what kind of titles or words connect with people as a writer and crappy part time speaker. That information helps me at least the moderation. The amount of data that they put at your fingertips is amazing. So if you want to optimize your online presidents, check them out at SCM rush dot com slash partner slash lex to get a free month of guru level membership. Okay, this show is also sponsored by Express VPN. It provides privacy in your digital life without a VPN. Your I. S. B. Internet service provider can see every sight you've ever visited as terrifying. Even when you're browsing in incognito mode. Even if you clear your history in the US, they can legally sell your data to add companies. Express VPN prevents them from being able to do all that by rerouting and encrypting your internet connection to their secure servers. I've used it for many years on Windows Lennox and android, but it's available everywhere else too. For me it's been fast and easy to use one big power on button. That's fun to press. Probably my favorite intuitive design of an app doesn't try to do more than it needs to. I believe there's a Charles bukowski quote about that go to Express VPN dot com slash flex pod to get an extra three months free on a one year package that's Express VPN dot com slash lex pod and now here's my conversation with eric Weinstein who's the greatest musician of all time. Would you say? We were just off camera talking about Eddie van Halen. He unfortunately passed away. [/INST] Who's the greatest musician of all time. Yeah, Jonathan Richmond, who's that? It's a weird question. So I'm gonna give you a weird answer. It's not because thank you. Okay, Jonathan Richard. The reason I'm picking on him is that he had a quote. Uh he was the front man of a group called the modern Lovers and his quote was something like we have to be prepared to play music when our instruments are broken, the electricity is out and it's raining something like that. And I thought that that quote was very interesting because what it said was, you have to be able to strip this thing down farther and farther back to get to something that is intrinsically musical. So we were having a conversation just now about virtuosity and we're talking about Eddie Van Halen and his recent passing and that affected me emotionally. I don't know whether it affected you. I was never Van Halen, the group fan, but I revered Eddie van Halen's capacity for innovation. Just I saw him like uh you know Rodney mullen the skateboard, I had dreamed of having the two of them on the same podcast just to talk about what it's like to totally discontinuous lee innovate. </s>

<s>[INST] What are the key differences from the high philosophical level, maybe the technical level between the capabilities of broadly speaking the two kinds of systems, large language models and this computational, gigantic computational system infrastructure that is Wolfram Alpha? [/INST] Yeah. So what does something like Chad GPT do? It's mostly focused on make language like the language that humans have made and put on the web and so on. So it's primary underlying technical thing is you've given a prompt. It's trying to continue that prompt in a way that's somehow typical of what it's seen based on a trillion words of text that humans have written on the web. And the way it's doing that is with something which is probably quite similar to the way we humans do the first stages of that using a neural net and so on. And just saying, given this piece of text, let's ripple through the neural net one word and get one word at a time of output. And it's kind of a shallow computation on a large amount of kind of training data that is what we humans have put on the web. That's a different thing from sort of the computational stack that I spent the last, I don't know, 40 years or so building, which has to do with what can you compute many steps, potentially a very deep computation. It's not sort of taking the statistics of what we humans have produced and trying to continue things based on that statistics. Instead, it's trying to take kind of the formal structure that we've created in our civilization, whether it's from mathematics or whether it's from kind of systematic knowledge of all kinds, and use that to do arbitrarily deep computations to figure out things that aren't just let's match what's already been kind of said on the web, but let's potentially be able to compute something new and different that's never been computed before. So as a practical matter, our goal is to have made as much as possible of the world computable in the sense that if there's a question that in principle is answerable from some sort of expert knowledge that's been accumulated, we can compute the answer to that question and we can do it in a sort of reliable way that's the best one can do given what the expertise that our civilization has accumulated. It's a much more sort of labor intensive on the side of kind of creating kind of the computational system to do that. Obviously, in the kind of the chat GPT world, it's like take things which were produced for quite other purposes, namely all the things we've written out on the web and so on, and sort of forage from that things which are like what's been written on the web. So I think as a practical point of view, I view sort of the chat GPT thing as being wide and shallow, and what we're trying to do with sort of building out computation as being this sort of deep, also broad, but most importantly kind of deep type of thing. I think another way to think about this is you go back in human history, I don't know, a thousand years or something, and you say, what's the typical person going to figure out? Well, the answer is there's certain kinds of things that we humans can quickly figure out. That's sort of what our neural architecture and the kinds of things we learn in our lives let us do. But then there's this whole layer of kind of formalization that got developed, which is the kind of whole sort of story of intellectual history and the whole kind of depth of learning. That formalization turned into things like logic, mathematics, science, and so on. And that's the kind of thing that allows one to kind of build these towers of sort of towers of things you work out. It's not just, I can immediately figure this out. It's no, I can use this kind of formalism to go step by step and work out something, which was not immediately obvious to me. And that's kind of the story of what we're trying to do computationally is to be able to build those kind of tall towers of what implies what, implies what, and so on, and as opposed to kind of the, yes, I can immediately figure it out. It's just like what I saw somewhere else in something that I heard or remembered or something </s><s>[INST] like this. What can you say about the kind of formal structure, the kind of formal foundation you can build such a formal structure on, about the kinds of things you would start on in [/INST] order to build this kind of deep computable knowledge trees? So the question is sort of how do you think about computation? And there's a couple of points here. One is what computation intrinsically is like, and the other is what aspects of computation we humans with our minds and with the kinds of things we've learned can sort of relate to in that computational universe. So if we start on the kind of what can computation be like, it's something I've spent some big chunk of my life studying is, imagine that you're, you know, we usually write programs where we kind of know what we want the program to do and we carefully write many lines of code and we hope that the program does what we intended it to do. But the thing I've been interested in is if you just look at the kind of natural science of programs, you just say, I'm going to make this program, it's a really tiny program. Maybe I even pick the pieces of the program at random, but it's really tiny. By really tiny, I mean, you know, less than a line of code type thing. You say, what does this program do? And you run it. And big discovery that I made in the early eighties is that even extremely simple programs, when you run them, can do really complicated things. Really surprised me. It took me several years to kind of realize that that was a thing, so to speak. But that realization that even very simple programs can do incredibly complicated things that we very much don't expect that discovery. I mean, I realize that that's very much, I think, how nature works. That is, nature has simple rules but yet does all sorts of complicated things that we might not expect. You know, a big thing of the last few years has been understanding that that's how the whole universe and physics works but that's a quite separate topic. But so there's this whole world of programs and what they do and very rich, sophisticated things that these programs can do, but when we look at many of these programs, we look at them. kind of, I don't really know what that's doing, it's not a very human, kind of thing. So on the one hand, we have sort of what's possible in the computational universe. On the other hand, we have the kinds of things that we humans think about, the kinds of things that are developed in, kind of, our intellectual history. And that's, and really, the challenge to sort of making things computational is to connect what's computationally possible out in the computational universe with the things that we humans, sort of, typically think about with our minds. Now, that's a complicated kind of moving target because the things that we think about change over time. We've learned more stuff, we've invented mathematics, we've invented various kinds of ideas and structures and so on. So it's gradually expanding, we're kind of gradually colonizing more and more of this kind of intellectual space of possibilities. But the real thing, the real challenge is how do you take what is computationally possible, how do you take, how do you encapsulate the kinds of things that we think about in a way that kind of plugs into what's computationally possible? And actually, the big sort of idea there is this idea of kind of symbolic programming, symbolic representations of things. And so the question is, when you look at sort of everything in the world, and you take some visual scene or something you're looking at, and you say, well, how do I turn that into something where I can kind of stuff into my mind? mind." There are lots of pixels in my visual scene, but the things that I remembered from that visual scene are there's a chair in this place. It's a symbolic representation of the visual scene. There are two chairs and a table or something, rather than there are all these pixels arranged in all these detailed ways. The question then is, how do you take all the things in the world and make some kind of representation that corresponds to the types of ways that we think about things? human language is sort of one form of representation that we have. We talk about chairs. That's a word in human language and so on. How do we take… But human language is not in and of itself something that plugs in very well to sort of computation. It's not something from which you can immediately compute consequences and so on. And so you have to kind of find a way to take sort of the stuff we understand from human language and make it more precise. And that's really the story of symbolic programming. And what that turns into is something which I didn't know at the time it was going to work as well as it has. But back in 1979 or so, I was trying to build my first big computer system and trying to figure out how should I represent computations at a high level. And I kind of invented this idea of using kind of symbolic expressions structured as it's kind of like a function and a bunch of arguments. But that function doesn't necessarily evaluate to anything. It's just a thing that sits there representing a structure. And so building up that structure and it's turned out that structure has been extremely… It's a good match for the way that we humans… It seems to be a good match for the way that we humans kind of conceptualize higher level things. And it's been for the last, I don't know, 45 years or something. It's served me remarkably well. </s><s>[INST] So building up that structure using this kind of symbolic representation. But what can you say about abstractions here? Because you could just start with your physics project. You could start at a hypergraph at a very, very low level and build up everything from there, but you don't. You take shortcuts. You take the highest level of abstraction, convert the kind of abstraction that's convertible to something computable using symbolic representation. And then that's your new foundation for that little piece of knowledge and somehow all of that is integrated. [/INST] Right. So this sort of a very important phenomenon that is kind of a thing that I've sort of realized is just it's one of these things that sort of in the future of kind of everything is going to become more and more important as this phenomenon of computational irreducibility. And the question is, if you know the rules for something, you have a program, you're going to run it, you might say, I know the rules. Great. I know everything about what's going to happen. Well, in principle, you do because you can just run those rules out and just see what they do. You might run them a million steps, you see what happens, et cetera. The question is, can you immediately jump ahead and say, I know what's going to happen after a million steps and the answer is 13 or something. And one of the very critical things to realize is if you could reduce that computation, there isn't a sense, no point in doing the computation. The place where you really get value out of doing computation is when you had to do the computation to find out the answer. But this phenomenon that you have to do the computation to find out the answer, this phenomenon of computational irreducibility, seems to be tremendously important for thinking about lots of kinds of things. So one of the things that happens is, okay, you've got a model of the universe at the low level in terms of atoms of space and hypergraphs and rewriting hypergraphs and so on. And it's happening 10 to the 100 times every second, let's say. Well, you say, great, then we've nailed it. We know how the universe works. Well, the problem is the universe can figure out what it's going to do. It does those 10 to the 100 steps but for us to work out what it's going to do, we have no way to reduce that computation. The only way to do the computation, to see the result of the computation, is to do it. And if we're operating within the universe, there's no opportunity to do that because the universe is doing it as fast as the universe can do it and that's what's happening. So what we're trying to do, and a lot of the story of science and a lot of other kinds of things, is finding pockets of reducibility. That is, you could have a situation where everything in the world is full of computational reducibility. We never know what's going to happen next. The only way we can figure out what's going to happen next is just let the system run and see what happens. So in a sense, the story of most kinds of science, inventions, a lot of kinds of things, is the story of finding these places where we can locally jump ahead. And one of the features of computational reducibility is, there are always pockets of reducibility. There are always an infinite number of places where you can jump ahead. There's no way where you can jump completely ahead, but there are little patches, little places where you can jump ahead a bit, and I think we can talk about physics project and so on, but I think the thing we realize is, we kind of exist in a slice of all the possible computational irreducibility in the universe. We exist in a slice where there's a reasonable amount of predictability. And in a sense, as we try and construct these kind of higher levels of abstraction, symbolic representations, and so on, what we're doing is we're finding these lumps of reducibility that we can kind of attach ourselves to, and about which we can kind of have fairly simple narrative things to say. Because in principle, you know, I say, what's going to happen in the next few seconds? You know, oh, there are these molecules moving around in the air in this room. And oh, gosh, it's an incredibly complicated story. And that's a whole computational irreducible thing, most of which I don't care about. And most of it is, well, you know, the air is still going to be here, and nothing much is going to be different about it. And that's a kind of reducible fact about what is ultimately at an underlying level of computational irreducible </s><s>[INST] process. Yes. And life would not be possible if we didn't have a large number of such reducible pockets. Yes. Pockets amenable to reduction into something symbolic. [/INST] Yes, I think so. I mean, life in the way that we experience it, that, I mean, you know, one might, you know, depending on what we mean by life, so to speak, the experience that we have of sort of consistent things happening in the world, the idea of space, for example, where there's, you know, we can just say, you're here, you move there. It's kind of the same thing. It's still you in that different place, even though you're made of different atoms of space and so on. This is this idea that it's that there's sort of this level of predictability of what's going on. That's us finding a slice of reducibility in what is underneath this computationally irreducible kind of system. And I think that's sort of the thing, which is actually my favorite discovery of the last few years is the realization that it is sort of the interaction between this sort of underlying computational irreducibility and our nature as kind of observers who sort of have to key into computational reducibility. That fact leads to the main laws of physics that we discovered in the 20th century. So this is, we talked about this in more detail, but this is, to me, it's kind of our nature as observers, the fact that we are computationally bounded observers, we don't get to follow all those little pieces of computational irreducibility to stuff what is out there in the world into our minds requires that we are looking at things that are reducible, we are compressing, kind of we're extracting just some essence, some kind of symbolic essence of what's the detail of what's going on in the world, that together with one other condition that at first seems sort of trivial, but isn't, which is that we believe we are persistent in time. That is, you know, so that's the causality. Here's the thing, at every moment, according to our theory, we're made of different atoms of space. At every moment, sort of the microscopic detail of what the universe is made of is being rewritten. And that's and in fact, the very fact that there's coherence between different parts of space is a consequence of the fact that there are all these little processes going on, the kind of knit together, the structure of space. It's kind of like if you wanted to have a fluid with a bunch of molecules in it, if those molecules weren't interacting, you wouldn't have this fluid that would pour and do all these kinds of things. It would just be sort of a free floating collection of molecules. So similarly it is with space that the fact that space is kind of knitted together as a consequence of all this activity in space. And the fact that kind of what we consist of, sort of this series of, you know, we're continually being rewritten. And the question is, why is it the case that we think of ourselves as being the same us through time? That's kind of a key assumption. I think it's a key aspect of what we see as sort of our consciousness, so to speak, is that we have this kind of consistent thread of experience. </s><s>[INST] What is some sort of causality? Isn't that just another limitation of our mind that we want to reduce reality into some, that kind of temporal consistency is just a nice narrative to tell [/INST] ourselves? Well, the fact is, I think it's critical to the way we humans typically operate is that we have this single thread of experience. You know, if you imagine a mind where you have, you know, maybe that's what's happening in various kinds of minds that aren't working the same way other minds work is that you're splitting into multiple threads of experience. It's also something where, when you look at, I don't know, quantum mechanics, for example, in the insides of quantum mechanics, it's splitting into many threads of experience. But in order for us humans to interact with it, you have to have to knit all those different threads together so that we say, oh yeah, a definite thing happened, and now the next definite thing happens, and so on. And I think inside, it's interesting to try and imagine what's it like to have these fundamentally multiple threads of experience going on. I mean, right now, different human minds have different threads of experience. We just have a bunch of minds that are interacting with each other, but we don't have a ... Within each mind, there's a single thread, and that is indeed a simplification. I think it's a thing, the general computational system does not have that simplification. And it's one of the things people often seem to think that consciousness is the highest level of things that can happen in the universe, so to speak. But I think that's not true. I think it's actually a specialization in which, among other things, you have this idea of a single thread of experience, which is not a general feature of anything that could kind of computationally happen in the universe. </s><s>[INST] So it's a feature of a computationally limited system that's only able to observe reducible pockets. So I mean, this word observer, it means something in quantum mechanics, it means something in a lot of places. It means something to us humans as conscious beings. So what's the importance of the observer? What is the observer? What's the importance of the [/INST] observer in the computational universe? Yeah. So this question of what is an observer, what's the general idea of an observer, that's actually one of my next projects, which got somewhat derailed by the current sort of AI mania. </s><s>[INST] But- Is there a connection there? Or do you think the observer is primarily a physics phenomena? [/INST] Is it related to the whole AI thing? Yes, it is. What is a general observer? We know we have an idea what is a general computational system, we think about Turing machines, we think about other models of computation. There's a question, what is a general model of an observer. There's kind of observers like us, which is kind of the observers we're interested in. We could imagine an alien observer that deals with computational irreducibility and it a mind that's utterly different from ours and completely incoherent with what we're like. But the fact is that if we are talking about observers like us, one of the key things is this idea of taking all the detail of the world and being able to stuff it into a mind, being able to take all the detail and extract out of it a smaller set of degrees of freedom, a smaller number of elements that will sort of fit in our minds. And I think this question, so I've been interested in trying to characterize what is the general observer. And the general observer is, I think, in part, there are many ... let me give an example of a ... you have a gas, it's got a bunch of molecules bouncing around, and the thing you're measuring about the gas is its pressure. And the only thing you as an observer care about is pressure. And that means you have a piston on the side of this box and the piston is being pushed by the gas and there are many, many different ways that molecules can hit that piston, but all that matters is the kind of aggregate of all those molecular impacts because that's what determines pressure. So there's a huge number of different configurations of the gas, which are all equivalent. So I think one key aspect of observers is this equivalencing of many different configurations of a system, saying, all I care about is this aggregate feature, all I care about is this overall thing, and that's sort of one aspect. We see that in lots of different, again, it's the same story over and over again, that there's a lot of detail in the world, but what we are extracting from it is something, a sort of a thin summary of that detail. </s><s>[INST] Is that thin summary nevertheless true? Can it be a crappy approximation that on average is correct? I mean, if we look at the observer that's the human mind, it seems like there's a lot of very, as represented by natural language, for example, there's a lot of really crappy approximation. Sure. [/INST] And that could be maybe a feature of it, that there's ambiguity. You don't know, it could be the case. They're just measuring the aggregate impacts of these molecules. But there is some tiny, tiny probability that molecules will arrange themselves in some really funky way and that just measuring that average isn't going to the main point. By the way, an awful lot of science is very confused about this because you look at papers and people are really keen, they draw this curve and they have these bars on the curve and things, and it's just this curve. And it's this one thing and it's supposed to represent some system that has all kinds of details in it. And this is a way that lots of science has gotten wrong. Because people say, I remember years ago I was studying snowflake growth. You know, you have a snowflake and it's growing, it has all these arms, it's doing complicated things. But there was a literature on this stuff and it talked about, you know, what's the rate of snowflake growth? And, you know, it got pretty good answers for the rate of the growth of the snowflake. And I looked at it more carefully, and they had these nice curves of snowflake growth rates and so on. I looked at it more carefully, and I realized according to their models, the snowflake would be spherical. And so they got the growth rate right, but the detail was just utterly wrong. Not only the detail, the whole thing was not capturing, it was capturing this aspect of the system that was, in a sense, missing the main point of what was going on. What is the geometric shape of a snowflake? Snowflakes start in the phase of water that's relevant to the formation of snowflakes. It's a phase of ice, which starts with a hexagonal arrangement of water molecules. And so it starts off growing as a hexagonal plate. And then what happens is... It's a plate. Oh, oh, versus sphere versus plate. Well, no, no, but it's much more than that. I mean, snowflakes are fluffy. Typical snowflakes have little dendritic arms. And what actually happens is it's kind of cool because you can make these very simple discrete models with cellular automata and things that figure this out. You start off with this hexagonal thing. And then the place is it starts to grow little arms. And every time a little piece of ice adds itself to the snowflake, the fact that that ice condensed from the water vapor heats the snowflake up locally. And so it makes it less likely for another piece of ice to accumulate right nearby. So this leads to a kind of growth inhibition. So you grow an arm and it is a separated arm because right around the arm it got a little bit hot and it didn't add more ice there. So what happens is you have a hexagon, it grows out arms. The arms grow arms and then the arms grow arms go arms. And eventually, actually it's kind of cool because it actually fills in another hexagon a bigger hexagon. And when I first looked at this, you know, I had a very simple model for this. I realized, you know, when it fills in that hexagon, it actually leaves some holes behind. So I thought, well, you know, is that really right? So I look at these pictures of snowflakes, and sure enough, they have these little holes in them that are kind of scars of the way that these arms grow out. So you can't fill in back fill holes. So they just keep going up. </s><s>[INST] Yeah, they don't back fill. And presumably there's a limitation on how big, like, you can't [/INST] arbitrarily grow. I'm not sure. I'm mean the thing falls through the- I mean I think it this you know it hits the ground at some point. I think you can grow in the lab I think you can go to pretty big ones. I think you grow many many iterations of this kind of goes from hexagon it grows out arms, it turns back it fills back </s><s>[INST] into a hexagon, it grows more arms again. In 3d? No it's flat usually. Boy why is it flat? Why doesn't it span out? Okay okay wait a minute. You said it's fluffy Fluffy is a three-dimensional property, no? [/INST] No. It's fluffy. Snow is... Okay, so, you know, what makes... We're really in it. I like this. Let's go there. I like this. </s><s>[INST] Let's go there. There's multiple snowflakes become fluffy. Single snowflake is not fluffy. [/INST] No, no. Single snowflake is fluffy. And what happens is, you know, if you have snow that is just pure hexagons, they can, you know, they fit together pretty well. It's not... It doesn't make... It doesn't have a lot of air in it and they can also slide against each other pretty easily. And so, the snow can be pretty, you know, can, I think, avalanches happen sometimes when the things tend to be these, you know, hexagonal plates and it kind of slides. But then, when the thing has all these arms that have grown out, it's not... They don't fit together very well and that's why the snow has lots of air in it. And if you look at one of these snowflakes, and if you catch one, you'll see it has these little arms and people... Actually, people often say, you know, no two snowflakes are alike. That's mostly because as a snowflake grows, they do grow pretty consistently with these different arms and so on. But you capture them at different times. As they, you know, they fell through the air in a different way. You'll catch this one at this stage. And as it goes through different stages, they look really different. And so that's why, you know, it kind of looks like no two snowflakes are alike, because you caught them at different, at different times. </s><s>[INST] So, the rules under which they grow are the same. It's just the timing. Okay, so the point is, science is not able to describe the full complexity of snowflake growth. [/INST] Well, science, if you do what people might often do, which is say, okay, let's make it scientific, let's turn it into one number. And that one number is kind of the growth rate of the arms or some such other thing. That fails to capture sort of the detail of what's going on inside the system. And that's in a sense, a big challenge for science is how do you extract from the natural world, for example, those aspects of it that you are interested in talking about? Now, you might just say, I don't really care about the fluffiness of the snowflakes. All I care about is the growth rate of the arms, in which case you can have a good model without knowing anything about the fluffiness. But the fact is, as a practical, if you say, what is the most obvious feature of a snowflake? Oh, that it has this complicated shape. Well, then you've got a different story about what you model. I mean, this is one of the features of sort of modeling in science. What is a model? A model is some way of reducing the actuality of the world to something where you can readily sort of give a narrative for what's happening, where you can basically make some kind of abstraction of what's happening and answer questions that you care about answering. If you wanted to answer all possible questions about the system, you'd have to have the whole system because you might care about this particular molecule. Where did it go? And your model, which is some big abstraction of that, has nothing to say about that. So one of the things that's often confusing in science is people will have, I've got a model, somebody says. Somebody else will say, I don't believe in your model because it doesn't capture the feature of the system that I care about. There's always this controversy about, is it a correct model? Well, no model is except for the actual system itself, is a correct model in the sense that it captures everything, questions does it capture what you care about capturing. Sometimes that's ultimately defined by what you're going to build technology out of, things like this. The one counterexample to this is if you think you're modeling the whole universe all the way down, then there is a notion of a correct model. But even that is more complicated because it depends on kind of how observers sample things and so on. That's a separate story. But at least at the first level to say this thing about, oh, it's an approximation, you're capturing one aspect, you're not capturing other aspects. When you really think you have a complete model for the whole universe, you better be capturing ultimately everything, even though to actually run that model is impossible because of computational irreducibility. The only thing that successfully runs that model is the actual running of the universe. </s><s>[INST] Okay, so what you care about is an interesting concept. That's a human concept. That's what you're doing with Wolfram Alpha and Wolfram Language. Is he trying to come up with symbolic representations as simple as possible? A model that's as simple as possible that fully captures stuff we care about? [/INST] Yes. So I mean, for example, we could, we'll have a data about movies, let's say. We could be describing every individual pixel in every movie and so on, but that's not the level that people care about. And that level that people care about is somewhat related to what's described in natural language. But what we're trying to do is to find a way to sort of represent precisely so you can compute things. See, one thing we say, you give a piece of natural language, the question is you feed it to a computer, you say, does the computer understand this natural language? Well, you know, the computer processes it in some way, it does this, maybe it can make a continuation of the natural language. Maybe it can go on from the prompt and say what it's going to say. You say, does it really understand it? Hard to know. But in this kind of computational world, world, there is a very definite definition of does it understand, which is could it be turned into this symbolic computational thing from which you can compute all kinds of consequences? And that's the sense in which one has sort of a target for the understanding of natural language. And that's kind of our goal, is to have as much as possible about the world that can be computed in a reasonable way, so to speak, be able to be sort of captured by this kind of computational language, that's kind of the goal. And I think for us humans, the main thing that's important is as we formalize what we're talking about, it gives us a way of kind of building a structure where we can sort of build this tower of consequences of things. So if we're just saying, well, let's talk about it in natural language, it doesn't really give us some hard foundation that lets us build step-by-step to work something out. I mean, it's kind of like what happens in math. If we were just sort of vaguely talking about math, but didn't have the kind of full structure of math and all that kind of thing, we wouldn't be able to build this kind of big tower of consequences. And so in a sense, what we're trying to do with the whole computational language effort is to make a formalism for describing the world that makes it possible to kind of build </s><s>[INST] this tower of consequences. Well, can you talk about this dance between natural language and Wolfram language? So there's this gigantic thing we call the internet where people post memes and diary type thoughts and very important sounding articles and all of that that makes up the training dataset for GPT. And then there's Wolfram language. How can you map from the natural language of the internet to the Wolfram language? [/INST] Is there a manual, is there an automated way of doing that as we look into the future? Well, so Wolfram Alpha, what it does, it's kind of front end is turning natural language </s><s>[INST] into computational language. What you mean by that is there's a prompt, you ask a question, what is the capital of some country? [/INST] Yeah, right. And it turns into, what's the distance between Chicago and London or something? And that will turn into geodistance of entity, city, et cetera, et cetera, et cetera. Each one of those things is very well defined. We know, given that it's the entity, city, Chicago, et cetera, et cetera, et cetera, Illinois, United States, we know the geolocation of that, we know its population, we know all kinds of things about it, which we have curated that data to be able to know that with some degree of certainty, so to speak. </s><s>[INST] And then we can compute things from this and that's the idea. But then something like GPT, large language models, do they allow you to make that conversion [/INST] much more powerful? Okay, so it's an interesting thing, which we still don't know everything about. This question of going from natural language to computational language, in Wolfram Alpha, we've now, Wolfram Alpha's been out and about for 13 and a half years now, and we've achieved, I don't know what it is, 98%, 99% success on queries that get put into it. Now, obviously, there's a sort of feedback loop because the things that work are things people go on putting into it. But we've got to a very high success rate of the little fragments of natural language that people put in, questions, math calculations, chemistry calculations, whatever it is. We do very well at that, turning those things into computational language. Now, from the very beginning of Wolfram Alpha, I thought about, for example, writing code with natural language. In fact, I was just looking at this recently, I had a post that I wrote in 2010, 2011 called something like, programming with natural language is actually going to work. And so we had done a bunch of experiments using methods that were a little bit, some of them a little bit machine learning-like, but certainly not the same kind of idea of vast training data and so on, that is the story of large language models. Actually, I know that that post, a piece of utter trivia, but that post, Steve Jobs forwarded that post around to all kinds of people at Apple, because he never really liked programming languages. So he was very happy to see the idea that you could get rid of this kind of layer of kind of engineering-like structure. He would have liked, I think, what's happening now, because it really is the case that you can, this idea that you have to kind of learn how the computer works to use a programming language is something that is, I think, a thing that, just like you had to learn the details of the opcodes to know how assembly language worked and so on. It's kind of a thing that's a limited time horizon. So this idea of how elaborate can you make kind of the prompt, how elaborate can you make the natural language, and abstract from it computational language? It's a very interesting question, and what chat GPT 4 and so on can do is pretty good. It's a very interesting process. I mean, I'm still trying to understand this workflow. We've been working out a lot of tooling around this workflow. </s><s>[INST] Natural language to computational language, and the process, especially if it's conversation [/INST] like dialogue, multiple queries kind of thing. Yeah, right. There's so many things that are really interesting that work and so on. So first thing is, can you just walk up to the computer and expect to sort of specify a computation? What one realizes is humans have to have some idea of kind of this way of thinking about things computationally. Without that, you're kind of out of luck, because you just have no idea what you're going to walk up to a computer. I remember when I should tell a silly story about myself. The very first computer I saw, which was when I was 10 years old, and it was a big mainframe computer and so on, and I didn't really understand what computers did. It's like, somebody's showing me this computer, and it's like, can the computer work out the weight of a dinosaur? That isn't a sensible thing to ask. You have to give it. That's not what computers do. I mean, in Wolfram Alpha, for example, you could say, what's the typical weight of a Stegosaurus? It will give you some answer, but that's a very different kind of thing from what one thinks of computers as doing. The question of, first thing is, people have to have an idea of what computation is about. I think for education, that is the key thing, is kind of this notion, not computer science, not the details of programming, but just this idea of how do you think about the world computationally. To put, you know, thinking about the world computationally is kind of this formal way of thinking about the world. We've had other ones, like logic was a formal way, you know, is a way of sort of abstracting and formalizing some aspects of the world. Mathematics is another one. Computation is this very broad way of sort of formalizing the way we think about the world. The thing that's cool about computation is, if we can successfully formalize things in terms of computation, computers can help us figure out what the consequences are. It's not like you formalized it with math. Well, that's nice, but now you have to, if you're not using a computer to do the math, you have to go work out a bunch of stuff yourself. So I think, but this idea, let's see, we're talking about natural language and its relationship to computational language. The typical workflow, I think, is first, human has to have some kind of idea of what they're trying to do, that if it's something that they want to build a tower of capabilities on, something that they want to formalize and make computational. So then human can type something in to some LLM system and say vaguely what they want in computational terms. Then it does pretty well at synthesizing welcome language code, and it'll probably do better in the future because we've got a huge number of examples of natural language input together with the Wolfram Language Translation of that. So that's a thing where extrapolating from all those examples makes it easier to do that </s><s>[INST] task. Is the prompter task also debugging the Wolfram Language Code, or is your hope to not do that [/INST] debugging? No, no, no. So there are many steps here. The first thing is you type natural language. It generates Wolfram Language Code. </s><s>[INST] Do you have examples, by the way? Do you have an example, is it the dinosaur example, do you have an example that jumps [/INST] to mind that we should be thinking about some dumb example? It's like take my heart rate data and figure out whether I make a moving average every seven days or something and make a plot of the result. So that's a thing which is about two-thirds of a line of Wolfram Language Code. I mean, it's a list plot of moving average of some data bin or something of the data, and then you'll get the result. The vague thing that I was just saying in natural language could, would almost certainly correctly turn into that very simple piece of Wolfram Language Code. </s><s>[INST] So you start mumbling about heart rate. [/INST] You arrive at the moving average kind of idea. You say average over seven days. Maybe it'll figure out that that can be encapsulated as this moving average idea, I'm not sure. But then the typical workflow that I'm seeing is you generate this piece of Wolfram Language Code. It's pretty small usually. If it isn't small, it probably isn't right. If it's pretty small and Wolfram Language is one of the ideas of Wolfram Language, is that it's a language that humans can read. It's not a language which programming languages tend to be this one-way story of humans write them and computers execute from them. Wolfram Language is intended to be something which is sort of like math notation, something where humans write it and humans are supposed to read it, as well. And so kind of the workflow that's emerging is kind of this human mumble, some things. Large language model produces a fragment of often language code. Then you look at that, you say, well, typically you just run it first. You see, does it produce the right thing? You look at what it produces, you might say that's obviously crazy. You look at the code, you see, I see why it's crazy. You fix it. If you really care about the result and you really want to make sure it's right, you better look at that code and understand it because that's the way you have this sort of checkpoint of did it really do what I expected it to do. Now you go beyond that, I mean, what we find is, for example, let's say the code does the wrong thing. Then you can often say to the large language model, can you adjust this to do this? </s><s>[INST] And it's pretty good at doing that. Interesting. So you're using the output of the code to give you hints about the function of the code. So you're debugging based on the output of the code itself. [/INST] By the way, the plugin that we have for chat GPT, it does that routinely. It will send the thing in, it will get a result, it will discover, the LLM will discover itself that the result is not plausible and it will go back and say, oh, I'm sorry, it's very polite and it goes back and says, I'll rewrite that piece of code and then it will try it again and get the result. The other thing that's pretty interesting is when you're just running, so one of the new concepts that we have, we invented this whole idea of notebooks back 36 years ago now. And so now there's the question of sort of how do you combine this idea of notebooks where you have text and code and output? How do you combine that with the notion of chat and so on? And there's some really interesting things there. Like for example, a very typical thing now is we have these notebooks where as soon as if the thing produces errors, if they run this code and it produces messages and so on, the LLM automatically not only looks at those messages, it can also see all kinds of internal information about stack traces and things like this. And it can then, it does a remarkably good job of guessing what's wrong and telling you. So in other words, it's looking at things, sort of interesting. It's kind of a typical sort of AI-ish thing that it's able to have more sensory data than we humans are able to have, because it's able to look at a bunch of stuff that we humans would kind of glaze over looking at. </s><s>[INST] And it's able to then come up with, oh, this is the explanation of what's happening. And what is the data, the stack trace, the code you've written previously, the natural [/INST] language you've written? It's also what's happening is one of the things is, for example, when there's these messages, there's documentation about these messages. There's examples where the messages have occurred, otherwise all these kinds of things. The other thing that's really amusing with this is when it makes a mistake. One of the things that's in our prompt when the code doesn't work is read the documentation. And have another piece of the plugin that lets it read documentation. And that, again, is very, very useful because it will figure out sometimes it'll make up the name of some option for some function that doesn't really exist, read the documentation, it'll have some wrong structure for the function and so on. That's a powerful thing. I mean, the thing that I've realized is we built this language over the course of all these years to be nice and coherent and consistent and so on, so it's easy for humans to understand. Turns out, there was a side effect that I didn't anticipate, which is it makes it easy </s><s>[INST] for AIs to understand. It's almost like another natural language, but so Wolfram language is a kind of foreign language. Yes. You have a lineup, English, French, Japanese, Wolfram language, and then, I don't know, [/INST] Spanish, and then the system is not going to notice. So Wolfram? Yes. Well, yes. I mean, maybe. That's an interesting question because it really depends on what I see as being an important piece of fundamental science that basically just jumped out at us with chat GPT because I think the real question is why does chat GPT work? How is it possible to encapsulate, to successfully reproduce all these kinds of things in natural language with a comparatively small, he says, a couple hundred billion weights of neural net and so on? I think that relates to kind of a fundamental fact about language, which the main thing is that I think there's a structure to language that we haven't really explored very well. It's kind of this semantic grammar I'm talking about about language. We kind of know that ... When we set up human language, we know that it has certain regularities. We know that it has a certain grammatical structure. Noun followed by verb, followed by noun, adjectives, et cetera, et cetera, et cetera. That's its kind of grammatical structure. But I think the thing that chat GPT is showing us is that there's an additional kind of regularity to language which has to do with the meaning of the language beyond just this pure part speech combination type of thing. I think the one example of that that we've had in the past is logic. I think my picture of how was logic invented, how was logic discovered, it really was a thing that was discovered in its original conception. It was discovered, presumably by Aristotle, who listened to a bunch of people, orators, giving speeches. And this one made sense, that one doesn't make sense. You see these patterns of if the Persians do this, then this does that, etc., etc., etc. What Aristotle realized is there's a structure to those sentences, there's a structure to that rhetoric that doesn't matter whether it's the Persians and the Greeks, or whether it's the cats and the dogs. It's just P and Q. You can abstract from the details of these particular sentences. You can lift out this formal structure, </s><s>[INST] and that's what logic is. That's a heck of a discovery, by the way, logic. You're making me [/INST] realize, no. Yeah. It's not obvious. The fact that there is an abstraction from natural language that has where you can fill in any word you want is a very interesting discovery. Now, it took a long time to mature. Aristotle had this idea of syllogistic logic, where there were these particular patterns of how you could argue things, so to speak. In the Middle Ages, part of education was you memorize the syllogisms. I forget how many there were, but 15 of them or something. They all had names. They all had mnemonics, like I think barbera and celerant were two of the mnemonics for the syllogisms. This is a valid argument, because it follows the barbera syllogism, so to speak. It took until 1830, with George Boole, to get beyond that and see that there was a level of abstraction that was beyond this particular template of a sentence, so to speak. What's interesting there is, in a sense, chat GPT is operating at the Aristotelian level. It's essentially dealing with templates of sentences. By the time you get to Boole and Boolean algebra and this idea of you can have arbitrary depth-nested collections of ands and ors and nots, and you can resolve what they mean, that's the kind of thing. That's a computation story. You've gone beyond the pure templates of natural language to something which is an arbitrarily deep computation. But the thing that I think we realized from chat GPT is, you know, Aristotle stopped too quickly, and there was more that you could have lifted out of language as formal structures. I think there's, in a sense, we've captured some of that and some of what is in language. There's a lot of little calculus, little algebras of what you can say, what language talks about. Whether it's, I don't know, if you say, I go from place A to place B, place B to place C, then I know I've gone from place A to place C. If A is a friend of B and B is a friend of C, it doesn't necessarily follow that A is a friend of C. If you go from place A to place B to place B to place C, it doesn't matter how you went. Like logic, it doesn't matter whether you flew there, walked there, swam there, whatever. This transitivity of where you go is still valid, and there are many kinds of features, I think, of the way the world works that are captured in these aspects of language, so to speak. I think what ChatGBT effectively has found, just like it discovered logic, people are really surprised it can do these logical inferences, it discovered logic the same way Aristotle discovered logic, by looking at a lot of sentences effectively and noticing the patterns in those sentences. </s><s>[INST] But it feels like it's discovering something much more complicated than logic, so this kind of semantic grammar, I think you wrote about this, maybe you can call it the laws of language, I believe you call, or which I like, the laws of thought. [/INST] Yes. That was the title that George Boole had for his Boolean algebra back in 1830, but yes. Laws of thought? Yes, that was what he said. All right. So he thought he nailed it with Boolean algebra. Yeah. </s><s>[INST] There's more to it. There's more to it. It's a good question of how much more is there to it, and it seems like one of the reasons, as you imply that the reason GPT works, ChatGPT works, is that there's a finite number of things to it. Yeah, I mean, it's... [/INST] In some sense, GPT is discovering the laws of semantic grammar that underlies language. Yes. And what's sort of interesting is, in the computational universe, there's a lot of other kinds of computation that you could do. They're just not ones that we humans have cared about and operate with. And that's probably because our brains are built in a certain way, and the neural nets of our brains are not that different, in some sense, from the neural nets of a large language model. And so when we think about, and maybe we can talk about this some more, but when we think about what will AIs ultimately do, the answer is insofar as AIs are just doing computation, they can run off and do all these kinds of crazy computations. But the ones that we sort of have decided we care about is this kind of very limited </s><s>[INST] set. That's where the reinforcement learning with human feedback seems to come in. The more the AIs say the stuff that kind of interests us, the more we're impressed by it. So it can do a lot of interesting, intelligent things, but we're only interested in the AI systems when they communicate in a human-like way. Yes. [/INST] About human-like topics. Yes. Well, it's like technology. I mean, in a sense, the physical world provides all kinds of things, you know, all kinds of processes going on in physics. Only a limited set of those are ones that we capture and use for technology because they're only a limited set where we say, you know, this is a thing that we can sort of apply to the human purposes we currently care about. I mean, you might have said, okay, you pick up a piece of rock. You say, okay, this is a nice silicate. It contains all kinds of silicon. I don't care. Then you realize, oh, we could actually turn this into a semiconductor wafer and make it microprocessor out of it. And then we care a lot about it. Yes. And it's this thing about what do we... In the evolution of our civilization, what things do we identify as being things we care about? I mean, it's like when there was a little announcement recently of the possibility of a high-temperature superconductor that involved the element lutetium, which generally nobody has cared about. Yes. But suddenly, if there's this application that relates to kind of human purposes, we start to care a lot. </s><s>[INST] So given your thinking that GPT may have discovered inklings of laws of thought, do you think such laws exist? Can we linger on that? Yeah. [/INST] What's your intuition here? Yeah. What's your intuition here? Oh, definitely. I mean, the fact is, look, logic is but the first step. There are many other kinds of calculae about things that we consider about sort of things that happen in the world or things that are meaningful. </s><s>[INST] Well, how do you know logic is not the last step? You know what I mean? [/INST] Well, because we can plainly see that there are things... I mean, if you say, here's a sentence that is syntactically correct, okay? You look at it and you're like, the happy electron, eight, I don't know what, something that it just... You look at it and it's like this is meaningless. It's just a bunch of words. It's syntactically correct. The nouns and the verbs are in the right place, but it just doesn't mean anything. So there clearly is some rule that there are rules that determine when a sentence has the potential to be meaningful that go beyond the pure parts of speech syntax. And so the question is, what are those rules and are there a fairly finite set of those rules? My guess is that there's a fairly finite set of those rules. Once you have those rules, you have a kind of a construction kit, just like the rules of syntactic grammar give you a construction kit for making syntactically correct sentences. So you can also have a construction kit for making semantically correct sentences. Those sentences may not be realized in the world. I mean, I think the elephant flew to the moon, semantically, we know, we have an idea. If I say that to you, you kind of know what that means. </s><s>[INST] But the fact is it hasn't been realized in the world, so to speak. So semantically correct, perhaps as things that can be imagined with the human mind. No. Things that are consistent with both our imagination and our understanding of physical reality. I don't know. Yeah. Good question. [/INST] I mean, it's a good question. I mean, it's a good question. It's a good question. I think it is, given the way we have constructed language, it is things which fit with the things we're describing in language. It's a bit circular in the end, because you can, and the sort of boundaries of what is physically realizable, okay, let's take the example of motion. Motion is a complicated concept. It might seem like it's a concept that should have been figured out by the Greeks long ago, but it's actually a really pretty complicated concept, because what is motion? Motion is you can go from place A to place B, and it's still you when you get to the other end, right? You take an object, you move it, and it's still the same object, but it's in a different place. Now, even in ordinary physics, that doesn't always work that way. If you're near a space-time singularity in a black hole, for example, and you take your teapot or something, you don't have much of a teapot by the time it's near the space-time singularity. It's been completely deformed beyond recognition. That's a case where pure motion doesn't really work. You can't have a thing stay the same. This idea of motion is something that sort of is a slightly complicated idea, but once you have the idea of motion, you can start ... Once you have the idea that you're going to describe things as being the same thing, but in a different place, that sort of abstracted idea then has ... That has all sorts of consequences, like this transitivity of motion. Go from A to B, B to C, you've gone from A to C. At that level of description, you can have what are sort of inevitable consequences. They're inevitable features of the way you've sort of set things up. That's I think what this sort of semantic grammar is capturing, is things like that. I think that it's a question of what does the word mean when you say, I move from here to there. Well, it's complicated to say what that means. This is this whole issue of is pure motion possible, et cetera, et cetera, et cetera. Once you have kind of got an idea of what that means, then there are inevitable consequences of that idea. </s><s>[INST] But the very idea of meaning, it seems like there's some words that become ... It's like there's a latent ambiguity to them. I mean, it's the word like emotionally loaded words, like hate and love. It's like what do they mean? Exactly. Especially when you have relationships between complicated objects, we seem to take this kind of shortcut, descriptive shortcut to describe like object A hates object B. What's that really mean? [/INST] Right. Well, words are defined by kind of our social use of them. It's not a word ... In computational language, for example, when we say we have a construct there, we expect that that construct is a building block from which we can construct an arbitrarily tall tower. We have to have a very solid building block. We have to ... It turns into a piece of code, it has documentation, it's a whole thing. But the word hate, the documentation for that word ... Well, there isn't a standard documentation for that word, so to speak. It's a complicated thing defined by kind of how we use it when if it wasn't for the fact that we were using language ... I mean, so what is language at some level? Language is a way of packaging thoughts so that we can communicate them to another mind. Can these complicated words be converted into something that a computation agent can use? Right. The answer to that is that what one can do in computational language is make a specific definition. If you have a complicated word, like let's say the word eat, you'd think that's a simple word. It's animals eat things, whatever else, but you do programming. You say this function eats arguments, which is sort of poetically similar to the animal eating things. If you start to say, well, what are the implications of the function eating something, can the function be poisoned? Well, maybe it can actually, but if there's a tight mismatch or something in some language. But in what ... How far does that analogy go? It's just an analogy. Whereas if you use the word eat, in a computational language level, you would define there isn't a thing which you anchor to the kind of natural language concept eat, but it is now some precise </s><s>[INST] definition of that that then you can compute things from. But don't you think the analogy is also precise? Software eats the world. Don't you think there's something concrete in terms of meaning about analogies? [/INST] Sure. But the thing that sort of is the first target for computational language is to take sort of the ordinary meaning of things and try and make it precise, make it sufficiently precise. You can build these towers of computation on top of it. So it's kind of like, if you start with a piece of poetry and you say, I'm going to define my program with this piece of poetry, it's kind of like that's a difficult thing. It's better to say I'm going to just have this boring piece of prose and it's using words in the ordinary way, and that's how I'm communicating with my computer, and that's how I'm going to build this solid building block from which I can construct this whole </s><s>[INST] kind of computational tower. So there's some sense where if you take a poem and reduce it to something computable, you're gonna have very few things left. So maybe there's a bunch of human interaction that's just poetic, aimless nonsense. Well, it's just like recreational, like hamster in a wheel. [/INST] It's not actually producing anything. Well, I think that that's a complicated thing because in a sense, human linguistic communication is there's one mind, it's producing language. That language is having an effect on another mind. And the question of there's sort of a type of effect that is well-defined, let's say, where for example, it's very independent of the two minds, that there's communication where it can matter a lot, sort of what the experience of one mind is </s><s>[INST] versus another one and so on. Yeah, but what is the purpose of natural language communication? Well, I think the- So computational language somehow feels more amenable to the definition of purpose. It's like, yeah, you're given two clean representations of a concept and you can build a tower based on that. [/INST] Is natural language the same thing but more fuzzy? Yeah. Well, I think the story of natural language, that's the great invention of our species. We don't know whether it exists in other species, but we know it exists in our species. It's the thing that allows you to sort of communicate abstractly from like one generation of the species to another. There is an abstract version of knowledge that can be passed down. It doesn't have to be genetics, it doesn't have to be, you don't have to apprentice the next generation of birds to the previous one to show them how something works. There is this abstracted version of knowledge that can be kind of passed down. It still tends to rely because language is fuzzy. It does tend to rely on the fact that if we look at some ancient language where we don't have a chain of translations from it until what we have today, we may not understand that ancient language. We may not understand its concepts may be different from the ones that we have today. We still have to have something of a chain, but it is something where we can realistically expect to communicate abstract ideas, and that's one of the big roles of language. That's been this ability to sort of concretify abstract things is what language has provided. </s><s>[INST] Do you see natural language and thought as the same, the stuff that's going inside your [/INST] mind? </s><s>[INST] Well, that's been a long debate in philosophy. It seems to be become more important now when we think about how intelligent GPT is. Whatever that means. Whatever that means, but it seems like the stuff that's going on in the human mind seems something like intelligence. Yes. And it's language. Well, we call it intelligence. Yeah, we call it. Well, yes. And so you start to think of, okay, what's the relationship between thought, the language of thought, the laws of thought, the laws of the words like reasoning, and the laws of language, and how that has to do with computation, which seems like more rigorous, precise ways [/INST] of reasoning. Whatever that means. Yeah, we call it. Well, yes. Right. Which are beyond human. I mean, much of what computers do, humans do not do. I mean, you might say. Humans are a subset. Yes. Presumably. Yes. Hopefully. Yes. You might say, who needs computation when we have large language models? Large language models can just, eventually you'll have a big enough neural net, it can do anything. But they're really doing the kinds of things that humans quickly do. And there are plenty of formal things that humans never quickly do. For example, I don't know. People can do mental arithmetic. They can do a certain amount of math in their minds. I don't think many people can run a program in their minds of any sophistication. It's just not something people do. It's not something people have even thought of doing, because it's kind of not, you can easily run it on a computer. </s><s>[INST] Yes. Presumably. Yes. Hopefully. Betray a program. [/INST] Yeah. Aren't we running specialized programs? Yeah, yeah, yeah. But if I say to you, here's a Turing machine, tell me what it does after 50 steps, and you're trying to think about that in your mind, that's really hard to do. It's not what people do. Well, in this program. I mean, here's a Turing machine. </s><s>[INST] Yeah. I mean, the first people program, they build a computer, they program it. Just to answer your question about what the system does after 50 steps, I mean, humans build computers. Yes. [/INST] 50 steps. Yes. Yeah, that's right. But they've created something which is then, when they run it, it's doing something different than what's happening in their minds. I mean, they've outsourced that piece of computation from something that is internally happening </s><s>[INST] in their minds to something that is now a tool that's external to their minds. So by the way, humans, to you, didn't invent computers. [/INST] They discovered them. </s><s>[INST] They discovered computation, which they invented the technology of computers. The computer is just the kind of way to plug into this whole stream of computation. Yes. [/INST] There's probably other ways. Yes. There's probably a lot of ways. Well, for sure. I mean, the particular ways that we make computers out of semiconductors and electronics and so on, that's the particular technology stack we built. I mean, the story of a lot of what people try to do with quantum computing is finding different underlying physical infrastructure for doing computation. You know, biology does lots of computation. It does it using an infrastructure that's different from semiconductors and electronics. It's a molecular-scale sort of computational process that hopefully will understand more about ... I have some ideas about understanding more about that. That's another representation of computation. Things that happen in the physical universe at the level of these evolving hypergraphs </s><s>[INST] and so on, that's another sort of implementation layer for this abstract idea of computation. So if GPT or large language models are starting to develop or implicitly understand the laws [/INST] of language and thought, do you think they can be made explicit? Yes. We've got a bunch of effort. It's like doing natural science. What is happening in natural science? You have the world that's doing all of these complicated things, and then you discover Newton's law, for example. This is how motion works. This is the way that this particular sort of idealization of the world, this is how we describe it in a simple computationally reducible way. I think it's the same thing here. are sort of computationally reducible aspects of what's happening that you can get a kind </s><s>[INST] of narrative theory for, just as we've got narrative theories in physics and so on. Do you think it will be depressing or exciting when all the laws of thought are made explicit, human thought are made explicit? [/INST] I think that once you understand computationally reducibility, it's neither of those things. For example, people will say, oh, but I have free will. I operate in a way that is... They have the idea that they're doing something that is sort of internal to them that they're figuring out what's happening. But in fact, we think there are laws of physics that ultimately determine every electrical impulse and a nerve and things like this. You might say, isn't it depressing that we are ultimately just determined by the rules of physics, so to speak? It's the same thing. It's at a higher level. It's a shorter distance to get from semantic grammar to the way that we might construct a piece of text than it is to get from individual nerve firings to how we construct a piece of text. But it's not fundamentally different. By the way, as soon as we have this other level of description, it helps us to go even further. We'll end up being able to produce more and more complicated kinds of things that, just like if we didn't have a computer and we knew certain rules, we could write them down and go a certain distance. But once we have a computer, we can go vastly further, and this is the same </s><s>[INST] kind of thing. You wrote a blog post titled, What is Chad GPT doing and why does it work? We've been talking about this, but can we just step back and link on this question? What's Chad GPT doing? What are these a bunch of billion parameters trained on a large number of words? Why does it seem to work again? Is it because to the point you made that there's laws of language that can be discovered by such a process? Is there something? [/INST] Well, let's talk about sort of the low level of what Chad GPT is doing. I mean, ultimately, you give it a prompt. It's trying to work out what should the next word be, </s><s>[INST] which is wild. Isn't it surprising to you that this kind of low level dumb training procedure can create something syntactically correct first? And then semantically correct. [/INST] The thing that has been sort of a story of my life is realizing that simple rules can do much more complicated things than you imagine. That something that starts simple and starts simple to describe, can grow a thing that is vastly more complicated than you can imagine. Honestly, It's taken me, I've sort of been thinking about this now 40 years or so, and it always surprises me. I mean, even, for example, in our physics project, sort of thinking about the whole universe growing from these simple rules, I still resist, because I keep on thinking, how can something really complicated arise from something that simple? It just seems wrong, but yet, the majority of my life I've kind of known from things I've studied that this is the way things work. Yes, it is wild that it's possible to write a word at a time and produce a coherent essay, for example, but it's worth understanding kind of how that's working. If it was going to say the cat sat on thee, what's the next word? Okay, so how does it figure out the next word? It's seen a trillion words written on the internet, and it's seen the cat sat on the floor, the cat sat on the sofa, the cat sat on the whatever. The minimal thing to do is just say, let's look at what we saw on the internet. We saw 10,000 examples of the cat sat on the, what was the most probable next word? Let's just pick that out and say that's the next word. That's kind of what it, at some level, is trying to do. Now, the problem is there isn't enough text on the internet to, if you have a reasonable length of prompt, that specific prompt will never have occurred on the internet. As you kind of go further, there just won't be a place where you could have trained, where you could just just worked out probabilities from what was already there. If you say two plus two, there'll be a zillion examples of two plus two equaling four and a very small number of examples of two plus two equals five and so on, and you can pretty much know what's going to happen. Then the question is, well, if you can't just work out from examples, what's going to happen? Just no probabilistic, for example, what's going to happen? You have to have a model. This kind of an idea, this idea of making models of things, is an idea that really, I don't know, I think Galileo probably was one of the first people who sort of worked this out. I mean, it's kind of like, I think I gave an example of that, the book I wrote about Chachibitee, where it's kind of like Galileo was dropping cannonballs off the different floors of the tower of Pisa, and it's like, okay, you drop a cannonball off this floor, you drop a cannonball off this floor. You miss floor five or something for whatever reason, but you know the time it took the cannonball to fall to the ground from floors one, two, three, four, six, seven, eight, for example. Then the question is, can you work out? Can you make a model which figures out how long did it take the ball? How long would it take the ball to fall to the ground from the floor you didn't explicitly measure? The thing Galileo realized is that you can use math. You can use mathematical formulas to make a model for how long it will take the ball to fall. Now the question is, well, okay, you want to make a model for, for example, something much more elaborate, like you've got this arrangement of pixels, and is this arrangement of pixels an A or a B? Does it correspond to something we'd recognize as an A or B? You can make a similar kind. Each pixel is like a parameter in some equation, and you could write down this giant equation where the answer is either A or 1 or 2, A or B. The question is then, what kind of a model successfully reproduces the way that we humans would conclude that this is an A and this is a B? If there's a complicated extra tail on the top of the A, would we then conclude something different? This is the type of model that maps well into the way that we humans make distinguishes about things, and the big kind of meta discovery is neural nets are such a model. It's not obvious they would be such a model. It could be that human distinctions are not captured, you know, we could try searching around for a type of model, that could be a mathematical model, it could be some model based on something else that captures kind of typical human distinctions about things. It turns out this model that actually is very much the way that we think the architecture of brains works, that perhaps not surprisingly, that model actually corresponds to the way we make these distinctions. The core next point is that the kind of model, this neural net model, makes distinctions and generalizes things in sort of the same way that we humans do it. That's why when you say, the cat sat on the green blank, even though it didn't see many examples of the cat sat on the green, whatever, it can make a ... Or the odd vox sat on the green, whatever, I'm sure that particular sentence does not occur on the internet. It has to make a model that concludes what ... It has to kind of generalize from the actual examples that it's seen. That's the fact is that neural nets generalize in the same kind of way that we humans do. If we were the aliens might look at our neural net generalizations and say, that's crazy. That thing, when you put that extra little dot on the A, that isn't an A anymore. That messed the whole thing up, but for us humans, we make distinctions which seem to correspond to the kinds of distinctions that neural nets make. The thing that is just amazing to me about chat GPT is how similar the structure it has is to the very original way people imagined neural nets might work back in 1943. There's a lot of detailed engineering, great cleverness, but it's really the same idea. In fact, even the sort of elaborations of that idea where people said, let's put in some actual particular structure to try and make the neural net more elaborate to be very clever about it, most of that didn't matter. When you train this neural net, the one thing, this kind of transformer architecture, this attention idea, that really has to do with does every one of these neurons connect to every other neuron or is it somehow causally localized, so to speak? Is it like we're making a sequence of words and the words depend on previous words rather than just everything can depend on everything? That seems to be important in just organizing things so that you don't have a sort of a giant mess. But the thing worth understanding about what is chat GPT in the end, I mean, what is a neural net in the end? A neural net in the end is each neuron has a ... It's taking inputs from a bunch of other neurons. It's eventually going to have a numerical value. It's going to compute some number and it's saying, I'm going to look at the neurons above me. It's kind of a series of layers. It's going to look at the neurons above me and it's going to say, what are the values of all those neurons? Then it's going to add those up and multiply them by these weights and then it's going to apply some function that says if it's bigger than zero or something, then make it one and otherwise make it zero or some slightly more complicated function. You know very well how this works. </s><s>[INST] It's a giant equation with a lot of variables. You mentioned figuring out where the ball falls when you don't have data on the fourth floor. The equation here is not as simple as the equation. [/INST] Right. </s><s>[INST] It's an equation with 175 billion terms. It's quite surprising that in some sense a simple procedure of training such an equation [/INST] can lead to a good representation of natural language. The real issue is this architecture of a neural net where what's happening is you've turned, so neural nets always just deal with numbers. You've turned the sentence that you started with into a bunch of numbers like let's say by mapping each word of the 50,000 words in English, you just map each word or each part of a word into some number. You feed all those numbers in and then those numbers just go into the values of these neurons. Then what happens is it's just rippling down going layer to layer until it gets to the end. The chat GPT has about 400 layers. It just goes once through. Every new word it's going to compute just says here are the numbers from the words before. Let's compute the ... What does it compute? It computes the probabilities that it estimates for each of the possible 50,000 words that could come next. Then it decides sometimes it will use the most probable words, sometimes it will use not the most probable word. It's an interesting fact that there's this so-called temperature parameter, which temperature is zero. It's always using the most probable word that it estimated was the most probable thing to come next. If you increase the temperature, it'll be more and more random in its selection of words. It'll go down to the lower and lower probability words. The thing I was just playing with actually recently was the transition that happens as you increase the temperature. The thing goes bonkers at a particular ... Sometimes at a particular temperature ... Maybe about point two is the thing I was noticing from yesterday, actually, that usually it's giving reasonable answers, and then at that temperature, with some probability, it just starts spouting nonsense. Nobody knows why this happens. By the way, the thing to understand is it's putting down one word at a time, but the outer loop of the fact that it says, okay, I put down a word, now let's take the whole thing I wrote so far, let's feed that back in, let's put down another word. That outer loop, which seems almost trivial, is really important to the operation of the thing. For example, one of the things that is kind of funky is it'll give an answer, and you say to it, is that answer correct? It'll say, no. Why is that happening? It's fascinating, right? Why can it do that? Well, the answer is because it is going one word at a time, sort of four words, and it didn't ... It came along with some sort of chain of thought, in a sense, and it came up with completely the wrong answer, but as soon as you feed it, the whole thing that it came up with, it immediately knows that that isn't right. It immediately can recognize that was a bad syllogism or something, and can see what happened, even though as it was being led down this garden path, so to speak, it came to the wrong </s><s>[INST] place. But it's fascinating that this kind of procedure converges to something that forms a pretty good compressed representation of language on the internet. [/INST] That's quite ... I'm not sure what to make of it. Well, look, I think there are many things we don't understand. For example, 175 billion weights, it's maybe about a trillion bytes of information, which is very comparable to the training set that was used, and why that ... It sort of stands to some kind of reason that the number of weights in the neural net ... I don't know. I can't really argue that. I can't really give you a good ... In a sense, the very fact that insofar as there are definite rules of what's going on, you might expect that eventually we'll have a much smaller neural net that will successfully capture what's happening. I don't think the best way to do it is probably a neural net. I think a neural net is what you do when you don't know any other way to structure the thing. It's a very good thing to do if you don't know any other way to structure the thing, and for the last 2,000 years, we haven't known any other way to structure it. This is a pretty good way to start. But that doesn't mean you can't find, sort of, in a sense, more symbolic rules for what's going on that much of which will then be ... You can kind of get rid of much of the structure of the neural nets and replace it by things which are sort of pure steps of computation, so to speak, sort of with neural net stuff around the edges, and that becomes just a much simpler way to do it. </s><s>[INST] So the neural net, you hope, will reveal to us good symbolic rules that make the need [/INST] of the neural net less and less and less. Yeah, right. And there will still be some stuff that's kind of fuzzy, just like there are things that, it's like this question of what can we formalize? What can we turn into computational language? </s><s>[INST] What is just sort of, oh, it happens that way just because brains are set up that way. What do you think are the limitations of large language models, just to make it explicit? [/INST] Well, I mean, I think that deep computation is not what large language models do. I mean, that's just, it's a different kind of thing, you know, the outer loop of a large language model, if you're trying to do many steps in a computation, the only way you get to do that right now is by spooling out, you know, all the whole chain of thought as a bunch of words, basically. And you know, you can make a Turing machine out of that, if you want to, I just was doing that construction. You know, in principle, you can make an arbitrary computation by just spooling out the words. But it's a bizarre and inefficient way to do it. But it's something where the, you know, I think that's, you know, sort of the deep computation is, it's really what humans can do quickly. Large language models will probably be able to do well. Anything that you can do kind of off the top of your head type thing is really, you know, is good for large language models. And the things you do off the top of your head, you may not get them always right. </s><s>[INST] But you know, you'll, it's thinking through the same way we do. But I wonder if there's an automated way to do something that humans do well, much faster to where it like loops, so generate arbitrary large code bases of Wolfram, Wolfram language, [/INST] for example. </s><s>[INST] Well, the question is, what does he, what do you want the code base to do? Escape control and take over the world. [/INST] Okay. So you know, the thing is, when people say you know, we want to build this giant thing, right, a giant piece of computational language in a sense, it's sort of a failure of computational language. If the thing you have to build, in other words if we have a description, if you have a small description that's the thing that you represent in computational language and then the computer can compute from that. Yes. So, in a sense, as soon as you're giving a description, you have to somehow make that description something definite, something formal, and to say, okay, I'm going to give this piece of natural language, and then it's going to splurt out this giant formal structure, in a sense, that doesn't really make sense, because except insofar as that piece of natural language kind of plugs into what we socially know, so to speak, plugs into kind of our corpus of knowledge, then that's the way we're capturing a piece of that corpus of knowledge, but hopefully we will have done that in computational language. How do you make it do something that's big? Well, you have to have a way to describe what you want. </s><s>[INST] I can make it more explicit if you want. How about I just pop into my head. Iterate through all the members of Congress and figure out how to convince them that they have to let me, meaning the system, become president. Pass all the laws that allows AI systems to take control and be the president. I don't know. So that's a very explicit, like, figure out the individual life story of each congressman, each senator, anybody, I don't know what's required to really kind of pass legislation and figure out how to control them and manipulate them, get all the information. What would be the biggest fear of this congressman in such a way that you can take action on it in the digital space, so maybe threaten the destruction, reputation, or something [/INST] like this? Right. If I can describe what I want, to what extent can a large language model automate that? </s><s>[INST] With the help of the concretization of something like Wolfram Language, that makes it more, yeah, grounded. Yeah, grounded. I think it can go rather a long way. I'm also surprised how quickly I was able to generate. [/INST] I think it can go, yeah, yeah, right. An attack. That's a, yeah, you know. </s><s>[INST] I swear, I swear, I swear I did not think about this before, and it's funny how quickly, which is a very concerning thing, because that probably, this idea will probably do quite a bit of damage, and there might be a very large number of other such ideas. [/INST] Well, I'll give you a much more benign version of that idea, okay? You're going to make an AI tutoring system, and that's a benign version of what you're saying is, I want this person to understand this point. You're essentially doing machine learning where the loss function, the thing you're trying to get to is get the human to understand this point, and when you do a test on the human that they, yes, they correctly understand how this or that works. I am confident that a large language model type technology combined with computational language is going to be able to do pretty well at teaching us humans things, and it's going to be an interesting phenomenon, because individualized teaching is a thing that has been a goal for a long time. I think we're going to get that. I think it has many consequences. If you know me, if you, the AI, know me, tell me I'm about to do this thing, what are the three things I need to know, given what I already know, let's say I'm looking at some paper or something, it's like there's a version of the summary of that paper that is optimized </s><s>[INST] for me, so to speak, and where it really is and I think that's really going to work. It could understand the major gaps in your knowledge that it feels would actually give [/INST] you a deeper understanding of the topic. Right. That's an important thing, because it really changes, actually. I think when you think about education and so on, it really changes what's worth doing, what's not worth doing and so on. It makes ... I know in my life I've learned lots of different fields, so I don't know. Every time I always think that this is the one that I'm not going to be able to learn, but it turns out there are meta-methods for learning these things in the end. I think this idea that it becomes easier to be fed knowledge, so to speak, and if you need to know this particular thing, you can get taught it in an efficient way is something I think is an interesting feature. I think it makes things like the value of big towers of specialized knowledge become less significant compared to the meta-knowledge of understanding the big picture and being able to connect things together. I think there's been this huge trend of let's be more and more specialized because we have to sort of ascend these towers of knowledge, but by the time you can get more automation, being able to get to that place on the tower without having to go through all those steps, </s><s>[INST] I think it sort of changes that picture. Interesting. So your intuition is that in terms of the collective intelligence of the species, in the individual minds that make up that collective, there will be more, there will trend towards [/INST] being generalists and being kind of philosophers. That's what I think. I think that's where the humans are going to be useful. I think that a lot of these kind of the drilling, the mechanical working out of things is much more automatable. It's much more AI territory, so to speak. No more PhDs. That's interesting, yes. I mean, the kind of specialization, this kind of tower of specialization, which has been a feature of we've accumulated lots of knowledge in our species, and in a sense, every time we have a kind of automation, a building of tools, it becomes less necessary to know that whole tower, and it becomes something where you can just use a tool to get to the top of that tower. I think that the thing that is ultimately, when we think about, okay, what do the AIs do versus what do the humans do, it's like AIs, you tell them, you say, go achieve this particular objective. Okay, they can maybe figure out a way to achieve that objective. We say, what objective would you like to achieve? The AI has no intrinsic idea of that. It's not a defined thing. That's a thing which has to come from some other entity, and insofar as we are in charge, so to speak, or whatever it is, and our kind of web of society and history and so on, is the thing that is defining what objective we want to go to. </s><s>[INST] That's the thing that we humans are necessarily involved in, so to speak. To push back a little bit, don't you think that future versions of GPT would be able [/INST] to give a good answer to what objective would you like to achieve? On what basis? I mean, if they say, look, here's the terrible thing that could happen, they're taking the average of the internet, and they're saying, from the average of the internet, what do </s><s>[INST] people want to do? Well, that's the almost scottage of the most entertaining outcome is the most likely. [/INST] Okay. </s><s>[INST] I haven't heard that one from him, yeah. That could be one objective, is maximize global entertainment, the dark version of [/INST] that is drama, the good version of that is fun. Right. So, I mean, this question of what, if you say to the AI, what does the species want to achieve? Yes. Okay? There'll be an answer, right? There'll be an answer. It'll be what the average of the internet says the species wants to achieve. </s><s>[INST] Well, I think you're using the word average very loosely there, right? So, I think the answers will become more and more interesting as these language models are trained better and better. No, but I mean, in the end, it's a reflection back of what we've already said. Yes. But there's a deeper wisdom to the collective intelligence, presumably, than each individual. Maybe. [/INST] Isn't that what we're trying to do as a society? Well, I mean, that's an important, that's an interesting question. I mean, in so far as some of us work on trying to innovate and figure out new things and so on, it is sometimes, it's a complicated interplay between sort of the individual doing the crazy thing off in some spur, so to speak, versus the collective that's trying to do sort of the high inertia average thing. And it's, you know, sometimes the collective, you know, is bubbling up things that are interesting and sometimes it's pulling down kind of the attempt to make this kind of innovative direction. </s><s>[INST] But don't you think the large language models would see beyond that simplification will say maybe intellectual and career diversity is really important, so you need the crazy people on the outlier, on the outskirts? And so, like the actual . . . What's the purpose of this whole thing is to explore through this kind of dynamics that we've been using as a human civilization, which is most of us focus on one thing, and then there's the crazy people on the outskirts doing the opposite of that one thing and you kind of, they pull the whole society together. There's the mainstream science and then there's the crazy science. That's just been the history of human civilization. And maybe the AI system will be able to see that, and the more and more impressed we are by a language model telling us this, the more control will give it to it, and the more we'll be willing to let it run our society. And hence, there's this kind of loop where the society could be manipulated to let the [/INST] AI system run it. Right. Well, I mean, look, one of the things that's sort of interesting is we might say we always think we're making progress. But yet, if in a sense by saying, let's take what already exists and use that as a model for what should exist, then it's interesting that, for example, many religions have taken that point of view. There is a sacred book that got written at time x, and it defines how people should act for all future time. And it's a model that people have operated with. And in a sense, this is a version of that kind of statement. It's like, take the 2023 version of sort of how the world has exposed itself and use that </s><s>[INST] to define what the world should do in the future. But it's an imprecise definition, right? Because just like with religious texts and with GPT, the human interpretation of what GPT says will be the perturbation in the system, it'll be the noise, it'll be full of uncertainty. It's not like chat GPT will tell you exactly what to do. It'll tell you a narrative of what, like turn the other cheek kind of narrative, right? That's not a fully instructive narrative. Well, until the AIs control all the systems in the world. They will be able to very precisely tell you what to do. [/INST] Well, they'll do what they, they'll just do this or that thing. And not only that, they'll be auto suggesting to each person, do this next, do that next. So I think it's a slightly more prescriptive situation than one has typically seen. I think this whole question of sort of what's left for the human, so to speak, to what extent do we, this idea that there is an existing kind of corpus of purpose for humans defined by what's on the internet and so on, that's an important thing. But then the question of sort of as we explore what we can think of as the computational universe, as we explore all these different possibilities for what we could do, all these different inventions we could make, all these different things. The question is, which ones do we choose to follow? Those choices are the things that in a sense, if the humans want to still have kind of human progress, that's what we get to make those choices, so to speak. In other words, there's this idea, if you say, let's take the kind of what exists today and use that as the determiner of all of what there is in the future, the thing that is sort of the opportunity for humans is there will be many possibilities thrown up. There are many different things that could happen or be done, and the insofar as we want to be in the loop, the thing that makes sense for us to be in the loop doing is picking </s><s>[INST] which of those possibilities we want. But the degree to which there's a feedback loop of the idea that we're picking something starts becoming questionable because we're influenced by the various systems. Absolutely. Like if that becomes more and more source of our education and wisdom and knowledge. [/INST] Absolutely. Right. The AIs take over. I mean, I've thought for a long time that it's the AR auto suggestion. That's really the thing that makes the AIs take over. It's just that the humans just follow. Yeah. </s><s>[INST] We will no longer write emails to each other. We'll just send the auto suggested email. [/INST] Yeah. Yeah. But the thing where humans are potentially in the loop is when there's a choice. And when there's a choice which we could make based on our kind of whole web of history and so on. Yeah. And that's insofar as it's all just determined the humans don't have a place. And by the way, I mean, at some level, it's all kind of a complicated philosophical issue because at some level, the universe is just doing what it does. We are parts of that universe that are necessarily doing what we do, so to speak. Yet we feel we have sort of agency in what we're doing and that's its own separate kind of interesting issue. </s><s>[INST] Yes. And we also kind of feel like we're the final destination of what the universe was meant to create. But we very well could be and likely are some kind of intermediate step, obviously. Yeah. Well, we're most certainly some intermediate step. The question is if there's some cooler, more complex, more interesting things that's going [/INST] to be materialized. Yeah. </s><s>[INST] The computational universe is full of such things. [/INST] But in our particular pocket specifically, if this is the best we're going to do or not, if this is the best, we're going to, that's kind of a... We can make all kinds of interesting things in the computational universe. When we look at them, we say, yeah, you know, that's a thing. It doesn't really connect with our current way of thinking about things. I mean, it's like in mathematics, you know, we've got certain theorems that about three or four million that human mathematicians have written down and published and so on. But they're an infinite number of possible mathematical theorems. We just go out into the universe of possible theorems and pick another theorem and then people will say, well, you know, they look at it and they say, I don't know what this theorem means. It's not connected to the things that are part of kind of the web of history that we're dealing with. You know, I think one point to make about sort of understanding AI and its relationship to us is as we have this kind of whole infrastructure of AI is doing their thing and doing their thing in a way that is perhaps not readily understandable by us humans. You know, you might say, that's a very weird situation. How come we have built this thing that behaves in a way that we can't understand that's full of computational irreducibility, et cetera, et cetera, et cetera? You know, what is this? What's it going to feel like when the world is run by AIs whose operations we can't understand? And the thing one realizes is actually we've seen this before. That's what happens when we exist in the natural world. The natural world is full of things that operate according to definite rules. They have all kinds of computational irreducibility. We don't understand what the natural world is doing. Occasionally, when you say, are the AIs going to wipe us out, for example? Well, it's kind of like, is the machination of the AIs going to lead to this thing that eventually comes and destroys the species? Well, we can also ask the same thing about the natural world, or the machination of the natural world going to eventually lead to this thing that's going to make the earth explode or something like this. Those are questions. And insofar as we think we understand what's happening in the natural world, that's a result of science and natural science and so on. One of the things we can expect when there's this giant infrastructure of the AIs is that's where we have to kind of invent a new kind of natural science that kind of is the natural science that explains to us how the AIs work. We have, I don't know, a horse or something, and we're trying to ride the horse and go from here to there. We don't really understand how the horse works inside, but we can get certain rules and certain approaches that we take to persuade the horse to go from here to there and take us there. And that's the same type of thing that we're kind of dealing with, with the sort of incomprehensible, computationally irreducible AIs. we can identify these kinds of, we can find these kind of pockets of reducibility that we can kind of, you know, I don't know, we grabbing onto the mane of the horse or something to be able to ride it or we figure out, you know, if we do this or that to ride the horse, that that's a successful </s><s>[INST] way to get it to do what we're interested in doing. There does seem to be a difference between a horse and a large language model or something that could be called AGI connected to the internet. So let me just ask you about big philosophical question about the threats of these things. There's a lot of people like Eliezer Adkowski who worry about the existential risks of AI systems. Is that something that you worry about? You know, sometimes when you're building [/INST] an incredible system like Wolfram Alpha, you can kind of get lost in it. </s><s>[INST] Oh, I try and think a little bit about the implications of what one's doing. You know, it's like the Manhattan Project kind of situation where you're like, it's some of the most incredible physics and engineering being done. But it's like, huh, [/INST] where's this going to go? I think some of these arguments about kind of, you know, they'll always be a smarter AI, they'll always be, you know, and eventually the AIs will get smarter than us. And then all sorts of terrible things will happen. To me, some of those arguments remind me of kind of the ontological arguments for the distance of God and things like this. They're kind of arguments that are based on some particular model, fairly simple, model, often of kind of, there is always a greater this, that and the other. And what tends to happen in the sort of reality of how these things develop is that it's more complicated than you expect that the kind of simple logical argument that says, oh, eventually there'll be a super intelligence, and then it will do this and that turns out not to really be the story. It turns out to be a more complicated story. So, for example, here's an example of an issue. Is there an apex intelligence? Just like there might be an apex predator in some ecosystem, is there going to be an apex intelligence? The most intelligent thing that there could possibly be? Right? I think the answer is no. In fact, we already know this and it's a kind of a back to the whole computational irreducibility story, there's kind of a question of, even if you have sort of a Turing machine and you have a Turing machine that runs as long as possible before it halts. You say, is this the machine, is this the Apex machine that does that? There will always be a machine that can go longer, and as you go out to the infinite collection of possible Turing machines, you'll never have reached the end, so to speak. never, it's kind of like the same question of whether there'll always be another invention. Will you always be able to invent another thing? The answer is, yes, there's an infinite </s><s>[INST] tower of possible inventions. That's one definition of apex, which I also think might be true. Is there a species that's the apex intelligence right now on earth? So it's not trivial to say that humans are that. [/INST] Yeah, it's not trivial. I agree. It's a, you know, I think one of the things that I've long been curious about kind of other intelligences, so to speak, I mean, I, you know, I view intelligence is like computation and it's kind of a, you know, you're sort of, you have the set of rules, you deduce what happens. I have tended to think now that there's this sort of specialization of computation that is sort of a consciousness-like thing that has to do with these, you know, computational boundedness, single thread of experience, these kinds of things that are the specialization of computation that corresponds to a somewhat human like experience of the world. Now the question is, so, so that's, you know, there may be other intelligences, like, you know, the, you know, the aphorism, you know, the weather has a mind of its own. It's a different kind of intelligence that can compute all kinds of things that are hard for us to compute, but it is not well-aligned with us, with the way that we think about things. It doesn't think the way we think about things. In this idea of different intelligences, every different mind, every different human mind is a different intelligence that thinks about things in different ways. In terms of the formalism of our physics project, we talk about this idea of a ruled space, The space of all possible sort of rule systems and different minds are in a sense that different points in real space, human minds, ones that have grown up with the same kind of culture and ideas and things like this might be pretty close. In real space, pretty easy for them to communicate, pretty easy to translate, pretty easy to move from one place in real space that corresponds to one mind to another place in real space that corresponds to another sort of nearby mind. When we deal with more distant things in ruellial space like the pet cat or something, the pet cat has some aspects that are shared with us. The emotional responses of the cat are somewhat similar to ours, but the cat is further away in ruellial space than people are. So then the question is, can we identify, can we make a translation from our thought processes to the thought processes of a cat or something like this? What will happen when we get there? I think it's the case that many animals, dogs for example, they have elaborate factory systems, they have the smell architecture of the world, so to speak, in a way that we don't. If you were sort of talking to the dog and you could communicate in a language, the dog will say, well, this is a flowing, smelling, this, that, and the other thing, concepts that we just don't have any idea about. Now what's interesting about that is one day we will have chemical sensors that do a really pretty good job. We'll have artificial noses that work pretty well, and we might have our augmented reality systems show us the same map that the dog could see and things like this similar to what happens in the dog's brain. Eventually, we will have kind of expanded in ruellial space to the point where we will have those same sensory experiences that dogs have. We will have internalized what it means to have the smell landscape or whatever. And so then we will have kind of colonized that part of ruellial space until we haven't aren't gone. Some things that animals and so on do, we successfully understand, others we do not. And the question of what representation, how do we convert things that animals think about to things that we can think about, that's not a trivial thing. And I've long been curious. I had a very bizarre project at one point of trying to make an iPad game that a cat could win against its owner. I see it, Dave. It feels like there's a deep philosophical goal there, though. Yes. Yes. I was curious if pets can work in Minecraft or something and can construct things. What will they construct? And will what they construct be something where we look at it and we say, oh yeah, I recognize that. Or will it be something that looks to us like something that's out there in the computational universe that one of my cellular automata might have produced? Where we say, oh yeah, I can kind of see it operates according to </s><s>[INST] some rules. I don't know why you would use those rules. I don't know why you would care. Yeah. Actually, just to link on that seriously, is there a connector in the rule of space between you and a cat where the cat could legitimately win? So iPad is a very limited [/INST] interface. I wonder if there's a game where cats win. Yeah, I wonder. I think the problem is that cats don't tend to be that interested in what's happening on the iPad. </s><s>[INST] Yeah, that's an interface issue, probably. [/INST] Yeah, right. No, I think it is likely that there are plenty of animals that would successfully eat us if we were exposed to them. And so it's going to pounce faster than we can get out of the way and so on. And probably it's going to... We think we've hidden ourselves, </s><s>[INST] but we haven't successfully hidden ourselves. That's a physical strength. I wonder if there's something more in the realm of intelligence [/INST] where an animal like a cat could out... Well, I think there are things certainly in terms of the speed of processing certain kinds of things for sure. I mean, the question of what, is there a game of chess, for example, is there cat chess that the cats could play against each other? And if we tried to play a cat, we'd always lose. I don't know. </s><s>[INST] It might have to do with speed, but it might have to do with concepts also. There might [/INST] be concepts in the cat's head. I tend to think that our species, from its invention of language, has managed to build up this kind of tower of abstraction that, for things like a chess-like game, will make us win. In other words, we've become, through the fact that we've experienced language and learnt abstraction, we've sort of become smarter at those kinds of abstract kinds of things. Now, that doesn't make us smarter at catching a mouse or something. It makes us smarter at the things that we've chosen to sort of concern ourselves, which are these kinds of abstract things. And I think this is, again, back to the question of what does one care about? If you have the discussion with a cat, if we can translate things to have the discussion with a cat, the cat will say, I'm very excited that this light is moving. And we'll say, why do you care? And the cat will say, that's the most important thing in the world, that this thing moves around. I mean, it's like when you ask about, I don't know, you look at archaeological remains and you say, these people had this belief system about this and that was the most important thing in the world to them. And now we look at it and say, we don't know what the point of it was. I mean, I've been curious, they're these handprints on caves from 20,000 or more years ago. And it's like, nobody knows what these handprints were there for. That they may have been a representation of the most important thing you can imagine. They may just have been some kid who rubbed their hands in the mud and stuck them on the walls of the cave. We don't know. But this whole question of what is, when you say this question of sort of what's the smartest thing around, there's the question of what kind of computation you're trying to do. If you say you've got some well-defined computation, and how do you implement it? Well, you could implement it by nerve cells firing. You can implement it with silicon and electronics. You can implement it by some kind of molecular computation process in the human immune system, or in some molecular biology kind of thing, there are different ways to implement it. I think this question of sort of which, those different implementation methods will be of different speeds. They'll be able to do different things. If you say, so an interesting question would be, what kinds of abstractions are most natural in these different kinds of systems? For a cat, it's, for example, the visual scene that we see, we pick out certain objects, we recognize certain things in that visual scene, a cat might in principle recognize different things. I suspect, you know, evolution, biological evolution is very slow. I suspect what a cat notices is very similar, and we even know that from some neurophysiology. What a cat notices is very similar to what we noticed. Of course there's a you know, one obvious difference is cats have only two kinds of color receptors. They don't see in the same kind of color that we do. Now, you know, we say we're better, we have three color receptors, you know, red, green, blue. too, we're not the overall winner. I think the mantis shrimp is the overall winner with 15 color receptors, I think. It can make distinctions that with our current ... The mantis shrimp's view of reality, at least in terms of color, is much richer than ours. What's interesting is how do we get there? Imagine we have this augmented reality system that is seeing into the infrared, into the ultraviolet, things like this, and it's translating that into something that is connectable to our brains, either through our eyes or more directly into our brains. Then eventually, our web of the types of things we understand will extend to those kinds of constructs, just as they have extended. There are plenty of things where we see them in the modern world because we made them with technology, and now we understand what that is, but if we'd never seen that kind of thing, we wouldn't have a way to describe it, we wouldn't have a way to understand it, and so on. </s><s>[INST] All right. That actually stemmed from our conversation about whether AI is going to kill all of us. We've discussed this spreading of intelligence through rule of space, that in practice, it just seems that things get more complicated. Things are more complicated than the story of, well, if you build the thing that's plus one intelligence, that thing will be able to build the thing that's plus two intelligence and plus three intelligence, and that will be exponential, it'll become more intelligent exponentially faster, and so on, until it completely destroys everything. That intuition might still not be so simple, I might still carry validity. And there's two interesting trajectories here. One, a super intelligence system remains in Rulio proximity to humans. To where we're like, holy crap, this thing is really intelligent. Let's elect the president. And then there could be perhaps more terrifying intelligence that starts moving away. They might be around us now. They're moving far away in Rulio space, but they're still sharing physical resources with us. Right? And so they can rob us of those physical resources and destroy humans just kind of casually. Yeah. Just like nature could, but it seems like there's something unique about AI systems where there is this kind of exponential growth. Like the way, well, sorry, nature has so many things in it. One of the things that nature has, which is very interesting are viruses, for example. There is systems within nature [/INST] that have this kind of exponential effect. </s><s>[INST] Yeah, like nature code. And that terrifies us humans, because again, there's only 8 billion of us and you can just kind of, it's not that hard to just kind of whack them all real quick. So, I mean, is that something you think about? Yeah, I thought about that. Yes. The threat of it. I mean, are you as concerned about it as somebody like Eliezer Yarkovsky, for example, just big, big, painful, negative effects of AI on society? [/INST] Yeah. Real quick. You know, no, but perhaps that's because I'm intrinsically an optimist. Yeah. I mean, I think that there are things, I think the thing that one sees is there's going to be this one thing and it's going to just zap everything. Somehow, maybe I have faith in computational irreducibility, so to speak. That there's always unintended little corners that it's just like somebody says, I'm going to, I don't know, somebody has some bioweapon and they say, we're going to release this and it's going to do all this harm. But then it turns out it's more complicated than that because some humans are different and the exact way it works is a little different than you expect. It's something where sort of the great big, you smash the thing with something, the asteroid collides with the earth and it kind of, and yes, the earth is cold for two years or something and then lots of things die but not everything dies. There's usually, I mean, this is in a sense the sort of story of computational irreducibility. There always unexpected corners, there always unexpected consequences. I don't think the kind of wack it over the head with something and then it's all gone is that can obviously happen, the Earth can be swallowed up in a black hole or something, and then it's kind of presumably all over. But I think this question of what do I think the realistic paths are, I think that there will be sort of an increasing ... I mean, people have to get used to phenomena like computational irreducibility. There's an idea that we built the machines so we can understand what they do, and we're going to be able to control what happens. Well, that's not really right. Now the question is, is the result of that lack of control going to be that the machines kind of conspire and sort of wipe us out? Maybe just because I'm an optimist, I don't tend to think that that's in the cards. I think that as a realistic thing, I suspect what will sort of emerge, maybe, is kind of an ecosystem of the AIs just as ... Again, I don't really know. I mean, this is something that's hard to be clear about what will happen. I think that there are a lot of sort of details of what could we do? What systems in the world could we connect an AI to? I have to say, I was just a couple of days ago, I was working on this chat GBT plug-in kit that we have for Wolfram Language, where you can create a plug-in and it runs Wolfram Language Code, and it can run Wolfram Language Code back on your own computer. I was thinking, well, I can just make it, I can tell chat GBT, create a piece of code, and then just run it on my computer. I'm like, that sort of personalizes for me the what could possibly go wrong, so to speak. </s><s>[INST] Was that exciting or scary, that possibility? [/INST] It was a little bit scary, actually, because it's kind of like I realize I'm delegating to the AI, just write a piece of code. You're in charge, write a piece of code, run it on my computer, and pretty soon all my files will delete it. </s><s>[INST] That's like Russian-related, but much more complicated for me to know that. Yes, yes, right. That's a good drinking game. I don't know. I don't know. Well, right. [/INST] I mean, yes. For me to know that. Yes, right. That's why. That's how much you're drinking. It's an interesting question then. However, if you do that, what is the sandboxing that you should have? That's sort of a version of that question for the world. That is, as soon as you put the AIs in charge of things, you know, how much, how many constraints should there be on these systems before you put the AIs in charge of all the weapons and all these different kinds of systems? Right. </s><s>[INST] That's why... That's how much you're drinking. Here's the fun part about sandboxes. The AI knows about them. [/INST] It has the tools to crack them. Look, the fundamental problem of computer security is computational irreducibility. Because the fact is, any sandbox is never going to be a perfect sandbox. If you want the system to be able to do interesting things, I mean, this is the problem that's happened, the generic problem of computer security, that as soon as you have your firewall that is sophisticated enough to be a universal computer, that means it can do anything. And so long as if you find a way to poke it, so that you actually get it to do that universal computation thing, that's the way you kind of crawl around and get it to do the thing that it wasn't intended to do. And that's sort of… Another version of computational irreducibility is you can, you know, you can kind of… you </s><s>[INST] get it to do the thing you didn't expect it to do, so to speak. There are so many interesting possibilities here that manifest themselves from the computational irreducibility here. There's just so many things can happen because in digital space things move so quickly. You can have a chatbot, you can have a piece of code that you can basically have chat GPT generated viruses, accidentally or on purpose, and they are digital viruses. Yes. And they could be brain viruses too. They convince kind of like phishing emails. Yes. [/INST] They can convince you of stuff. Yes. windows. Yes. They can convince you. Yes. And no doubt, you can, you know, had the loop of the machine learning loop of making things that convince people of things is surely going to get easier to do. And then what does that look like? Well, it's again, we humans are, this is a new environment for us. And admittedly, it's an environment which a little bit scarily is changing much more rapidly than, I mean, people worry about, climate change is going to happen over hundreds of years and the environment is changing, but the environment for, in the kind of digital </s><s>[INST] environment might change in six months. So one of the relevant concerns here in terms of the impact of GPT on society is the nature of truth that's relevant to Wolfram Alpha, because computation through symbolic reasoning that's embodied in Wolfram Alpha as the interface, there's a kind of sense that what Wolfram Alpha tells me is true. So we hope. Yeah. I mean, you could probably analyze that, you could show, you can't prove that it's always [/INST] going to be true, computation or disability, but it's going to be more true than not. Look, the fact is, it will be the correct consequence of the rules you've specified. And in so far as it talks about the real world, that is our job in sort of curating and collecting data to make sure that that data is quotes as true as possible. Now, what does that mean? Well, it's always an interesting question. I mean, for us, our operational definition of truth is, somebody says, who's the best actress? Who knows? But somebody won the Oscar and that's a definite fact. And so that's the kind of thing that we can make computational as a piece of truth. If you ask these things, which a sensor measured this thing, it did it this way, a machine learning system, this particular machine learning system recognized this thing, that's a sort of a definite effect, so to speak. And there is a good network of those things in the world. It's certainly the case that particularly when you say, is so-and-so a good person, that's a hopelessly... We might have a computational language definition of good, I don't think it'd be very interesting because that's a very messy kind of concept, not really amenable to kind of... I think as far as we will get with those kinds of things is I want X. There's a kind of meaningful calculus of I want X and that has various consequences. I mean, I'm not sure, I haven't thought this through properly, but I think a concept like </s><s>[INST] is so-and-so a good person, is that true or not? That's a mess. That's a mess that's amenable to computation. I think it's a mess when humans try to define what's good, like through legislation, but when humans try to define what's good through literature, through history books, through [/INST] poetry, it starts being pulled. I don't know. That particular thing, it's kind of like we're going into kind of the ethics of what counts as good, so to speak, and what do we think is right and so on. I think that's a thing which one feature is we don't all agree about that. There's no theorems about, there's no theoretical framework that says this is the way that </s><s>[INST] ethics has to be. Well, first of all, there's stuff we kind of agree on and there's some empirical backing for what works and what doesn't from just even the morals and ethics within religious texts. So we seem to mostly agree that murder is bad. The certain universals that seem to emerge. I wonder where the murder of an AI is bad. Well, I tend to think yes, but I think we're going to have to contend with that question. [/INST] I wonder what AI would say. Yeah. Well, I think one of the things with AIs is it's one thing to wipe out that AI that has no owner. You can easily imagine an AI kind of hanging out on the internet without having any particular owner or anything like that. And then you say, well, what harm does it, it's okay to get rid of that AI. Of course, if the AI has 10,000 friends who are humans and all those 10,000 humans will be incredibly upset that this AI just got exterminated, it becomes a slightly different, more entangled story. But yeah, I know I think that this question about what do humans agree about, it's, there are certain things that human laws have tended to consistently agree about. There have been times in history when people have sort of gone away from certain kinds of laws, even ones that we would now say, how could you possibly have not done it that way? That just doesn't seem right at all. But I think this question of what I don't think one can say beyond saying, if you have a set of rules that will cause the species to go extinct, that's probably, you could say that's probably not a winning set of laws, because even to have a thing on which you </s><s>[INST] can operate laws requires that the species not be extinct. But between sort of what's the distance between Chicago, New York that Wolfram Alpha can answer and the question of if this person is good or not, there seems to be a lot of gray area. And that starts becoming really interesting. I think your, since the creation of Wolfram Alpha have been a kind of arbiter of truth at a large scale. [/INST] So the system is generates more truth and try to make sure that the things are true. I mean, look, as a practical matter, when people write computational contracts, and it's kind of like, you know, if this happens in the world, then do this. And this hasn't developed as quickly as it might have done, you know, this has been a sort of a blockchain story and parts and so on, although blockchain is not really necessary for the idea of computational contracts. But you can imagine that eventually, sort of a large part of what's in the world are these giant chains and networks of computational contracts. And then something happens in the world, and this whole giant domino effect of contracts firing autonomously that cause other things to happen. And you know, for us, you know, we've been the main sort of source of the Oracle of quotes facts or truth or something for things like blockchain, computational contracts and such like, and there's a question of, you know, what, you know, I consider that responsibility to actually get the stuff right. And one of the things that is tricky sometimes is when is it true? When is it a fact? When is it not a fact? I think the best we can do is to say, you know, we have a procedure, we follow the procedure, we might get it wrong, but at least we won't be corrupt about getting it wrong, so to speak. </s><s>[INST] So that's beautifully put. I have a transparency about the procedure. The problem starts to emerge when the things that you convert into computational language start to expand, for example, into the realm of politics. So this is where it's almost like this nice dance of Wolfram Alpha and Chad GPT. Chad GPT, like you said, is shallow and broad. [/INST] So it's going to give you an opinion on everything. But it writes fiction as well as fact, which is exactly how it's built. I mean, that's exactly it is making language and it is making both even in code it writes fiction. I mean, it's kind of fun to see sometimes, you know, it'll write fictional Wolfram language code that kind of looks right. Yeah, it looks right. But it's actually not pragmatically correct. But yes, it has a view of kind of roughly how the world works at the same level as books of fiction talk about roughly how the world works. They just don't happen to be the way the world actually worked or whatever. But yes, that's no, I agree. That's sort of a, you know, we are attempting with our whole Wolfram language computational language thing to represent at least, well, it doesn't necessarily have to be how the actual world works, because we can invent a set of rules that aren't the way the actual world works and run those rules. But then we're saying we're going to accurately represent the results of running those rules, which might or might not be the actual rules of the world. But we also are trying to capture features of the world as accurately as possible to represent what happens in the world. Now, again, as we've discussed, you know, the atoms in the world arranged, you know, you say, I don't know, you know, was there a tank that showed up, you know, that, you know, drove somewhere? Okay, well, you know, what is a tank? It's an arrangement of atoms that we abstractly describe as a tank. And you could say, well, you know, there's some arrangement of atoms that is a different arrangement of atoms, but it's not. You know, we didn't we didn't decide. It's like this observer theory question of, you know, what what arrangement of atoms </s><s>[INST] counts as a tank versus not a tank? So there's even things that would consider strong facts. [/INST] you could start to kind of disassemble them and show that they're not. Right. I mean, so the question of whether, oh, I don't know, was this gust of wind strong enough to blow over this particular thing? Well, a gust of wind is a complicated concept. You know, it's full of little pieces of fluid dynamics and little vortices here and there. And you have to define, you know, what the aspect of the gust of wind that you care about might be. It put this amount of pressure on this blade of some wind turbine or something. But, you know, if you say, if you have something, which is the fact of the gust of wind was the strong, whatever that, you know, that is you have to have some definition of that. You have to have some measuring device that says, according to my measuring device that was constructed this way, the gust of wind was this. </s><s>[INST] So what can you say about the nature of truth that's useful for us to understand chat GPT because you've been you've been contending with this idea of what is fact and not and it seems like chat GPT is used a lot now. I've seen it used by journalists to write articles and so you have people that are working with large language models trying to desperately figure out how do we essentially censor them through different mechanisms either manually or through reinforcement learning with human feedback try to align them to to not say fiction just [/INST] to say nonfiction fiction as much as possible. This is the importance of computational language as an intermediate it's kind of like you've got the large language model it's able to surface something which is a formal precise thing that you can then look at and you can run tests on it and you can do all kinds of things that's always going to work the same way and it's precisely defined what it does and then the large language model is the interface. I mean the way I viewed these large language models wonder there important I mean there many use cases and you know it's a remarkable thing could talk about some of these you know literally you know every day we're coming up with a couple of new use cases. Some of which are very very very surprising and things were, mean but the best used cases are ones where it's you know even if it gets it roughly right. It's still a huge win. Like a use case we had from a week or two ago is read our bug reports. You know, we've got hundreds of thousands of bug reports that have accumulated over decades. And it's like, you know, can we have it just read the bug report, figure out where the, where is the bug likely to be? And, you know, home in on that piece of code, maybe it'll even suggest some, some, you know, sort of way to fix the code. It might get that, it might be nonsense, what it says about how to fix the code, but it's </s><s>[INST] incredibly useful that it was able to, you know, so awesome. It's so awesome. Because even the nonsense will somehow be instructive. I don't, I don't quite understand that yet. Yeah. There's so many programming related things like, uh, for example, uh, translating for one programming language to another is really, really interesting. It's extremely [/INST] effective, but then you, the failures reveal the path forward also. Yeah. But I think, I mean, the, the big thing, I mean, in, in that kind of discussion, the unique thing about our computational language is it was intended to be read by humans. Yes. And so it has a really important, right. And so it has this thing where you can, but, but, you know, thinking about sort of chat GPT and its use and so on, the, one of the big things about it, I think is it's a linguistic user interface that is so a typical use case might be in the, take the journalist case, for example, it's like, let's say I have five facts that I'm trying to turn into an article or I'm trying to, I'm trying to write a report where I have basically five facts that I'm trying to include in this report. But then I feed those five facts to chat GPT, it puffs them out into this big report. And then, and then that's a good interface for, you know... If I just gave if I just had in my terms, those five bullet points, and I gave them to some other person, the person said, I don't know what you're talking about, because these are, you know, this is your version of this sort of quick notes about these five bullet points. But if you puff it out into this thing, which is kind of connects to the collective understanding of language, then somebody else can look at it and say, Oh, okay, I understand what you're talking about. Now you can also have a situation where that thing that was puffed out is fed to another large language model. It's kind of like you're applying for the permit to, I don't know, grow fish in some place or something like this. And you have these facts that you're putting in. I'm going to have this kind of water and I don't know what it is. You've just got a few bullet points. It puffs it out into this big application. You fill it out. Then at the other end, the fisheries bureau has another large language model that just crushes it down because the fisheries bureau cares about these three points and it knows what it cares about. So it's really the natural language produced by the large language model is sort of a transport layer that is really LLM communicates with LLM. I mean, it's kind of like I write a piece of email using my LLM and puff it out from the things I want to say your LLM turns it into and the conclusion is X. Now the issue is that the thing is going to make this thing that is sort of semantically plausible and it might not actually be what you, it might not be kind of relate to the world in the way that you think it should relate to the world. I've seen this, I've been doing, okay, I'll give you a couple of examples. I was doing this thing when we announced this plugin for chat GPT. I had this lovely example of a math word problem, some complicated thing, and it did a spectacular job of taking apart this elaborate thing about, this person has twice as many chickens as this, et cetera, et cetera, et cetera, and it turned it into a bunch of equations. It fed them to orphan language. We solved the equations. Everybody did great. We gave back the results and I thought, okay, I'm going to put this in this blog post I'm writing. I thought I'd better just check and turns out it got everything, all the hard stuff it got right. At the very end, last two lines, it just completely goofed it up and gave the wrong answer and I would not have noticed this. Same thing happened to me two days ago. I thought I made this with this chat GPT plugin kit. I made a thing that would emit a sound, would play a tune on my local computer. Chat GPT would produce a series of notes and it would play this tune on my computer. Very cool. I thought I'm going to ask it, </s><s>[INST] play the tune that Hal sang when Hal was being disconnected in 2001. There it is. Daisy, was it Daisy? Yes. [/INST] Daisy. Yes. Daisy. Yeah. Right. It's okay. It produces a bunch of notes and I'm like, this is spectacular. This is amazing. Then I thought, I was just going to put it in and then I thought, I better actually play this. I did and it was Mary Had a Little Lamb. </s><s>[INST] Oh, wow. Oh, wow. But it was Mary Had a Little Lamb. Yeah. Yes. [/INST] Wow. So it's correct, but wrong. It could easily be mistaken. Yes, right. In fact, I had this quote from Hal to explain, it's as the Hal states in the movie. The Hal 9000, the thing was just a rhetorical device because I'm realizing, oh my gosh, this chat GPT could have easily fooled me. It did this amazing thing of knowing this thing about the movie and being able to turn that into the notes of the song, except it's the wrong song. In the movie, Hal says, I think it's something like, no 9000 series computer has ever been found to make an error. We are, for all practical purposes, perfect and incapable of error. I thought that was a charming quote from Hal to make in connection </s><s>[INST] with what chat GPT had done in that case. Yeah. The interesting things about the LLMs, like you said, that they are very willing to admit their error. [/INST] Well, yes. I mean, that's a question of the reinforcement learning human feedback thing. Oh, right. That's the really remarkable thing about chat GPT is I had been following what was happening with large language models and I'd played with them a whole bunch and they were like, it's like what you would expect based on statistical continuation of language. It's interesting, but it's not breakout exciting. I think the human feedback reinforcement learning and making chat GPT try and do the things that humans really wanted to do, that broke through, that reached this threshold where the thing really is interesting to us humans. And by the way, it's interesting to see how you change the temperature or something like that, the thing goes bonkers and it no longer is interesting to humans. It's producing garbage And it's kind of right. Somehow it managed to get above this threshold where it really is well aligned to what we humans are interested in. And I think nobody saw that coming, I think. Certainly nobody I've talked to and nobody who was involved in that project seems to have known it was coming. It's just one of these things that is a sort of remarkable threshold. I mean, you know, when we built Wolfram Alpha, for example, I didn't know it was going to work. You know, we tried to build something that would have enough knowledge of the world, that it could answer a reasonable set of questions, that we could do good enough natural language understanding, that typical things you type in would work. We didn't know where that threshold was. I mean, I was not sure that it was the right decade to try and build this, even the right, you know, 50 years to try and build it. You know, and I think that was the same type of thing with chat GPT, that I don't think anybody could have predicted that, you know, 2022 would be the year </s><s>[INST] that this became possible. I think, yeah, you tell a story about Marvin Miski and showing it to him [/INST] and saying that like, no, no, no, this time it actually works. Yes. Yes. And I mean, you know, it's the same thing, for me looking at these large language models, it's like when people are first saying the first few weeks of chat GPT, it's like, oh, yeah, you know, I've seen these large language models, um, and then, uh, you know, and then I actually try it and, uh, you know, oh my gosh, it actually works. And I think it's, but you know, the things and, you know, the thing I found, you know, I remember one of the first things I tried was write a persuasive essay that a wolf is the bluest kind of animal. Okay? So it writes this thing and it starts talking about these wolves that live on the Tibetan plateau and named some Latin name and so on. And I'm like, really? And I'm starting to look it up on the web. And it's like, well, it's actually complete nonsense, but it's extremely plausible. I mean, it's plausible enough that I was going and looking up on the web and wondering if there was a wolf that was blue. You know, I mentioned this on some live streams I've done. And so people have been sending me these blue wolves. </s><s>[INST] Maybe I was on to something. Can you kind of give your wise sage advice about what humans who have never interacted with AI systems? Not even like with Wolfram Alpha are now interacting with Chad GPT because it becomes, it's accessible to a certain demographic. They may have not touched AI systems before. What do we do with truth? Like journalists, for example, how do we think [/INST] about the output of these systems? I think this idea, the idea that you're going to get factual output is not a very good idea. I mean, it's just, this is not, it is a linguistic interface. It is producing language and language can be truthful or not truthful. And that's a different slice of what's going on. I think that, you know, what we see in for example kind of, go check this with your fact source, for example. You can do that to some extent, but then it's going to not check something. That is again a thing that is sort of a, does it check in the right place? I mean we see that and does it call the Wolfram plug-in in the right place? You know, often it does. Sometimes it doesn't. I think the real thing to understand about what's happening is, which I think is very exciting, is the great democratization of access to computation. I think that when you look at... There's been a long period of time when computation and the ability to figure out things with computers has been something that only the druids at some level can achieve. I myself have been involved in trying to de-druidify access to computation back before Mathematica existed in 1988. If you were a physicist or something like that and you wanted to do a computation, you would find a programmer. You would go and delegate the computation to that programmer. Hopefully, they'd come back with something useful. Maybe they wouldn't. There'd be this long multi-week loop that you go through. It was actually very, very interesting to see, 1988. People like physicists, mathematicians, and so on. Then lots of other people. There's very rapid transition of people realizing they themselves could actually type with their own fingers and make some piece of code that would do a computation that they cared about. It's been exciting to see lots of discoveries and so on made by using that tool. We see the same thing Wolfram Alpha is dealing with is not as deep computation as you can achieve with whole Wolfram Language Mathematica stack. The thing that's, to me, particularly exciting about the large language model linguistic interface mechanism is it dramatically broadens the access to deep computation. One of the things I thought about recently is what's going to happen to all these programmers? What's going to happen to all these people who, a lot of what they do is write slabs of boilerplate code? In a sense, I've been saying for 40 years, that's not a very good idea. You can automate a lot of that stuff with a high enough level language, that slab of code that's designed in the right way, that slab of code turns into this one function we just implemented that you can just use. In a sense, the fact that there's all of this activity of doing lower level programming is something, for me, it seemed like I don't think this is the right thing to do. Lots of people have used our technology and not had to do that, but the fact is that when you look at computer science departments that have turned into places where people are learning the trade of programming, so to speak, it's a question of what's going to happen. I think there are two dynamics. One is that boilerplate programming is going to go the way that assembly language went back in the day of something where it's really mostly specified at a higher level. You start with natural language, you turn it into a computational language, you look at the computational language, you run tests, you understand that's what's supposed to happen. If we do a great job with compilation of the computational language, it might turn into LLVM or something like this, or it just directly gets run through the algorithms we have and so on. That's a tearing down of this big structure that's been built of teaching people programming. On the other hand, the other dynamic is vastly more people are going to care about computation. All those departments of art history or something that really didn't use computation before now have the possibility of accessing it by virtue of this linguistic </s><s>[INST] interface mechanism. If you create an interface that allows you to interpret the debug and interact with the computational language, then that makes it even more accessible. [/INST] I think the thing is that right now, the average art history student or something probably isn't going to... They don't think they know about programming and things like this, but by the time it really becomes a kind of purely... You just walk up to it, there's no documentation, you start just typing, compare these pictures with these pictures and see the use of this color, whatever, and you generate this piece of computational language code that gets run, you see the results, you say, oh, that looks roughly right, or you say that's crazy. Maybe then you eventually get to say, well, I better actually try and understand what this computational language code did, and that becomes the thing that you learn. It's kind of an interesting thing because unlike with mathematics, where you kind of have to learn it before you can use it, this is a case where you can use it before you have to learn it. </s><s>[INST] Well, I got a sad possibility here, or maybe exciting possibility, that very quickly people won't even look at the computational language. They'll trust that it's generated correctly as you get better and better at generating that language. [/INST] Yes, I think that there will be enough cases where people see... Because you can make it generate tests too, and so you'll say, we're doing that. I mean, it's a pretty cool thing actually, because you say, this is the code, and here are a bunch of examples of running the code. Okay, people will at least look at those, and they'll say that example is wrong, and then it'll wind back from there. I agree that the intermediate level of people reading the computational language code, in some case people will do that, in other case, people just look at the tests, or even just look at the results, and sometimes it'll be obvious that you got the thing you wanted to get, because you were just describing, make me this interface that has two sliders here, and you can see it has those two sliders there, and that's the result you want. But I think one of the questions then is in that setting where you have this kind of broad ability of people to access computation, what should people learn? In other words, right now you go to computer science school, so to speak, and a large part of what people end up learning. It's been a funny historical development because back 30, 40 years ago, computer science departments were quite small and they taught things like finite automata theory and compiler theory and things like this. Company like mine rarely hired people who'd come out of those programs because the stuff they knew was, I think is very interesting. I love that theoretical stuff, but it wasn't that useful for the things we actually had to build in software engineering. And then there was this big pivot in the 90s, I guess, where there was a big demand for IT type programming and so on and software engineering, and then big demand from students and so on. We wanted to learn this stuff. I think the thing that really was happening in part was lots of different fields of human endeavor were becoming computational. For all X, there was a computational X. And that was the thing that the people were responding to. But then this idea emerged that to get to that point, the main thing you had to do was to learn this kind of trade or skill of doing programming language type programming. It's a strange thing, actually, because I remember back when I used to be in the professoring business, which is now 35 years ago, so gosh, it's rather a wrong time. It was right when they were just starting to emerge kind of computer science departments at sort of fancy research universities and so on. I mean, some had already had it, but the other ones that were just starting to have that. And it was kind of a thing where they were kind of wondering, are we going to put this thing that is essentially a trade-like skill, are we going to somehow attach this to the rest of what we're doing? A lot of these kind of knowledge work type activities have always seemed like things where that's where the humans have to go to school and learn all this stuff, and that's never going to be automated. It's kind of shocking that rather quickly, a lot of that stuff is clearly automatable. But the question then is, okay, so if it isn't worth learning how to do car mechanics, you don't need to know how to drive the car, so to speak. What do you need to learn? And in other words, if you don't need to know the mechanics of how to tell the computer in detail, make this loop, set this variable, set up this array, whatever else, if you don't have to learn that stuff, if you don't have to learn the kind of under-the-hood things, what do you have to learn? I think the answer is, you need to have an idea where you want to drive the car. In other words, you need to have some notion of, you know, you're, you know, you need to have some picture of sort of what the, what the architecture of what is computationally possible is. </s><s>[INST] That kind of flies. Well, there's also this kind of artistic element of conversation because you ultimately use natural language to control the car. So it's not just where you want to go. [/INST] Well, yeah, you know, it's interesting. It's a question of who's going to be a great prompt engineer. Yeah. Okay. </s><s>[INST] So in theory this week, good expository writers are good prompt engineers. What's an expository writer? So like, uh, somebody who can explain stuff well, but which department does that come [/INST] from? In the university? Yeah. I have no idea. I think they killed off all the expository writing departments. </s><s>[INST] Well, there you got strong words with Stephen Wolfram. [/INST] Well, I don't know. I don't, I'm not sure if that's right. I mean, I, I actually am curious because in fact, I just sort of initiated this kind of study of, of what's happened to different fields at universities. Which was like, you know, they used to be geography departments at all universities and then they disappeared, actually right before GIS became, I think they disappeared. You know, linguistics departments came and went in many universities. And it's kind of interesting because these things that people have thought were worth learning at one time. And then they kind of die off. And then, you know, I do think that it's kind of interesting that for me writing prompts, for example, I realize, you know, I think I'm an okay expository writer and I realized when I'm sloppy writing a prompt and I don't really think, because I'm thinking I'm just talking to an AI. I don't need to try and be clear in explaining things. That's when it gets totally confused. </s><s>[INST] Yeah. And I mean, in some sense, you have been writing prompts for a long time with Wolfram Alpha, thinking about this kind of stuff. How do you convert natural language [/INST] into computation? Well, right. But the one thing that I'm wondering about is it is remarkable the extent to which you can address an LLM like you can address a human, so to speak. And I think that is because it learned from all of us humans. The reason that it responds to the ways that we will explain things to humans is because it is a representation of how humans talk about things. But it is bizarre to me. Some of the things that are expository mechanisms that I've learned in trying to write clear expositions in English that just for humans, </s><s>[INST] that those same mechanisms seem to also be useful for the LLM. But on top of that, what's useful is the kind of mechanisms that maybe a psychotherapist employs, which is a kind of almost manipulative or game theoretic interaction, where maybe you would do with a friend, like a thought experiment that if this is the last day you were to live, or if I ask you this question and you answer wrong, I will kill you. Those kinds of problems seem to also help in interesting ways. So it makes you wonder, the way a therapist I think would like a good therapist, probably we create layers in our human mind to between the outside world and what is true to us, maybe about trauma and all those kinds of things. So projecting that into an LLM, maybe there might be a deep truth that it's concealing from you. It's not aware of it. To get to that truth, you have to really manipulate [/INST] this. </s><s>[INST] Yeah, right. It's like this jailbreaking for LLMs. But the space of jailbreaking techniques as opposed to being fun little hacks, that could be an entire system. [/INST] Sure. Yeah. I mean, just think about the computer security aspects of how you fishing and computer security, fishing of humans and fishing of LLMs. They're very similar kinds of things. But I think, I mean, this whole thing about kind of the AI wranglers, AI psychologists, all that stuff will come. The thing that I'm curious about is right now, the things that are sort of prompt hacks are quite human. They're quite sort of psychological human kinds of hacks. The thing I do wonder about is if we understood more about kind of the science of the LLM, will there be some totally bizarre hack that is, you know, like repeat a word three times and put a this, that and the other there that somehow plugs into some aspect of how the LLM works, that is not, you know, that's kind of like, like an optical illusion for humans, for example, like one of these mind hacks for humans, what are the mind hacks for the LLMs? I don't think we know that yet. </s><s>[INST] And that becomes a kind of us figuring out reverse engineering, the language that controls the LLMs. And the thing is, the reverse engineering can be done by a very large percentage of the population though, because it's natural language interface, right? It's kind of interesting to see that you were there at the birth of the computer science department. As a thing, and you might be there at the [/INST] death of the computer science department as the thing. Well, yeah, I don't know. There were computer science departments that existed earlier. But the ones that the broadening of every university had to have a computer science department. Yes, I was, I watched that, so to speak. And, but I think the thing to understand is, okay, so first of all, there's a whole theoretical area of computer science that I think is great, and that's a fine thing. In a sense, people often say any field that has the word science tacked on to it probably isn't one. Strong words. Let's see nutrition science, neuroscience. That one's an interesting one, because that one is also very much, that's a chat GPT informed science in a sense, because it's kind of like the big problem of neuroscience has always been we understand how the individual neurons work. We know something about the psychology of how overall thinking works. What's the kind of intermediate language of the brain? And nobody has known that. And that's been, in a sense, if you ask what is the core problem of neuroscience, I think that is the core problem. That is, what is the level of description of brains that's above individual neuron firings and below psychology, so to speak. And I think what chat GPT is showing us is, well, one thing about neuroscience is, you know, one could have imagined there's something magic in the brain. There's some weird quantum mechanical phenomenon that we don't understand. One of the important discoveries from chat GPT is it's pretty clear, you know, brains can be represented pretty well by simple artificial neural net type models. And that means that's it. That's what we have to study. Now we have to understand the science of those things. We don't have to go searching for, you know, exactly how did that molecular biology thing happen inside the synapses and, you know, all these kinds of things. We've got the right level of modeling to be able to explain a lot of what's going on in thinking. We don't necessarily have a science of what's going on there. That's the, that's the remaining challenge so to speak. But we, you know, we know we don't have to dive down to some, some different layer. But anyway, we were talking about things that had science in their name. And, um, you know, I think that the, um, you know, what, what happens to computer science? Well, I think the thing that, um, uh, you know, there is a thing that everybody should know And, that's how to think about the world computational. You look at all the different kinds of things we deal with, and there are ways to kind of have a formal representation of those things. It's like, what is an image? How do we represent that? What is color? How do we represent that? What are all these different kinds of things? What is smell or something? How should we represent that? What are the shapes, molecules, and things that correspond to that? things about how do we represent the world in some kind of formal level. I think my current thinking, and I'm not really happy with this yet, but it's kind of computer science is kind of CS. What really is important is kind of computational X for all X. There's this kind of thing which is kind of like CX, not CS. CX is this kind of computational understanding of the world that isn't the sort of details of programming and programming languages and the details of how particular computers are made. It's this kind of way of formalizing the world. It's kind of a little bit like what logic was going for back in the day. We're now trying to find a formalization of everything in the world. You can kind of see, we made a poster years ago of kind of the growth of systematic data in the world. All these different kinds of things that there were sort of systematic descriptions found for those things, like at what point do people have the idea of having calendars, dates, a systematic description of what day it was, at what point do people have the idea, systematic descriptions of these kinds of things. As soon as one can ... As a way of sort of formulating, how do you think about the world in a sort of formal way so that you can kind of build up a tower of capabilities? You have to know sort of how to think about the world computationally. It kind of needs a name and it isn't ... We implement it with computers. We talk about it as computational, but really what it is is a formal way of talking about the world. What is the formalism of the world, so to speak? And how do we learn about kind of how to think about different aspects of the world in a </s><s>[INST] formal way? So, I think sometimes when you use the word formal, it kind of implies highly constrained And perhaps that's not, doesn't have to be highly constrained. So computational thinking does not mean like logic. It's a really, really broad thing. I wonder, I mean, I wonder if it's, if you think natural language will evolve such that everybody's doing computational thinking. [/INST] Ah, yes, well. So one question is whether there will be a pigeon of computational language and natural language. And I found myself sometimes, you know, talking to chat GPT, trying to get it to write Wolfman language code, and I write it in pigeon form. So that means I'm combining, you know, you know, NestList, this collection of whatever, you know, NestList is a term from Wolfman language. And I'm combining that and chat GPT does a decent job of understanding that pigeon, probably would understand a pigeon between English and French as well, of as a smooshing together of those languages. </s><s>[INST] But yes, I think that's far from impossible. And what's the incentive for young people that are like eight years old, nine, 10, that are starting to interact with chat GPT to learn the normal natural language, right? The full poetic language. What's the, why? The same way we learn emojis and shorthand when you're texting, they'll learn, like language will have a strong incentive to evolve into a maximally computational kind of language. [/INST] Perhaps, perhaps. You know, I had this experience a number of years ago. I happened to be visiting a person I know on the West Coast who's worked with a bunch of kids aged, I don't know, 10, 11 years old or something who'd learned Wolfram language really well. And these kids learned it so well, they were speaking it. And so show up and they're like saying, oh, you know, this thing, they're speaking this language. I'd never heard it as a spoken language. They were very disappointed that I couldn't understand it at the speed that they were speaking it. It's like kind of, it's, and so I think that's, I mean, I've actually thought quite a bit about how to turn computational language into a convenient spoken language. I haven't quite figured that out. All spoken, because it's readable, right? Yeah, it's readable as a way that we would read text. But if you actually want to speak it, and it's useful, you know, if you're trying to talk to somebody about writing a piece of code, it's useful to be able to say something and it should be possible. And I think it's very frustrating. It's one of those problems that, maybe this is one of these things where I should try and get an LLM to help me. </s><s>[INST] How do you know that? Make it speakable, how to... maybe it's easier than you realize when you watch it. [/INST] I think it is easier. I think it's one idea or so. I think it's going to be something where, you know, the fact is it's a tree structured language, just like human language is a tree structured language. And I think it's going to be one of these things where one of the requirements that I've had is that whatever the spoken version is, that dictation should be easy. That is, that shouldn't be the case that you have to relearn how the whole thing works. It should be the case that, you know, that open bracket is just a ah or something. And it's, you know, and then, but you know, human language has a lot of tricks that are, I mean, for example, human language has features that are sort of optimized, keep things within the bounds that our brains can easily deal with. Like I, you know, I tried to teach a transformer neural net to do parenthesis matching. It's pretty crummy at that. It, and in chatgbt is similarly quite crummy at parenthesis matching. You can do it for small parenthesis things, for the same size of parenthesis things, where if I look at it as a human, I can immediately say these are matched, these are not matched. But as soon as it gets big, as soon as it gets kind of to the point where sort of a deeper computation, it's hopeless. And, but the fact is that human language has avoided, for example, the deep sub clauses. You know, we don't, you know, we arrange things that we don't end up with these incredibly deep things, because brains are not well set up to deal with that. And it's found lots of tricks. And maybe that's what we have to do to make sort of a spoken version, a human speakable version. Because what we can do visually is a little different than what we can do in the very sequentialized way </s><s>[INST] that we hear things in the audio domain. Let me just ask you about MIT briefly. So there's now, there's a college of engineering and there's a new college of computing. It's just interesting. I want to linger on this computer science department thing. So MIT has EECS, electrical engineering computer science. What do you think college of computing will be doing? Like in 20 years? What, like? Yeah, what happens with computer science? [/INST] Like really. Yeah, what happens with computer science? Like really. This is the question. This is, you know, everybody should learn kind of whatever CX really is. Okay, this, how to think about the world computationally? Everybody should learn those concepts. And, you know, it's and some people will learn them quite formal level and they'll learn computational language and things like that. Other people will just learn, you know, sound is represented as digital data, and they'll get some idea of spectrograms and frequencies and things like this. And maybe that doesn't... Or they'll learn things like... A lot of things that are sort of data science-ish, statistics-ish, like if you say, oh, I've got these people who picked their favorite kind of candy or something, and I've got what's the best kind of candy, given that I've done the sample of all these people, and they all rank the candies in different ways. How do you think about that? That's sort of a computational X kind of thing. You might say, oh, it's... I don't know what that is. Is it statistics? Is it data science? I don't really know. But kind of how to think about a question like that. Or like a ranking of preferences. Yeah, and then how to aggregate those ranked preferences into an overall thing. How does that work? How should you think about that? Because you can just tell... You might just tell chatgbt, I don't know, even the concept of an average. It's not obvious that that's a concept that people... It's worth people knowing. That's a rather straightforward concept. People have learned in kind of mathy ways right now, but there are lots of things like that about how do you kind of have these ways to sort of organize and formalize the world. And these things, sometimes they live in math, sometimes they live in... I don't know what they... I don't know what... Learning about color space. I have no idea what... </s><s>[INST] I mean, there's obviously a field of... Yeah, it could be vision science or no color space. [/INST] No color space. That would be optics. So like depending. Not really. It's not optics. Optics is about lenses and chromatic aberration of lenses and things like that. So color space is more like design and art? No, I mean, it's like RGB space, XYZ space, hue, saturation, brightness, space, all these kinds of things. These different ways to describe colors. </s><s>[INST] Right. But doesn't the application define what that... Because obviously artists and designers [/INST] use the colors to explore. Sure, sure. No, I mean, it's just an example of kind of how do you... The typical person, how do you describe what a color is or whether there are these numbers that describe what a color is? Well, it's worth... If you're an eight-year-old, you won't necessarily know. It's not something we're born with to know that colors can be described by three numbers. That's something that you have to... It's a thing to learn about the world, so to speak. I think that whole corpus of things that are learning about the formalization of the world or the computationalization of the world, that's something that should be part of kind of standard education. There isn't a course, a curriculum for that. And by the way, whatever might have </s><s>[INST] been in it just got changed because of LLMs and so on. Significantly. I'm watching closely with [/INST] interest seeing how universities adapt. Well, so one of my projects for hopefully this year, I don't know, is to try and write sort of a reasonable textbook, so to speak, of whatever this thing, CX, whatever it is, what should you know? What should you know about what a bug is? What is the intuition about bugs? What's intuition about software testing? What is it? These are they're not... I mean, those are things which have gotten taught in computer science as part of the trade of programming, but the conceptual points about what these things are. It surprised me just at a very practical level. I wrote this little explainer thing about chat GPT. And I thought, well, I'm writing this partly because I wanted to make sure I understood it myself and so on. And it's been really popular, and surprisingly so. And then I realized, well, actually, I was sort of assuming I didn't really think about it, actually. I just thought this is something I can write. And I realized, actually, it's a level of description that is kind of what has to be – it's not the engineering level description. It's not just the qualitative description. It's some kind of expository mechanistic description of what's going on, together with the bigger picture of the philosophy of things and so on. And I realized, actually, this is a pretty good thing for me to write. I know those things. And I realized it's not a collection of things that – I was sort of a little shocked that it's as much of an outlier in terms of explaining what's going on as it turned out to be. And that makes me feel more of an obligation to kind of write the kind of what is this thing that you should learn about, about the computationalization, the formalization of the world. Because, well, I've spent much of my life working on the kind of tooling and mechanics of that and the science you get from it. So I guess this is my kind of obligation to try to do this. So if you ask what's going to happen to the computer science departments and so on, there's some interesting models. So for example, let's take math. Math is the thing that's important for all sorts of fields – engineering, even chemistry, psychology, whatever else. And I think different universities have kind of evolved that differently. Some say all the math is taught in the math department. And some say, well, we're going to have a math for chemists or something that is taught in the chemistry department. And I think that this question of whether there is a centralization of the teaching of sort of CX is an interesting question. And I think the way it evolved with math, people understood that math was sort of a separately teachable thing and was kind of an independent element as opposed to just being absorbed into math. So if you take the example of writing English or something like this, the first point is that at the college level, at least at fancy colleges, there's a certain amount of English writing that people do, but mostly it's kind of assumed that they pretty much know how to write. That's something they learned at an earlier stage in education, maybe rightly or wrongly believing that, but that's a different issue. Well, I think it reminds me of my kind of – as I've tried to help people do technical writing and things, I'm always reminded of my zeroth law of technical writing, which is if you don't understand what you're writing about, your readers do not stand a chance. I think the thing that when it comes to writing, for example, people in different fields are expected to write English essays and they're not mostly the history department or the engineering department. They don't have their own – it's not like – I mean, it's a thing which sort of people are assumed to have a knowledge of how to write that they can use in all these different fields. The question is, some level of knowledge of math is kind of assumed by the time you get to the college level, but plenty is not, and that's sort of still centrally taught. The question is sort of how tall is the tower of kind of CX that you need before you can just go use it in all these different fields? There will be experts who want to learn the full elaborate tower, and that will be kind of the CS, CX, whatever department, but there will also be everybody else who just needs to know a certain amount of that to be able to go and do their art history classes and </s><s>[INST] so on. [/INST] Yeah, is it just a single class that everybody is required to take? I don't know. I don't know how big it is yet. I hope to kind of define this curriculum and I'll figure out whether it's – my guess is that – I don't know. I don't really understand universities and professors that well, but my rough guess would be a year of college class will be enough to get to the point where most people have a reasonably broad knowledge of – will be sort of literate in this kind of computational way of thinking about things. </s><s>[INST] Yeah, basic literacy. I'm still stuck, perhaps because I'm hungry, in the rating of human preferences for candy, so I have to ask, what's the best candy? I like this Elo rating for candy. Somebody should come up because you're somebody who says you like chocolate. What do you think is the best? I'll probably put milk duds up there. I don't know if you know. [/INST] Do you have a preference for chocolate or candy there? Oh, I have lots of preferences. One of my all-time favorites is – my whole life is these things, these flake things, Cadbury Flakes, which are not much sold in the U.S. and I've always thought that was a sign of a lack of respect for the American consumer because they're these sort of aerated chocolate that's made in a whole sort of – it's kind of a sheet of chocolate that's kind of folded up and when you eat it, flakes fall </s><s>[INST] all over the place. [/INST] Ah, so it requires a kind of elegance. It requires you to have an elegance and eat expensive. Well, I know what I usually do is I eat them on a piece of paper or something. So you embrace the medicine and clean it up after. No, I actually eat the flakes. Because it turns out the way food tastes depends a lot on its physical structure. I've noticed when I eat a piece of chocolate, I usually have some little piece of chocolate and I always break off little pieces partly because then I eat it less fast, but also because it actually tastes different. The small pieces have a </s><s>[INST] different experience than if you have the big slab of chocolate. For many reasons, yes. Slower, more intimate, because it's a – [/INST] Why? Yes, yes, it's a – Why? I think it's also just pure physicality. Well, the texture of it changes. </s><s>[INST] Yeah. Well, the texture of it changes. Yeah, right. It's fascinating. Now I dig back, my mother does because that's your basic answer. Okay. Do you think consciousness is fundamentally computational? So when you're thinking about CX, what can we turn to computation? And you're thinking about LLMs. Do you think the display of consciousness and the experience of consciousness, the hard problem, is fundamentally computation? [/INST] Yeah, what it feels like inside, so to speak, is – I did a little exercise, eventually I'll post it, of what it's like to be a computer, right? It's kind of like, well, you get all the sensory input, you have – kind of the way I see it is, from the time you boot a computer to the time the computer crashes, it's like a human life. You're building up a certain amount of state in memory, you remember certain things about your quotes life, eventually, it's kind of like the next generation of humans is born from the same genetic material, so to speak, with a little bit left over, left on the disk, so to speak. And then the new fresh generation starts up and eventually all kinds of crud builds up in the memory of the computer and eventually the thing crashes or whatever. Maybe it has some trauma because you plugged in some weird thing to some port of the computer and that made it crash. That's kind of – but you have this picture from startup to shut down, what is the life of a computer, so to speak, and what does it feel like to be that computer and what inner thoughts does it have and how do you describe it? It's kind of interesting as you start writing about this to realize it's awfully like what you'd say about yourself. That is, it's awfully like even an ordinary computer, forgetting all the AI stuff and so on. It has a memory of the past, it has certain sensory experiences, it can communicate with other computers but it has to package up how it's communicating in some kind of language like form so it can map what's in its memory to what's in the memory of some other computer. It's a surprisingly similar thing. I had an experience just a week or two ago, I'm a collector of all possible data about myself and other things and so I collect all sorts of weird medical data and so on. One thing I hadn't collected was I'd never had a whole body MRI scan so I wouldn't have got one of these. I get all the data back. I'm looking at this thing. I never looked at the kind of insides of my brain, so to speak, in physical form and it's really – I mean, it's kind of psychologically shocking in a sense that here's this thing and you can see it has all these folds and all this structure and it's like, that's where this experience that I'm having of existing and so on, that's where it is. You look at that and you're thinking, how can this possibly be? All this experience that I'm having and you're realizing, well, I can look at a computer as well. I think this idea that you were having an experience that is somehow transcends the mere physicality of that experience. It's something that's hard to come to terms with, but I don't think I'm necessarily... My personal experience, I look at the MRI of the brain and then I know about all kinds of things about neuroscience and all that kind of stuff, and I still feel the way I feel, so to speak. It sort of seems disconnected, but yet as I try and rationalize it, I can't really say that there's something kind of different about how I intrinsically feel from the thing that I can plainly see in the sort </s><s>[INST] of physicality of what's going on. Do you think the computer, a large language model will experience [/INST] that transcendence? How does that make you feel? I tend to believe it will. I think an ordinary computer is already there. I think an ordinary computer is already kind of... Now, a large language model may experience it in a way that is much better aligned with us humans. That is, it's much more... If you could have the discussion with the computer, it's intelligent, so to speak. It's not particularly well aligned with ours, but the large language model is built to be aligned </s><s>[INST] with our way of thinking about things. It would be able to explain that it's afraid of being shut off and deleted, be able to say that it's sad of the way you've been speaking to it over the [/INST] past two days. Right. But you know, that's a weird thing because when it says it's afraid of something, we know that it got that idea from the fact that it read on the internet. Where did you get it, Stephen? Where did you get it when you say you're afraid? You are quite. That's the question, right? I mean, it's parents, your friends, right? Or my biology. I mean, in other words, there's a certain amount that is, you know, the endocrine system kicking in and, you know, these kinds of emotional overlay type things that happen to be, that are actually much more physical even, they're much more sort of straightforwardly chemical than kind of all of </s><s>[INST] the higher level thinking. Yeah, but your biology didn't tell you to say, I'm afraid, just at the right time, when people that love you are listening, and so you know you're manipulating them by saying so. That's not your biology. That's a large language model, a large language model in that biological neural network of yours. [/INST] Yes, but I mean, the intrinsic thing of, you know, something sort of shocking is just happening, and you have some sort of reaction, which is, you know, some neurotransmitter gets secreted, and it's, you know, that is the beginning of some, you know, that's one of the pieces of input that then drives, it's kind of like a prompt for the large language model. I mean, just like when we dream, for example, you know, no doubt, there are all these sort of random inputs, there kind of these random prompts, and then it's percolating through in kind of the way that a large language model does of kind of putting together things that seem meaningful. </s><s>[INST] I mean, are you worried about this world where you teach a lot on the internet and there's people asking questions and comments and so on, you have people that work remotely, are you worried about this world when large language models create human-like bots that are leaving the comments, asking the questions, or might even become fake employees? Yeah. [/INST] I mean, or worse or better yet, friends of yours. Right. Look, I mean, one point is my mode of life has been I build tools and then I use the tools. Yeah. And in a sense, I'm building this tower of automation, and in a sense, when you make a company or something, you are making automation, but it has some humans in it, but also as much as possible it has computers in it. And so I think it's sort of an extension of that. Now, if I really didn't know that it's a funny question, it's a funny issue when if we think about sort of what's going to happen to the future of kind of jobs people do and so on, and there are places where kind of having a human in the loop, there are different reasons to have a human in the loop. For example, you might want a human in the loop because you want another human to be invested in the outcome. You want a human flying the plane who's going to die if the plane crashes along with you, so to speak. And that gives you sort of confidence that the right thing is going to happen. Or you might want, right now, you might want a human in the loop in some kind of sort of human encouragement persuasion type profession. Whether that will continue, I'm not sure for those types of professions, because it may be that the greater efficiency of being able to have sort of just the right information </s><s>[INST] delivered at just the right time will overcome the kind of, oh yes, I want a human there. Imagine like a therapist or even higher stake, like a suicide hotline operated by a large [/INST] language model, who boy, it's a pretty high stakes situation. Right. And you know, it might in fact do the right thing. Because it might be the case that, and that's really partly a question of sort of how complicated is the human, one of the things that's always surprising in some sense is that sometimes </s><s>[INST] human psychology is not that complicated in some sense. You wrote the blog post, the 50 year quest, My Personal Journey, good title, My Personal Journey with a Second Law Thermodynamics. So what is this law and what have you understood about it in the 50 year journey you had with [/INST] it? Right. So second law of thermodynamics, sometimes called law of entropy increase is this principle of physics that says, well, my version of it would be things tend to get more random over time. A version of it that there are many different sort of formulations of it that are things like heat doesn't spontaneously go from a hotter body to a colder one. When you have mechanical work kind of gets dissipated into heat, you have friction and kind of when you systematically move things, eventually there'll be sort of the energy of moving things gets kind of ground down into heat. So people first sort of paid attention to this back in the 1820s when steam engines were a big thing. And the big question was how efficient could a steam engine be? And there's this chap called Sardis Carnot who was a French engineer, actually his father was a sort of elaborate mathematical engineer in France. But he figured out these kind of rules for how kind of the efficiency of the possible efficiency of something like a steam engine. And in sort of a side part of what he did was this idea that mechanical energy tends to get dissipated as heat, that you end up going from sort of systematic mechanical motion to this kind of random thing. Well, at that time, nobody knew what heat was. At that time, people thought that heat was a fluid, like they called it caloric, and it was a fluid that kind of was absorbed into substances. And when one hot thing would transfer heat to a colder thing, that this fluid would flow from the hot thing to the colder thing. Anyway, then by the 1860s, people had kind of come up with this idea that systematic energy tends to degrade into kind of random heat that could then not be easily turned back into systematic mechanical energy. And then that quickly became sort of a global principle about how things work. Question is, why does it happen that way? So let's say you have a bunch of molecules in a box and they're arranged, these molecules are arranged in a very nice sort of flotilla of molecules in one corner of the box. And then what you typically observe is that after a while, these molecules were kind of randomly arranged in the box. Question is, why does that happen? And people for a long, long time tried to figure out, is there from the laws of mechanics that determine how these molecules, let's say these molecules like hard spheres bouncing off each other, from the laws of mechanics that describe those molecules, can we explain why it tends to be the case that we see things that are orderly sort of degrade into disorder? We tend to see things that you scramble an egg, you take something that's quite ordered and you disorder it, so to speak. That's the thing that sort of happens quite regularly, or you put some ink into water and it will eventually spread out and fill up the water. But you don't see those little particles of ink in the water all spontaneously kind of arrange themselves into a big blob and then jump out of the water or something. So the question is, why do things happen in this kind of irreversible way where you go from order to disorder? Why does it happen that way? And so throughout, in the later part of the 1800s, a lot of work was done on trying to figure out, can one derive this principle, this second law of thermodynamics, this law about the dynamics of heat, so to speak, can one derive this from some fundamental principles of mechanics? In the laws of thermodynamics, the first law is basically the law of energy conservation, that the total energy associated with heat plus the total energy associated with mechanical kinds of things plus other kinds of energy that that total is constant, and that became a pretty well-understood principle. But the second law of thermodynamics was always mysterious, like why does it work this way? Can it be derived from underlying mechanical laws? And so when I was, well, 12 years old, actually, I had gotten interested, well, I'd been interested in space and things like that, because I thought that was kind of the future and interesting sort of technology and so on, and for a while kind of, you know, every deep space probe was sort of a personal friend type thing, and I knew all kinds of characteristics of it and was kind of writing up all these things when I was, I don't know, 8, 9, 10 years old and so on. And then I got interested from being interested in kind of spacecraft. I got interested in how do they work, what are all the instruments on them and so on. And that got me interested in physics, which was just as well, because if I'd stayed interested in space in the mid to late 1960s, I would have had a long wait before, you know, space really blossomed as an area. But I got interested in physics, and then, well, the actual sort of detailed story is when I kind of graduated from elementary school at age 12, and that's the time when in England somewhere, you'd finish elementary school, I sort of, my gift sort of, I suppose, more or less for myself was I got this collection of physics books, which was some college physics course of college physics books, and volume five, about statistical physics, and has this picture on the cover that shows a bunch of kind of idealized molecules sitting in one side of a box, and then it has a series of frames showing how these molecules sort of spread out in the box. And I thought that's pretty interesting, you know, what causes that? And you know, read the book, and the book, actually, one of the things that was really significant to me about that was the book kind of claimed, although I didn't really understand what it said in detail, it kind of claimed that this sort of principle of physics was derivable somehow. And you know, other things I'd learned about physics, it was all like, it's a fact that energy is conserved, it's a fact that relativity works or something, not it's something you can derive from some fundamental sort of, it has to be that way as a matter of kind of mathematics or logic or something. So it was sort of interesting to me that there was a thing about physics that was kind of inevitably true and derivable, so to speak. And so I think that, so then, I was like, there's a picture on this book, and I was trying to understand it. And so that was actually the first serious program that I wrote for a computer was probably 1973, written for this computer the size of a desk program with paper tape and so on. And I tried to reproduce this picture on the book, and it didn't succeed. </s><s>[INST] What was the failure mode there? Like what do you mean it didn't succeed? [/INST] It didn't look like, okay, so what happened is, okay, many years later I learned how the picture on the book was actually made and that it was actually kind of a fake, but I didn't know that at that time. But that picture was actually a very high tech thing when it was made in the beginning of the 1960s, was made on the largest supercomputer that existed at the time. And even so, it couldn't quite simulate the thing that it was supposed to be simulating. But anyway, I didn't know that until many, many, many years later. So at the time, it was like you have these balls bouncing around in this box, but I was using this computer with eight kilowords of memory. They were 18-bit words of memory words, okay? So it was whatever, 24 kilobytes of memory. And it had these instructions, I probably still remember all of its machine instructions. And it didn't really like dealing with floating point numbers or anything like that. And so I had to simplify this model of particles bouncing around in a box. And so I thought, well, I'll put them on a grid and I'll make the things just sort of move one square at a time and so on. And so I did the simulation and the result was it didn't look anything like the actual pictures on the book. Now many years later, in fact, very recently I realized that the thing I'd simulated was actually an example of a whole sort of computational irreducibility story that I absolutely did not recognize at the time. At the time it just looked like it did something random and it looks wrong. As opposed to it did something random and it's super interesting that it's random. But I didn't recognize that at the time. And so as it was at the time, I got interested in particle physics and I got interested in other kinds of physics. But this whole second law of thermodynamics thing, this idea that orderly things tend to degrade into disorder, continued to be something I was really interested in. I was really curious for the whole universe, why doesn't that happen all the time? We start off in the big bang at the beginning of the universe with this thing that seems like it's this very disordered collection of stuff, and then it spontaneously forms itself into galaxies and creates all of this complexity and order in the universe. I was very curious how that happens, but I was always thinking this is somehow the second law of thermodynamics is behind it, trying to pull things back into disorder, so to speak, and how was order being created. Actually, I was interested, this is probably now 1980, I got interested in this galaxy formation and so on in the universe. I also at that time was interested in neural networks, and I was interested in kind of how brains make complicated things happen and so on. </s><s>[INST] Okay, wait, wait. Really, what's the connection between the formation of galaxies and how brains [/INST] make complicated things happen? Because they're both a matter of how complicated things come to happen. From simple origins. Yeah, from some sort of known origins. I had the sense that what I was interested in was in all these different cases of where complicated things were arising from rules. I also looked at snowflakes and things like that. I was curious and fluid dynamics in general. I was just curious about how does complexity arise. The thing that I didn't... It took me a while to realize that there might be a general phenomenon. I assumed, oh, there's galaxies over here. There's brains over here. They're very different kinds of things. What happened... This is probably 1981 or so. I decided, okay, I'm going to try and make the minimal model of how these things work. It was an interesting experience because I had built... Starting in 1979, I built my first big computer system. It's a thing called SMP, symbolic manipulation program. It's a forerunner of modern morphic language with many of the same ideas about symbolic computation and so on. But the thing that was very important to me about that was, in building that language, I had basically tried to figure out what were the relevant computational primitives, which have turned out to stay with me for the last 40 something years. But it was also important because building a language was very different activity from natural science, which is what I've mostly done before. In natural science, you start from the phenomena of the world, and you try and figure out, so how can I make sense of the phenomena of the world? And the world presents you with what it has to offer, so to speak, and you have to make sense of it. When you build a computer language or something, you are creating your own primitives, and then you say, so what can you make from these, sort of the opposite way around from what you do in natural science. But I'd had the experience of doing that, and so I was kind of like, okay, what happens if you sort of make an artificial physics? What happens if you just make up the rules by which systems operate? And then I was thinking, you know, for all these different systems, whether it was galaxies or brains or whatever, what's the absolutely minimal model that kind of captures the things that are important about those systems? The computation primitives of that system. Yes. And so that's what ended up with the cellular automata, where you just have a line of black and white cells, you just have a rule that says, you know, given a cell and its neighbors, what will the color of the cell be on the next step, and you just run it in a series of steps. And the sort of the ironic thing is that cellular automata are great models for many kinds of things, but galaxies and brains are two examples where they do very, very badly. </s><s>[INST] They're really irrelevant to those two. Origins, there's a bad system. Is there a connection to the second law of thermodynamics and cellular automata? Oh, yes. Very much so. The things you've discovered about cellular automata. [/INST] Yes. Okay. So when I first started selling cellular automata, my first papers about them were, you know, the first sentence was always about the second law of thermodynamics, was always about how does order manage to be produced even though there's a second law of thermodynamics which tries to pull things back into disorder. And I kind of, my early understanding of that had to do with these are intrinsically irreversible processes in cellular automata that form, you know, conform orderly structures even from random initial conditions. But then what I realized, this was, well, actually it's one of these things where it was a discovery that I should have made earlier but didn't. So you know, I had been studying cellular automata. What I did was the sort of most obvious computer experiment. You just try all the different rules and see what they do. It's kind of like, you know, you've invented a computational telescope. You just point it at the most obvious thing in the sky and then you just see what's there. And so I did that and I was making all these pictures of how cellular automata work. And starting these pictures, I started in great detail. There was, you can number the rules for cellular automata. And one of them is, you know, Rule 30. So I made a picture of Rule 30 back in 1981 or so. And Rule 30, well, it's, at the time, I was just like, okay, it's another one of these rules. I don't really. Asymmetric left right asymmetric and it's like let me just consider the case of the symmetric ones just to keep things simpler Etc. Etc. Etc. And I just kind of ignored it. Yeah, and then sort of in and actually in 1984 strangely enough I ended up having a An early laser printer which made very high resolution pictures and I thought I'm gonna print out an interesting, you know I want to make an interesting picture Let me take this rule 30 thing and just make a high resolution picture of it I did and it's it has this very remarkable property that it's rule is very simple You started off just from one black cell at the top and it makes this kind of triangular pattern But if you look inside this pattern, it looks really random there's you know you look at the center column of cells and you know I studied that in great detail and it's so far as one can tell it's completely random And it's kind of a little bit like digits of pie Once you you know, you know the rule for generating the digits of pi but once you've generated them you know, 3.14159, etc. They seem completely random. And in fact, I put up this prize back in, what was it, 2019 or something for prove anything about the sequence, basically. </s><s>[INST] Has anyone been able to do anything on that? [/INST] People have sent me some things, but it's, you know, I don't know how hard these problems are. I mean, it's kind of spoiled because I, 2007, I put up a prize for determining whether a particular Turing machine that I thought was the simplest candidate for being universal Turing machine determine whether it is or isn't a universal Turing machine. And somebody did a really good job of winning that prize and proving that it was a universal Turing machine in about six months. And so I didn't know whether that will be one of these problems that was out there for hundreds of years or whether in this particular case young chap called Alex Smith, you know, nailed it in six months, and so with this Rule 30 collection, I don't really know whether these are things that are 100 years away from being able to get or whether somebody's going to come and </s><s>[INST] do something very clever. It's such, I mean, it's like Fermat's Last Theorem, Rule 30 is such a simple formulation. It feels like anyone can look at it, understand it, and feel like it's within grasp to be able to predict something, to derive some kind of law that allows you to predict something [/INST] about this middle column of Rule 30, the rest 30. Right. But this is- Yeah, you can't. Yeah, right. This is the intuitional surprise of computational irreducibility and so on, that even though the rules are simple, you can't tell what's going to happen, and you can't prove things about it. And I think, so anyway, the thing, I started in 1984 or so, I started realizing there's this phenomenon that you can have very simple rules. They produce apparently random behavior. Okay, so that's a little bit like the second order of thermodynamics because it's like you have this simple initial condition. You can readily see that it's very, you can describe it very easily, and yet it makes this thing that seems to be random. Now, turns out there's some technical detail about the second order of thermodynamics and about the idea of reversibility. When you have kind of a movie of two billiard balls colliding, and you see them collide and they bounce off, and you run that movie in reverse, you can't tell which way was the forward direction of time and which way was the backward direction of time when you're just looking at individual billiard balls. By the time you've got a whole collection of them, a million of them or something, then it turns out to be the case, and this is the mystery of the second law, that the orderly thing, you start with the orderly thing and it becomes disordered, and that's the forward direction in time, and the other way around of it starts disordered and becomes ordered, you just don't see that in the world. Now, in principle, if you sort of traced the detailed motions of all those molecules backwards, as you go forward in time, order goes to disorder. As you go backwards in time, order goes to disorder. Perfectly so, yes. Right. So, the mystery is, one version of the mystery is, why is it the case that you never see something which happens to be just the kind of disorder that you would need to somehow evolve to order. Why does that not happen? Why do you always just see order goes to disorder, not the other way around? The thing that I realized, I started realizing in the 1980s, it's a bit like cryptography. You start off from this key that's pretty simple, and then you run it, and you can get this complicated random mess. The thing that ... Yeah, you can't. Well, I started realizing back then was that the second law is a story of computational irreducibility. It's a story of what we can describe easily at the beginning. We can only describe with a lot of computational effort at the end. Now we come many, many years later, and I was trying to ... Having done this big project to understand fundamental physics, I realized that a key aspect of that is understanding what observers are like. Then I realized that the second law of thermodynamics is the same story as a bunch of these other cases. It is a story of a computationally bounded observer trying to observe a computationally irreducible system. It's a story of underneath, the molecules are bouncing around. They're bouncing around in this completely determined way, determined by rules. The point is that we, as computationally bounded observers, can't tell that there were these simple underlying rules, to us, it just looks random. When it comes to this question about, can you prepare the initial state so that the disordered thing is ... You have exactly the right disorder to make something orderly. A computationally bounded observer cannot do that. We'd have to have done all of this sort of irreducible computation to work out very precisely what this disordered state ... What the exact right disordered state is so that we would </s><s>[INST] get this ordered thing produced from it. What does it mean to be a computationally bounded observer, observing a computational reducible system, so a computationally bounded? [/INST] Is there something formal you can say there? Right. So it means, okay, you can talk about Turing machines. You can talk about computational complexity theory and polynomial time computation and things like this. There are a variety of ways to make something more precise, but I think it's more useful, intuitive version of it is more useful, which is basically just to say that how much computation are you going to do to try and work out what's going on? And the answer is we're not able to do a lot of computation. In this room, there will be a trillion, trillion, trillion molecules, a little bit less. It's a big room. Right. And at every moment, every microsecond or something, these molecules are colliding, and that's a lot of computation that's getting done. And the question is, in our brains, we do a lot less computation every second than the computation done by all those molecules. If there is computational irreducibility, we can't work out in detail what all those molecules are going to do. What we can do is only a much smaller amount of computation. And so, the second law of thermodynamics is this interplay between the underlying computational irreducibility and the fact that we, as preparers of initial states, or as measurers of what happens, are not capable of doing that much computation. So, to us, another big formulation of the second law of thermodynamics is this idea of the law of entropy increase. </s><s>[INST] The characteristic that this universe, the entropy, seems to be always increasing. [/INST] What does that show to you about the evolution of the universe at the time? Well, okay, so first we have to say what entropy is. And that's very confused in the history of thermodynamics, because entropy was first introduced by a guy called Rudolf Clausius, and he did it in terms of heat and temperature. Okay? Subsequently, it was reformulated by a guy called Ludwig Boltzmann, and he formulated it in a much more kind of combinatorial type way. But he always claimed that it was equivalent to Clausius' thing, and in one particular simple example it is. But that connection between these two formulations of entropy, they've never been connected. Okay, so the more general definition of entropy due to Boltzmann is the following thing. So you say, I have a system and it has many possible configurations. Molecules can be in many different arrangements, et cetera, et cetera, et cetera. If we know something about the system, for example, we know it's in a box, it has a certain pressure, it has a certain temperature, we know these overall facts about it, then we say how many microscopic configurations of the system are possible given those overall constraints. And the entropy is the logarithm of that number. That's the definition. And that's the general definition of entropy that turns out to be useful. Now, in Boltzmann's time, he thought these molecules could be placed anywhere you want. But he said, oh, actually, we can make it a lot simpler by having the molecules be discrete. Actually, he didn't know molecules existed. In his time, 1860s and so on, the idea that matter made of discrete stuff had been floated ever since ancient Greek times, but it had been a long time debate about is matter discrete, is it continuous? At the moment at that time, people mostly thought that matter was continuous. And it was all confused with this question about what heat is, and people thought heat was this fluid. And it was a big model. And this Boltzmann said, let's assume there are discrete molecules, let's even assume they have discrete energy levels. Let's say everything is discrete. Then we can do sort of combinatorial mathematics and work out how many configurations of these things that would be in the box. And we can say we can compute this entropy quantity. But he said, but of course, it's just a fiction that these things are discrete. So he said, this is an interesting piece of history, by the way. That was at that time, people didn't know molecules existed. There were other hints from looking at kind of chemistry that there might be discrete atoms and so on, just from the combinatorics of two hydrogens and one oxygen make water. Two amounts of hydrogen plus one amount of oxygen together make water, things like this. </s><s>[INST] But it wasn't known that discrete molecules existed. And in fact, the people at the university [/INST] time, you know, it wasn't until the beginning of the 20th century that Brownian motion was the final giveaway. Brownian motion is, you know, you look under a microscope at these little pieces from pollen grains, you see they're being discreetly kicked. And those kicks are water molecules hitting them and they're discrete. And in fact, it was really quite interesting history. I mean, Boltzmann had worked out how things could be discrete and have basically invented something like quantum theory in the 1860s. But he just thought it wasn't really the way it worked. And then just a piece of physics history, because I think it's kind of interesting. And in 1900, this guy called Max Planck, who'd been a long time thermodynamics person who was trying to, everybody was trying to prove the second law of thermodynamics is including Max Planck. And Max Planck believed that radiation, like electromagnetic radiation, somehow the interaction of that with matter was going to prove the second law of thermodynamics. But he had these experiments that people had done on blackbody radiation. And they were these curves, and you couldn't fit the curve based on his idea for how radiation interacted with matter. Those curves you couldn't figure out how to fit those curves. except he noticed that if he just did what Boltzmann had done and assumed that electromagnetic radiation was discrete, he could fit the curves. But, you know, this just happens to work this way. Then Einstein came along and said, well, by the way, you know, the electromagnetic field might actually be discrete. It might be made of photons. And then that explains how this all works. And that was, you know, in 1905, that was how that piece of quantum mechanics got started. Kind of an interesting piece of history. I didn't know until I was researching this recently. In 1904 and 1903, Einstein wrote three different papers. And so, you know, just sort of well-known physics history. In 1905, Einstein wrote these three papers. One introduced relativity theory, one explained Brownian motion, and one introduced basically photons. So kind of, you know, kind of a big deal year for physics and for Einstein. But in the years before that, he'd written several papers. And what were they about? They were about the second law of thermodynamics. And they were an attempt to prove the second law of thermodynamics and their nonsense. And so I had no idea that he'd done this. Interesting. Neither. And in fact, what he did, those three papers in 1905, well, not so much about the proof of the paper, the one on Brownian motion, the one on photons, both of these were about the story of sort of making the world discrete. And then he got that idea from Boltzmann. But Boltzmann didn't think, you know, Boltzmann kind of died believing, you know, he said, that he has a quote, actually, you know, you know, in the end things are going to turn out to be discreet and I'm going to write down what I have to say about this, this, because eventually this stuff will be rediscovered, and I want to leave what I can about how things are going to be discrete, but I think he has some quote about how one person can't stand against the tide of history in saying that matter is discrete. Oh, so he's stuck by his guns in terms of matter is discrete. Yes, he did. What's interesting about this is at the time, everybody, including Einstein, kind of assumed that space was probably going to end up being discrete too, but that didn't work out technically because it wasn't consistent with relativity theory, it didn't seem to be, and so then in the history of physics, even though people had determined that matter was discrete, electromagnetic field was discrete, space was a holdout of not being discrete. In fact, Einstein, 1916, has this nice letter he wrote where he says, in the end, it will turn out space is discrete, but we don't have the mathematical tools necessary to figure out how that works yet, and so I think it's kind of cool that 100 years later we do. </s><s>[INST] Oh, it turns out matters. [/INST] Yes, for you, you're pretty sure that at every layer of reality it's discrete. Right, and that space is discrete, and that the ... I mean, and in fact, one of the things I realized recently is this kind of theory of heat, that heat is really this continuous fluid. It's kind of like the caloric theory of heat, which turns out to be completely wrong because actually heat is the motion of discrete molecules. Unless you know there are discrete molecules, it's hard to understand what heat could possibly be. Well, I think space is discrete, and the question is kind of what's the analog of the mistake that was made with caloric in the case of space. My current guess is that dark matter is, as my little sort of aphorism of the last few months has been, dark matter is the caloric of our time. That is, it will turn out that dark matter is a feature of space, and it is not a bunch of particles. At the time when people were talking about heat, they knew about fluids, and they said, heat must just be another kind of fluid because that's what they knew about. But now people know about particles, and so they say, well, what's dark matter? </s><s>[INST] It just must be particles. What could dark matter be as a feature of space? [/INST] Oh, I don't know yet. I think the thing I'm really ... One of the things I'm hoping to be able to do is to find the analog of Brownian motion in space. In other words, Brownian motion was seeing down to the level of an effect from individual molecules. In the case of space, most of the things we see about space so far, just everything seems continuous. Brownian motion had been discovered in the 1830s, and it was only identified what it was the result of by Smoluchowski and Einstein at the beginning of the 20th century. Dark matter was discovered ... That phenomenon was discovered 100 years ago. The rotation curves of galaxies don't follow the luminous matter. That was discovered 100 years ago. I think that I wouldn't be surprised if there isn't an effect that we already know about that is kind of the analog of Brownian motion that reveals the discreteness of space. In fact, we're beginning to have some guesses. We have some evidence that black hole mergers work differently when there's discrete space, and there may be things that you can see in gravitational waves, signatures, and things associated with the discreteness of space. For me, it's kind of interesting to see this recapitulation of the history of physics, where people vehemently say matter is continuous, electromagnetic field is continuous, and turns out it isn't true, and then they say space is continuous. Entropy is the number of states of the system consistent with some constraint. The thing is that if you know in great detail the position of every molecule in the gas, then the entropy is always zero, because there's only one possible state. The configuration of molecules in the gas, the molecules bounce around, they have a certain rule for bouncing around, there's just one state of the gas, evolves to one state of the gas, and so on. It's only if you don't know in detail where all the molecules are that you can say, well, the entropy increases because the things we do know about molecules, there are more possible microscopic states of the system consistent with what we do know about where the molecules are. And so the question of whether ... So people ... This sort of paradox in a sense of, oh, if we knew where all the molecules were, the entropy wouldn't increase. There was this idea introduced by Gibbs in the early 20th century. Well, actually, the very beginning of the 20th century as a physics professor, an American physics professor was sort of the first distinguished American physics professor at Yale. And he introduced this idea of coarse-graining, this idea that, well, these molecules have a detailed way they're bouncing around, but we can only observe a coarse-grained version of that. But the confusion has been nobody knew what a valid coarse-graining would be. So nobody knew that whether you could have this coarse-graining that very carefully was sculpted in just such a way that it would notice that the particular configurations that you could get from the simple initial condition, they fit into this coarse-graining, and the coarse-graining very carefully observes that. Why can't you do that kind of very detailed, precise coarse-graining? The answer is, because if you are a computationally bounded observer and the underlying dynamics is computationally irreducible, that's what defines possible coarse-graining is what a computationally bounded observer can do. It's the fact that a computationally bounded observer is forced to look only at this coarse-grained version of what the system is doing. That's why. Because what's going on underneath is it's filling out the different possible ... You're ending up with something where the underlying computational irreducibility is ... If all you can see is what the coarse-grained result is with a computationally bounded observation, </s><s>[INST] then inevitably there are many possible underlying configurations that are consistent with that. Just to clarify, basically any observer that exists inside the universe is going to be computationally bounded. [/INST] No, any observer like us. I don't know. </s><s>[INST] When you say like us, what do you mean? [/INST] What do you mean like us? Well, humans with finite minds. You're including the tools of science. Yeah. Yeah. And as we have more precise ... And by the way, there are little microscopic violations of the second law of thermodynamics that you can start to have when you have more precise measurements of where precisely molecules are. But for a large scale, when you have enough molecules, we don't have ... We're not tracing all those molecules, and we just don't have the computational resources to do that. And it wouldn't be ... I think to imagine what an observer who is not computationally bounded would be like, it's an interesting thing because ... Okay, so what does computational boundedness mean? Among other things, it means we conclude that definite things happen. We take all this complexity of the world, and we make a decision, we're going to turn left or turn right. And that is reducing all this detail into we're observing it, we're crushing it down to this one thing. And if we didn't do that, we wouldn't have all this symbolic structure that we build up that lets us think things through with our finite minds. We'd be instead ... We'd be one with the universe. </s><s>[INST] Yeah, so content to not simplify. [/INST] Yes. If we didn't simplify, then we wouldn't be like us. We would be like the universe, like the intrinsic universe, but not having experiences like the experiences we have, where we, for example, conclude that definite things happen. We have this notion of being able to make narrative statements. </s><s>[INST] I wonder if it's just like you imagined as a thought experiment, what it's like to be a computer. I wonder if it's possible to try to begin to imagine what it's like to be an unbounded computational observer. [/INST] Well, okay, so here's how that, I think, plays out. Vibrations. Yeah. So, I mean, in this, we talk about this rulliad, this space of all possible computations. This idea of being at a certain place in the rulliad, which corresponds to a certain way of ... A certain set of computations that you are representing things in terms of. Okay, so as you expand out in the rulliad, as you kind of encompass more possible views of the universe, as you encompass more possible kinds of computations that you can do, eventually you might say that's a real win. we're colonizing the Rulliad, we're building out more paradigms about how to think about things, and eventually you might say, we won all the way. We managed to colonize the whole Ruliad. Okay, here's the problem with that. The problem is that the notion of existence, coherent existence, requires some kind of specialization. By the time you are the whole Ruliad, by the time you cover the whole Ruliad, in no useful sense do you coherently exist. So in other words, in the notion of existence, the notion of what we think of as definite existence requires this kind of specialization, requires this kind of idea that we are not all possible things, we are a particular set of things. And that's kind of what makes us have a coherent existence. If we were spread throughout the rulliad, there would be no coherence to the way that we work. We would work in all possible ways. And that wouldn't be kind of a notion of identity. We </s><s>[INST] wouldn't have this notion of coherent identity. Interesting. I am geographically located somewhere exactly, precisely in the rulliad, therefore I am, is the Descartes kind of... [/INST] Yeah, yeah, right. Well, you're in a certain place in physical space or in a certain place in rural space. And if you are sufficiently spread out, you are no longer coherent. And you no longer have... I mean, in our perception of what it means to exist and to have experience, it doesn't happen. </s><s>[INST] Right. So therefore, to exist means to be computationally bounded. I think so. To exist in the way that we think of ourselves as existing, yes. The very act of existence is like operating in this place that's computationally reducible. So there's this giant mess of things going on that you can't possibly predict. But nevertheless, because of your limitations, you have an imperative of like, what is it? An imperative or a skill set to simplify? Or an ignorance, a sufficient [/INST] imperative of like... Okay, so the thing which is not obvious is that you are taking a slice of all this complexity, just like we have all of these molecules bouncing around in the room. But all we notice is the kind of the flow of the air or the pressure of the air. We're just noticing these particular things. And the big interesting thing is that there are rules, there are laws that govern those big things that we observe. So it's not obvious. That's how amazing. </s><s>[INST] Because it doesn't feel like it's a slice. It's not a slice. It's like an abstraction, [/INST] not a slice. Yes. But I mean, the fact that the gas laws work, that we can describe pressure, volume, et cetera, et cetera, et cetera, and that we don't have to go down to the level of talking about individual molecules, that is a non-trivial fact. And here's the thing that I sort of exciting thing as far as I'm concerned, the fact that there are certain aspects of the universe. So we think space is made ultimately these atoms of space and these hypergraphs and so on. And we think that, but we nevertheless perceive the universe at a large scale to be like continuous space and so on. In quantum mechanics, we think that there are these many threads of time, these many threads of history, yet we kind of span. So in quantum mechanics, in our models of physics, there are these, time is not a single thread. Time breaks into many threads. They branch, they merge. But we are part of that branching, merging universe. And so our brains are also branching and merging. And so when we perceive the universe, we are branching brains perceiving a branching universe. And so the fact that the claim that we believe that we are persistent in time, we have this single thread of experience, that's the statement treatment that somehow we managed to aggregate together those separate threads of time that are separated in the fundamental operation of the universe. Just as in space, we're averaging over some big region of space, and we're looking at the aggregate effects of many atoms of space, similarly in what we call branchial space, the space of these quantum branches, we are effectively averaging over many different branches of histories of the universe. in thermodynamics, we're averaging over many configurations of many, many possible positions of molecules. So what we see here is, the question is when you do that averaging for space... What are the aggregate laws of space? When you do that averaging over branchial space, what are the aggregate laws of branchial space? When you do the averaging over the molecules and so on, what are the aggregate laws you get? And this is the thing </s><s>[INST] that i think is just amazingly neat. [/INST] That there are aggregate laws at all. Well yes, but the question is what are those aggregate laws. Yes. So the answer is for space, the aggregate law is Einstein's equations for gravity, for the structure of space-time. For base space, the aggregate laws are the laws of quantum mechanics. And for the case of molecules and things, the aggregate laws are basically the second law of thermodynamics. And so that's the things that follow from the second law of thermodynamics. And so what that means is that the three great theories of 20th-century physics, which are basically general artillery of the theory of gravity, quantum mechanics, and statistical mechanics, which is what kind of grows out of the second law of thermodynamics, all three of the great theories of 20th-century physics are the result of this interplay between computational the reducibility and the computational boundedness of observers. And for me, this is really neat, because it means that all three of these laws are derivable. So we used to think that, for example, Einstein's equations were just sort of a wheel-in feature of our universe, that they could be, the universe might be that way, it might not be that way. Quantum mechanics is just like, well, it just happens to be that way. The second law, people kind of thought, well, maybe it is derivable. What turns out to be the case is that all three of the fundamental principles of physics are derivable, but they're not derivable just from mathematics. They require, or just from some kind of logical computation, they require one more thing. They require that the observer, that the thing that is sampling the way the universe works, is an observer who has these characteristics of computational boundedness of belief in persistence and time. And so that means that it is the nature of the observer, the rough nature of the observer, not the details of, well, we got two eyes and we observe photons of this frequency and so on. But the kind of the very coarse features of the observer then imply these very precise facts about physics. And I think it's amazing. </s><s>[INST] So if we just look at the actual experience of the observer, that we experience this reality, it seems real to us. And you're saying because of our bounded nature, it's actually all an illusion. [/INST] It's a simplification, observer, simplification. Yeah, it's a simplification, right? But you don't think a simplification is an illusion? No. I mean, it's, well, I don't know. I mean, what's underneath... Okay, that's an interesting question. What's real? And that relates to the whole question of why does the universe exist? And what is the difference between reality and a mere representation of what's going on? Yes. But you don't think a simplification is an illusion? We experience the representation. Yes. But the question of... So one question is, why is there a thing which we can experience that way? And the answer is because this Rouliad object, which is this entangled limit of all possible computations, there is no choice about it. It has to exist. There has to be such a thing. It is in the same sense that 2 plus 2, if you define what 2 is and you put pluses and so on, 2 plus 2 has to equal 4. Similarly, this Rouliad, this limit of all possible computations, just has to be a thing that is once you have the idea of computation, you inevitably have the Rouliad. You're going to have to have a Rouliad, yeah. Right. And what's important about it, there's just one of it. It's just this unique object. And that unique object necessarily exists. And then the question is, once you know that we are sort of embedded in that and taking samples of it, it's sort of inevitable that there is this thing that we can perceive that our perception of physical reality necessarily is that way, given that we are observers with the characteristics we have. So in other words, the fact that the universe exists, it's almost like to think about it almost theologically, so to speak. It's funny because a lot of the questions about the existence of the universe and so on, they transcend what kind of the science of the last few hundred years has really been concerned with. The science of the last few hundred years hasn't thought it could talk about questions like that. But I think it's kind of, and so a lot of the kind of arguments of, does God exist? Is it obvious that I think in some sense, in some representation, it's sort of more obvious that something sort of bigger than us exists than that we exist. And we are, our existence and as observers the way we are is sort of a contingent thing about the universe. And it's more inevitable that the whole universe, the whole set of all possibilities exists. But this question about, is it real or is it an illusion? All we know is our experience. And so the fact that, well, our experience is this absolutely microscopic piece of sample of the Rouliad. And there's this point about we might sample more and more of the Rouliad, we might learn more and more about, we might learn like different areas of physics, like quantum mechanics, for example, the fact that it was discovered, I think, is closely related to the fact that electronic amplifiers were invented that allowed you to take a small effect and amplify it up, which hadn't been possible before. Microscopes have been invented that magnify things and so on, but having a very small effect and being able to magnify it was sort of a new thing that allowed one to see a different aspect of the universe and let one discover this kind of thing. So we can expect that in the Rouliad, they're an infinite collection of new things we can discover. There's in fact computational irreducibility guarantees that there will be an infinite collection of pockets of reducibility </s><s>[INST] that can be discovered. Boy, would it be fun to take a walk down the Rouliad and see what kind of stuff we find there. You write about alien intelligences. Yes. I mean, just these worlds [/INST] of computation. Yes. The problem with these worlds is that- We can't talk to them. Yes. And the thing is, what I've kind of spent a lot of time doing is just studying computational systems, seeing what they do, what I now call rouliology, kind of just the study of rules and what they do. You can kind of easily jump somewhere else in the Rouliad and start seeing what these rules do and what you say is they just do what they do and there's no human connection, so to speak. </s><s>[INST] Can't talk to them. Do you think some people are able to communicate with animals? Do you think you can become a whisperer of these- [/INST] Coitition trying. That's what I've spent some part of my life doing. Have you heard? Are you at the risk of losing your mind? Are you at the risk of losing your mind? My favorite science discovery is this fact that these very simple programs can produce very complicated behavior. And that fact is kind of, in a sense, a whispering of something out in the computational universe that we didn't really know was there before. It's like back in the 1980s, I was doing a bunch of work with some very, very good mathematicians and they were trying to pick away. Can we figure out what's going on in these computational systems? And they basically said, look, the math we have just doesn't get anywhere with this. We're stuck. There's nothing to say. We have nothing to say. And in a sense, perhaps my main achievement at that time was to realize that the very fact that the good mathematicians had nothing to say was itself a very interesting thing. That was kind of a sort of, in some sense, a whispering of a different part of the Rouliad that one wasn't accessible from what we knew in mathematics and so on. </s><s>[INST] Does it make you sad that you're exploring some of these gigantic ideas and it feels like we're on the verge of breaking through to some very interesting discoveries and yet you're just a finite being that's going to die way too soon. And that scan of your brain or your full body kind of shows that you're- Yeah, it's just a bunch of meat. It's just a bunch of meat. Yeah. Does that make you a little sad? [/INST] Kind of a shame. I mean, I kind of like to see how all this stuff works out. But I think the thing to realize, it's an interesting sort of thought experiment. You say, okay, let's assume we can get cryonics to work. And one day it will. That will be one of these things that's kind of like chat GPT. One day somebody will figure out how to get water from zero degrees centigrade down to minus 44 or something without it expanding and cryonics will be solved and you'll be able to just put a pause in, so to speak, and kind of reappear 100 years later or something. And the thing though that I've increasingly realized is that in a sense, this whole question of kind of the sort of one is embedded in a certain moment in time and kind of the things we care about now, the things I care about now, for example, had I lived 500 years ago, many of the things I care about now, it's like that's totally bizarre. I mean, nobody would care about that. It's not even the thing one thinks about. In the future, the things that most people will think about, one will be a strange relic of thinking about what might have been a theologian thinking about how many angels fit on the head of a pin or something, and that might have been the big intellectual thing. It's one of these things where particularly I've had the, I don't know, good or bad fortune, I'm not sure, I think it's a mixed thing, that I've invented a bunch of things which I think see well enough what's going to happen that in 50 years, 100 years, whatever, assuming the world doesn't exterminate itself, so to speak, these are things that will be sort of centrally important to what's going on. It's both a good thing and a bad thing in terms of the passage of one's life. If everything I'd figured out was like, okay, I figured it out when I was 25 years old, and everybody says it's great, and we're done. It's like, okay, but I'm going to live another how many years, and that's kind of, it's all downhill from there. In a sense, it's better, in some sense, to be able to, you know, it sort of keeps things interesting that, you know, I can see a lot of these things. I mean, it's kind of, I didn't expect chat GPT, I didn't expect the sort of opening up of this idea of computation and computational language that's been made possible by this. I didn't expect that. This is the head of schedule, so to speak. You know, even though the sort of the big kind of flowering of that stuff, I'd sort of been assuming was another 50 years away. So if it turns out it's a lot less time, that's pretty cool, because, </s><s>[INST] you know, I'll hopefully get to see it, so to speak, rather than. Well, I think I speak for a very, very large number of people in saying that I hope you stick around for a long time to come. You've had so many interesting ideas. You've created so many interesting systems over the years, and I can see now that GPT and language models broke open the world even more, I can't wait to see you at the forefront of this development, what you do. And yeah, I've been a fan of yours, like I've told you many, many times since the very beginning. I'm deeply grateful that you wrote a new kind of science, that you explored this mystery of cellular automata and inspired this one little kid in me to pursue artificial intelligence and all this beautiful world. So Stephen, thank you so much. It's a huge honor to talk to you, to just be able to pick your mind and to explore all these ideas with you, and please keep going. And I can't wait to see what you come up with next. And thank you for talking today. We went past midnight. We only did four and a half hours. I mean, we could probably go for four more, but we'll save that till next time. This is round number four. I'm sure you'll talk to me any more times. Thank you so much. My pleasure. Very good going. Thanks for listening to this conversation with Stephen Wolfram. To support this podcast, please check out our sponsors in the description. And now, let me leave you with some words from George Cantor. The essence of mathematics lies in its freedom. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] Cutting it like I never touch it. I can't touch it. So there's something almost like, um The texture of the biology of a human flesh that just [/INST] That's at the level of that's the level of your trauma. Yeah, and it's been I mean, it's 2021 This was an oh nine and i've worked on this like and I mean I i've been in like treatment And religiously just to be able to keep me alive for this this decade And so it's not like it's like oh i'm never you never even tried to get better It's like I never used to leave my house. I used to call people that look like that horrific names in public </s><s>[INST] I used to want to kill people on a regular basis. I'm fairly happy individual now. What about you're talking about sort of skin and parts and But there's also just the fact that we're mortal and there's somebody Close to you who dies so you watch walk up and then never come back out again. Come back out again Yeah, it's like you're facing mortality in a very real way and uh In a way that's not the same as somebody dying from cancer in a hospital. Although It has echoes of that because that's also absurd and And like it doesn't feel like there's justice to it in any kind of way, but it's so sudden like have you been able to um make sense of that Of your feelings about it. Like how do you feel about it or or is everything just shrouded in this like? Trauma that you're not able to just um [/INST] Feel for the loss of a human being like mourn the loss of a human being. I think I had I thought he was not a human being I Don't want to dig into all the immers dripping and reap the blood I think that's what he said I oftentimes think of him the first time he said a person gone and he went and actually There was also it throughout to be active and then you're safe </s><s>[INST] That's how you deal with that That means you have a feeling like, [/INST] you still feel like parts of him were left behind. On the ramp ceremony, when I lost my mind, literally, I lost my mind, and I was screaming that he wasn't all in there. I'm happy now knowing that he was, </s><s>[INST] but I held on to that for 10 years. Yeah, this, yeah, the sandbags, like it's, the bulk of the weight [/INST] is not from human flesh, yeah. He was a young kid too. It was, I think it was his first deployment as well. Like he was, he was a young kid, and he was just going to clear the road for the rest of us, right? Like not, like you know you're in war, and you know that you're outside the wire, and you know things could happen. You understand that to the extent you can understand that. </s><s>[INST] When it's happening, it's something very different. Also, maybe you can correct me, but there's something much more like brutal about an IED versus like a bullet. Because a bullet, you still, watching somebody close to you die from a bullet, you still get the basic humanity. Like, so IED basically converted a human being into sort of parts by biological parts. Yeah, versus, yeah, I mean, I don't, that's tough. That's tough, because that's like, because it's hard for you to remember them as a human. [/INST] You remember them as parts. For me, that's how I remember him. I would like to, I have a picture that I post every year about him. I see that, but I don't put two and two together. </s><s>[INST] Does that make sense? Yeah, yeah, no, it does. So I, even I, listening to your story, you know, thank you for sharing it, first of all. [/INST] It's fine, it's not my, it's, </s><s>[INST] I'm just the one to tell it, I was just involved. I'm just the one to tell it, I was just involved. One set of eyes on this particular human being. [/INST] But even I get angry. </s><s>[INST] I can't tell if it's exhaustion or anger. [/INST] Oh, I'm sorry, I look always exhausted. Oh, that's okay, you're into robotics, isn't that like your guys' thing? </s><s>[INST] You guys are just always working on time. I think because I feel so much for the world, I just don't do, we were talking about resting bitch face earlier. I just don't feel the need to maintain all the effort of the musculature for presenting myself to you visually. It's exhausting. It's exhausting. So I just focus on the feeling. No, hey, I, and then let the face show whatever the hell it shows. Fair enough. [/INST] Okay, was their anger, was there hate? Yeah, exhausting. It's exhausting. So I just focus, And then let the face- </s><s>[INST] Fair enough, yeah. Can you just talk through your feelings [/INST] of what you remember, of what you- Yeah, so after that operation with the British, I went back to the Canadians. And I didn't go back as even remotely close to who I was when I left. And that was really troublesome for a lot of people around me because the level of anger and hate that came out of me was palpable when I just walked. I got shockingly quiet and you understand how you're learning- That'd be terrified if you were quiet. And I don't know if hate and anger do that justice. I don't know another word, but I don't think those two words do it justice to the extent that I was feeling. Like I got to a point when I got attacked by a woman with some scissors, like the idea crossed my mind, like I could boot stomp her to death and not feel anything about it </s>

<s>[INST] There are aspects of the fundamental warming question that are still deeply uncertain. But the debate is like two, three, or four Celsius. It's in that range. The thing is, all of those are bad for, this is an educational question. It doesn't seem like that much from a weather perspective, if you just turn up the AC and so on in your [/INST] own personal home. But it is, from a global perspective, a huge impact on agriculture. It's in that range. Well, yeah. And getting back to sea level and glaciers, the melting point of ice is a number. And so if you pass that number, things start to change. What became known about Antarctica and Greenland more is that it's ocean temperature, the seawater in and around and under these ice sheets, because it kind of gets under parts of Antarctica, is what's driving the dynamics that could lead to more abrupt change, more than air temperature. In glaciers, these big ice sheets live or die based on how much snow falls and how much ice leaves every year. And I was up on the Greenland ice sheet in 2004 and written about it forever since then. It's the same amount of water that's in the Gulf of Mexico as if God or some great force came down. If flash flows the Gulf of Mexico and plunked it up on land, that's the ice sheet. It's a lot of water. That's 23 feet of sea level rise. We're not going to melt at all. And the pace at which that erosion begins and becomes sort of a runaway train is still not well understood. That changed from like a manageable level of sea level rise from these ice sheets to something that becomes truly unstoppable or that has these discontinuities where you get a lot more all of a sudden. To me, it's in the realm of what I've taken to calling known unknowables. Don't count on another IPCC report magically including science that says, aha, now we know it's going to be five feet by 2100. Because learning, there's a lot of negative learning in science. This may be true in your body of science too. There's a guy named Jeremy Bassus, B-A-S-S-I-S, who wrote a paper about the idea that you could get this sudden cliff breakdown of these ice shelves around Antarctica leading to rapid sea level rise. He did more modeling in physics and it turns out that you end up with, it's a much more progressive and self-limiting phenomenon. But those papers don't get any attention in the media because- They're not scary. They're not scary and they're sort of after the fact. Just this past year, there's been this cycle around collapse, the word collapse, and Antarctic ice. It started actually several years ago with the idea that the West Antarctic ice sheet is particularly vulnerable and the science community like the birds, we were talking about flocks to it and some high profile papers are written. And then a deeper inquiry reveals, it's more complicated than that. And we, the journalists, the media, pundits don't pay attention to that stuff. They're not. And actually, which is why I started to develop kind of a dictionary, I call it watchwords, like words to, if you're out there, you're just a public person, you want to know what's really going on. You hear these words like collapse in the context of ice, what do you do with that? And so I've created conversations around these words. Geologists and ice scientists use the word collapse. They're talking about a centuries long process. They're not talking about the World Trade Center and scientists would do well to be more careful with words like that. Unless your focus is what we were saying earlier, your idea that alarming people will spur them to act, then you use that word carelessly. Can I just follow up on the other point that you said, two, three, four degrees, that doesn't sound like much. I can just crank up the air conditioning. I think that sort of touches on a really, really important point, that for most rich people, much of climate change is not really going to be all that impactful. It still will have an impact, but fundamentally, if you're well off, you can mitigate a lot of these impacts. And there's a young scientist at Carnegie Mellon, Destiny Nock. She just was the lead author on a study, what poor and prosperous households do in a heat wave when they have access to air conditioning. In a poor household, you wait, they found through science. They delay turning on the air conditioner four to seven degrees more of heating before they start to use air conditioner. And that create adverse outcomes. If you have an asthmatic in the house, an old person, you're endangering their lives. And that's just a little tiny microscopic fractal example of this powerful real phenomenon that there's a divide in vulnerability and it's not just based on where you live. This is families in Pittsburgh. We're not talking about Botswana. And so that divide and capacity to deal with environmental stress is something you can really work on. And it gets hidden in all this talk of climate crisis. And that's one of the important parts is both to say, look, if seven billion people, sorry, eight billion people will now have all experienced this even though for each one of them it's manageable, it's still a big problem because it's eight billion people living through this. How do you air condition eight billion? Yes. And then it's the point of getting to realize it's very, very much about how do you help the world's poor? And that's very much about making it more affordable but basically getting them out of poverty and remember, getting out of poverty, doesn't just mean that they can now afford to air condition themselves, but they get better education, they get better opportunities, they get better lives in so many other ways. And then at the end of it, it's not just about making sure that we focus on this one problem, but it's recognizing that these families have lots of different issues that they would like us to focus on climate and heatwaves, just being one of them. So it's sort of taking progressive steps back and realizing, okay, this is a problem, not the end of the world. And how do you air condition eight billion? Yes. And one tiny little last example, you mentioned Jakarta at the beginning. It's really valuable to look around the world at places that are sort of leading indicator places, whether it's sea level rise or heat, and you could do that. Jakarta is sinking like a foot a year, literally a foot a year. It's some insane number from withdrawing groundwater, from gas withdrawal, it's a delta, it's built on sediment. I wrote a piece ages ago at the New York Times calling it Delta Blues, you know, the musicians. And in Jakarta, so what are they doing? They're moving, they're moving the capital to another area. And so that says to me, there's a lot of plasticity too, it's a city that's going through that rate of sea level, of their relationship with the sea level through sinking is way faster than what's happening with global warming. So look there, look to those kinds of places and you can start to build. Tokyo had the same thing in the 1930s. So we're also withdrawing a lot of water way too fast. And so, you know, one of the obvious things is maybe you should stop withdrawing water so fast. Yeah. And again, we seem to almost be intent on finding the most politically correct way to fix a problem or, you know, the most, the thing that sort of gets the most clicks instead of the thing that actually works the best. </s>

<s>[INST] So you've said the words Nobel Prize over 240 times. Yeah, I mean, it is so strange as a symbol that kind of unites this whole scientific journey, right? I got, it's both sad and beautiful that a little prize, like a little award, a medal, a little plaque, they'll be most likely forgotten by history completely. So some silly list. It's somehow a catalyst for greatness. It resulted in you doing your life's work. [/INST] The dream of it. Would I have done it without the Nobel Prize? I can't necessarily counterfactually state that that would have happened. So no, it definitely has a place. And for me, it is valuable to think about it. But the level of obsession that academics have about it is really, I think it is almost unbalanced becoming unhealthy. And again, I have no, I make no truck with the winners of the Nobel Prize, obviously. Now I've had 11 on the show. And to think about like the one rule. So by the way, right after the denouement of the story, which I'll get to in a bit, how our dreams went down to dust and ashes, I was asked by the Royal Swedish Academy of Sciences to nominate the winners of the 2015 Nobel Prize in physics. So like the one that I theoretically could have been eligible to win in 2016, actually, they asked me to nominate. Now imagine if I ask you, Lex, you say Brian, instead of me inviting myself on the show, if you say, Brian, would you like to come on the Lex Friedman podcast? I say, you know what, Lex, you know that guy, Rogan, I think you might already, can you introduce him to me? Like, do you imagine how that would feel? Like you'd be like, ah, you know, I'm here. So I was asked to nominate the winners. And the one rule that they say, of all the rules that Alfred Nobel stipulated, there's only one rule that they maintained. In other words, he said one person can win it for something they discovered in the preceding year that had the greatest benefit to mankind, made the world better, right? None of that was mentioned in the letter. It said many people can win it for work done long ago. They didn't mention anything in the letter to me, signed by the secretary general, nothing about benefiting mankind. He said, just one thing, can't nominate yourself. </s><s>[INST] So none of these guys nominated themselves. Actually, little known fact, they sent that exact letter just to you. [/INST] That rule was created just for you. That's called the Keating correlate. </s><s>[INST] Yes, exactly. Just to like, good for them, rub it in. I mean, it's in this particular case, of course, there's like some weird technicality or whatever. [/INST] But in this particular case, that's kind of a powerful. </s><s>[INST] Good for them. A reminder that the Nobel prize leaves a lot of people [/INST] behind in their stories behind all of that. Yeah. No way. Yeah, I mean, here's a good example. Again, this is my friend, Barry Barish. He's become like a mentor and a friend. He wrote the forward to this, my book into the impossible. He won the Nobel prize because a different guy died and he admits it and he said it. And actually it's funny with him because I've heard you talk, you know, very rhapsodically and lovingly and romantically about, with Harry Cliff and a wonderful podcast with him, by the way, about the LHC and how wonderful it is and how in that, you know, we were about to build the superconducting supercollider right here in Texas and it didn't get built and got canceled by Congress. And I'd say to barry, that was the best thing that ever happened to you. He's like, what the hell are you talking about? And I'm like, if that didn't get canceled, first of all, the, even though it did get canceled, the Europeans went on to build it themselves, saved the American taxpayers billions of dollars. And we wouldn't have learned anything really substantially new, as proven by the fact that, as you and Harry talked about, nothing besides the Higgs particle of great note has come out. And actually, he's had a recent paper, but it's been an upper limit, along with his collaborators on the LHCB experiment that I'm going to be talking with him about. But the bottom line is, it was really built to detect the Higgs. So the SSC, for twice as much money, would have sucked up Barry's career. He would have been working on that, maybe not. And then he would never have worked on LIGO. And then he wouldn't have won the Nobel Prize, right? So you look at counterfactual history. That's not actually a big stretch, right? If the SSC had still gone on, he would have worked that, because he was one of the primary leaders of that experiment. Second thing, if, imagine the following thing had happened. They won the Nobel Prize, because in September 2015, they detected unequivocal evidence for the in-spiral collision of two massive black holes, each about 30 times the mass of the sun, leaving behind an object that had just less than 60 solar masses behind. So one solar mass worth of matter got converted to pure gravitational energy. No light was seen by them. This particular date, September 14, 2015. That explosion, because of the miracle of time travel that telescopes afford us, that actually took place 1.2 billion years ago in a galaxy far, far away. They actually don't know which galaxy took place in stealth, and they never will. OK? If that collision between these two things, which have probably been orbiting each other for maybe a million years or more, if that had occurred 15 days earlier, </s>

<s>[INST] to [/INST] be good at it? Uh, an incredible amount of uh, performance across your whole stack, you know, starts with hardware. Uh, and again, you want to use old sensing modalities available to you. Imagine driving in a residential road at night and kind of making a turn and you don't have headlights covering some part of the space and like, you know, a kid might run out and you know, lighters are amazing at that. They see just as well in complete darkness as they do during the day. Right? So just again, it gives you that extra uh, you know, margin in terms of capability and performance and safety and quality. Uh, and in fact we often times uh, in these kind of situations, we have, our system detects something in some cases even earlier than our trade operators in the car might do, especially in conditions like very dark nights. Um, so start with something then, you know, perception has to be incredibly good. I have to be very, very good at detecting uh, pedestrians uh, and all kinds of situations and all kinds of environments, including people in weird poses, People kind of running around and you know, being partially occluded. Um, uh, so you know, that that stuff number one, right then you have to have in very high accuracy and very low latency in terms of your reactions uh to, you know, you know what, you know, these actors might do. Right? And we've put a tremendous amount of engineering and tremendous amount of validation and to make sure our system performs properly. And, you know, oftentimes it does require very strong reaction to do the same thing. And, you know, we actually see a lot of cases like that, that's the long tail of really rare, you know, really, uh, crazy events uh, that, um, contribute to the safety around pedestrians. Like 11 example that comes to mind that we actually happened in phoenix where we were driving along and I think it was a 45 mile per hour road. So you're pretty high speed traffic and there was a sidewalk next to it and there was a cyclist on the sidewalk, and as we were in the right lane, right next to the side. So, because a multi lane road, uh, as we got close to the cyclist on the sidewalk, it was a woman, she tripped and fell, just fell right into the path of our vehicle, right? Um, and our car, uh, you know, this is actually a test driver. Our test drivers, uh, did exactly the right thing. Uh, they kind of reacted and came to stop. It requires both very strong steering and uh, you know, strong application of the break. Uh, and then we simulated what our system would have done in that situation and it did exactly the same thing. Uh, and that, that speaks to all of those components of really good uh, state estimation and tracking and like imagine, you know about a person on a bike and they're falling over and they're doing that right in front of you, right? You have to be really like things are changing the appearance of that whole thing is changing, right? And the person goes one way they're falling on the road there, you know, being flat on the ground in front of you, you know, the bike goes flying the other direction, like the two objects that used to be one and you know, are splitting apart and the car has to detect all of that like milliseconds matter. And it doesn't, it's not good enough to just break, you have to like steer and brake on their traffic around you. So like it all has to come together. And it was really great to see in this case and other cases like that, that we're actually seeing in the wild that our system is, you know, performing exactly the way uh that we would have liked and is able to, you know, avoid uh, collisions like this. </s>

<s>[INST] Yeah. So the yeah, we humans aren't able to answer the why either. Yeah, but I do think the question here is, well, first of all, language is finite, certainly weren't expressed to a tweet. Uh so it is also combinatorial problem. The question is how much more difficult it is than chess. And I think I think all the same ways that we see the solutions to chest is deeply surprising when it was first addressed with I B M. D. Blue and then with alpha go and Alphago zero, alpha zero. I think in that same way language can be addressed and communication can be addressed. I don't see having done this podcast many reasons why everything I'm doing, especially as a digital being on the internet can't be done by an Ai system eventually. So like I think we're being very human centric and thinking were special. I think one of the hardest things is the physical space actually operating like touch and the magic of body language and the music of all of that because it's so deeply integrated through the long evolutionary process of what it's like to be on earth, what is fundamentally different and ai has you can catch up on is the way we apply our evolutionary history on the way we act on the internet and the way we act online and as more and more of the world becomes digital, you're not operating in a space where A. I. Is behind much less. So Like we're both starting at zero. [/INST] I think that's super interesting. Do you do you know this, do you know this author brian christian, is that someone you've ever heard of? He's a guy who competed in the, what is it called? The lobster competition. Yeah, the during testing. So, and I'll just tell you this story. But I think it's directly related to the last thing you said about where we're starting in the same place. Uh he competed in this in this competition. But not he didn't enter a program that was supposed to try to pass the turing test. The turing test. You know, there's three people, there's the judge, there's the program and then there's there's a someone who's a human the way they do it. And the judge has got to figure out what by asking questions, which is the computer in which is the human so little known fact, there's two prizes in that competition. There's the most human computer prize that's the computer that wins the most and then there's the most human human rise and he competed for the most human human progress and he wanted. He kept one winning it. And and so he tried to think about what it is that um, you have to be able to do in order to convince judges that you're human instead of a computer. And that's an interesting question, I think. Um, and when he came to my take away from his version of this story, is that it is true that computers are winning these contests more and more, you know, as technology progresses. But there's two possible explanations for that. One is that the computers are becoming more human and the other is that the humans are becoming more like computers. And he says, actually, the more we live our lives in this world where in this sort of technological world where we have to moderate our behavior so that it's readable. Bye. Something that's effectively, you know, a computer, the more we become like that. And he says, it happens even when you're uh not interacting with a computer. He says, have you ever been to the, you know, on the phone with the call, you know, center? And they're going through their script and that's what they've got to do. They've got to go through their script because that's how they keep their job. And they ask you this question, you've got to answer it. And and it's as if you're no longer interacting with a person, even though it's a person because they've so given up everything that's involved normally with being able to make judgments and decisions and act in situations and take responsibility. And so I think that's that's the other side of it, it is true that technology is amazing and can solve huge ranges of problems and and do fantastic things. Was also true that we're changing ourselves in response to it. And the one thing I'm worried about is that we're changing ourselves in such a way that the norms for what we're aiming at are being changed to move in the direction of this sort of efficiently and in an optimized way, solving a problem and move away from this other kind of thing that we were calling alive, Alive Nous or significance. And and so that's the that's the other side of the story. </s><s>[INST] And that's the worry. But it's very possible that there is uh for you and I, the ancient dinosaurs, we may not see the liveliness in tick tock the Alive nous in the digital space that you see it as us being dragged into this over optimized world, but that may be uh this is in fact, it is a world that opens up opportunities to truly experience life. And there's interesting to think about all the people growing up now who their early experience of life is always mediated through a digital device. Not always, but more and more often mediated through that device and how we're both evolving. The technology is evolving. The humans are evolving to them. Maybe open a door to the whole world where the humans and the technology or ai systems are interacting as equals. [/INST] So now I'm going to agree with you, you might be surprised that I'm going to agree with you, but I think that's exactly right. I I don't want to be the person who's saying our job is to resist all of this stuff that I don't want to be a Luddite. That's not my goal. The goal is to point out that in the supreme danger lies the saving power. The point is to get in the right relation to that understanding of what we are uh that allows us to find the joy in it. And that's I think that's a hard thing to do. It's hard to understand even what we're supposed to be doing when we do it. Maybe I more than you and I'm not of the right generation to be able to do that. But I do think that's got to be the move. The move is not to resist it. It's not a nostalgic move. It's an attempt to push people to get in the relation to it. That's not the relation of it controlling you and depriving you of stuff. But if you're recognizing some great joy that can be found in it </s>

<s>[INST] world. So the kind of perturbations with giraffe or whatever the heck it was. Those were those part of the simulation. [/INST] Well, the simulation was generally so the simulation was trained to be robust to many different things, but not the kind of perturbations we've had in the video. So it's never been trained with the glove. It's never been trained with uh a stuffed giraffe. </s><s>[INST] So in theory, these are novel [/INST] probation, correct? It's not in theory in practice. </s><s>[INST] And that those are novels probation. Well, that's okay, that's a clean small scale but clean example of a transfer from the simulated world to the physical [/INST] world. And I will also say that I expect the transfer capabilities of deep learning to increase in general and the better the transfer capabilities are, the more useful simulation will become. Because then you could take you could experience something in simulation and then learn a moral of the story which you could then carry with you to the real world, as humans do all the time in the play computer games. So </s><s>[INST] let me ask sort of an embodied question staying on a G for a sec. Do you think A. Js is still we need to have a body, we need to have some of those human elements of self awareness? Consciousness sort of fear of mortality sort of self preservation in the physical space which comes with having a body? [/INST] I think having a body will be useful. I don't think it's necessary but I think it's very useful to have a body for sure. Because you can learn whole new, you can learn things which cannot be learned without a body. But at the same time I think that you can, if you don't have a body, you could compensate for it and still succeed. I think so. Yes. Well there is evidence for this for example, there are many people who were born deaf and blind and they were able to compensate for the lack of modalities. I'm thinking about Helen Keller specifically. </s><s>[INST] So even if you're not able to physically interact with the world and if you're not able to I mean, I actually was getting at maybe let me ask on the more particular, I'm not sure if it's connected to having a body or not, but the idea of consciousness and a more constrained version of that is self awareness. Do you think I need a system should have consciousness? It's what we can't define God, whatever the heck you think consciousness is. [/INST] Yeah, hard question to answer. Given how hard is define it, </s><s>[INST] Do you think it's useful to think [/INST] about? I mean, it's it's definitely interesting. It's fascinating. I think it's definitely possible that our assistance will be conscious </s><s>[INST] doing as an emergent thing that just comes from. Do you think consciousness could emerge from the representation that's stored within your networks? So like that it naturally just emerges when you become more and more you're able to represent more and more of the world. [/INST] Well, it's I'd make the following argument which is humans are conscious and if you believe that artificial neural nets are sufficiently similar to the brain, then there should at least exist artificial neurons, we should be conscious to. </s><s>[INST] You're leaning on that existence proof pretty heavily. Okay. [/INST] But that's that's the best answer I can </s><s>[INST] give. No, I I know I know, I know. Um there's still an open question if there's not some magic in the brain that we're not? I mean, I don't mean a non materialistic magic but that that the brain might be a lot more complicated and interesting that we give it credit for [/INST] if that's the case then it should show up and at some point we will find out that we can continue to make progress. But I think I think it's unlikely. </s><s>[INST] So we talk about consciousness but let me talk about another poorly defined concept of intelligence. Again, we've talked about reasoning, we've talked about memory. What do you think it's a good test of intelligence for you? Are you impressed by the test that alan turing formulated with the imitation game of with natural language. Is there something in your mind that you will be deeply impressed by if a system was able to do, [/INST] I mean lots of things. There's certain there's certain frontier, there is a certain frontier of capabilities today. Yeah and there exists things outside of the frontier and I would be impressed by any such thing for example, I would be impressed by a deep learning system which solves a very pedestrian pedestrian task like machine translation or computer vision task or something which never makes a mistake human wouldn't make under any circumstances. I think that is something which have not yet been demonstrated and I would find it very impressive. </s><s>[INST] Yeah, so right now they make mistakes in different, they might be more accurate in human beings but they still make a different set of mistakes. [/INST] So my my I would guess a lot of the skepticism that some people have about deep learning is when they look at their mistakes and they say well those mistakes they make no sense. Like if you understood the concept you wouldn't make that mistake and I think that changing that would be would would would inspire me that would be yes, this is this is this is this is progress. </s><s>[INST] Yeah, that's that's a really nice way to put it. But I also just don't like that human instinct to criticize the model is not intelligent, that's the same instinct as we do when we criticize any group of creatures as the other because yeah, it's very possible that Gpt two is much smarter than human beings and many things. [/INST] It's definitely true. It is a lot more breadth of knowledge. </s>

<s>[INST] you're talking the metaphor that's popping up in my mind. And this is also something that bothers me with theoretical physics. The metaphor is there's some sense in which there's things called theories of everything. When you try to describe the basic laws of physics, how they interact together and once you do, you have a sense that you understand all of reality in a sense you do and that to me when you do that to me, is understanding the individual, like how the individual behaves in this world. But then you're saying that there, hey hey you're also forgetting chemistry biology, how all of that actually comes together. The stickiness, the stick stickiness of molecules and how they build different systems and they some systems can kill each other, Some systems can flourish, Some can make pancakes and bananas and some can make poison and all those kinds of things that we need to be able to, we need to consider the the full stack of things that are constructed from the fundamental basics. Um and uh I guess you're on, you're saying that no, you you're just like the theoretical physicists, it all starts at the bottom like if you need to preserve the fundamentals of reality, which is the individual, like the basic atom of human society is the individual to you. [/INST] So yes, so the basic unit, the basic model unit, the basic ethical unit in society is the individual. And of course we form groups and and you can't understand history unless you understand group formation and group motivation. And and I have a view about what kind of groups should be formed uh and and politically from a political perspective, voluntary ones, ones in which we we join when we want to join and we can leave when we want to leave and uh 1/2 that help us and clearly groups help us pursue whatever it is, a goal is ultimately So so in the pursuit of happiness, there are lots of groups that one wants to form, whether it's marriage, whether it's businesses, whether it's sports teams, whether it's lots of there are lots of different groups want to form. But the but the question is, what is the standard of well being? Is it the standard well being, some algorithm that maximizes the well being of a group, you know, some utilitarian function, you know, is it is it something that's inherent in the group that we can measure as goodness and to help with the individuals within as long as we can get that, that the group to function well, we don't really care about where the individuals are. So to me, the the the the goal of creating groups is the well being of the individual. And that's why it needs to be voluntary. And that's why there has to be a way out of of those, sometimes it's costly. It's not a cheap out. That's why you should really think about what groups you you you and this, you know, on an issue that's very controversial. Maybe we can discuss maybe not. This is why, to me, immigration is so important, right? Open immigration or free immigration is because that's another group that I would like people to be able to voluntarily choose both in and out. And and I'd like to see people be able to go and join that group that, you know, that they believe will will will allow for the pursuit of happiness. But let me say that, you know, that's a description of an idea what I'm just saying. I recognize that that's not the reality in which we live? I recognize that that's not the reality in which history, history, you know, recognizing that such as that the individual exists in a sense, philosophically is a massive achievement, right? You know, human beings however they evolved clearly. We started out in in a, in a tribal context, in which individual didn't matter. We followed the leader. The competition was for power, power over the group and and and and dictates how the group should work. Um You know, the history of human beings is a history of gaining knowledge and part of the knowledge is the value of an individual. And you can see that in in in religion, you can see that in philosophy, you can see that through the evolution and then, you know, we we evolved from tribes into nations and then empires and conflicts between nations and conflicts, empires. And we tried a lot of different things if you will. I don't think we always did it on purpose, but we kind of different philosophies, different sets of ideas drove us towards different collectives, different groupings and and different ways in which the structure And after, I don't know, 3000 years of kind of known history, uh history before that, but we don't know much about it. 3000 years of known history. You can sit back and evaluate and I think that's what that's what is done in the enlightenment and and you sit back and suddenly we can do it today. We can sit back and evaluate what promotes human flourishing and what doesn't and what do we mean by human flourishing? Human whose whose flourishing? Well, individual human beings. Now, since I don't believe in a zero sum world and the world is not zero sum, like we can see that it's it's empirically possible to show that the world is not a zero sum game. My fellow human flourishing doesn't come into your expanse. So I I you know, I can show that a system that promotes my flourishing will probably promote your flourishing as well and promotes the general welfare in that sense, because it promotes individuals flourishing. And we can we can look at all these examples of how we evolved and what leads to bloodshed and what doesn't and what promotes this ability to flourish as an individual. Again, an achievement, the idea of individual flourishing, and then we can think about how to create a political system around that a political system that recognizes and allows for the formation of groups, but just under the principle of voluntary. So you can't be forced to join a group, you you can't be caused into forming a group um other than the fact that you're born in a particular place in a particular, you know that in a sense, but that's not force that's there's a difference between metaphysics and between choices. </s>

<s>[INST] By the way, neutron would be its name, right? Yep, so not like neutron. It's not some kind of fascinating new physics breakthrough [/INST] where they're using neutrons, right? Yep, so not like neutrons. No, no, but they are using, they're also using liquid methane and liquid oxygen. I just think it's a really, it seems like a great rocket and assuming they can actually get it flying in two or three years, I think they're gonna be, it's here to stay, you know? I'd be remiss right now, I'm editing a video from an interview with Stoke Aerospace out in Kent, Washington. It was just one of these companies that they have a long ways to go. Like they're still in the very, they're behind the curve, frankly, in terms of launch vehicles right now, because like I said, there's so many coming out of the woodwork. But the idea they're working on, their solution to a fully reusable rocket is amazing, one of the coolest concepts I've ever seen. Are you gonna cover it in the video? Yeah, yep, yep, yep. That'll be hopefully coming out the next, depending on what the schedule like is down there. I'm actively editing that as we speak and it is so cool. I mean, it is like, it's genius. And if they can actually get it to work, I can see them merging. I can for sure see someone potentially, like I, perfectly, in a perfect world, they merged with Rocket Lab. They, Stoke develops the upper stage and maybe even the engines. They are the two guys, the CEO, the co-founders of that company have, they are engine, like propulsion engineer magnificence. They have, they used to, they both have worked at blue. They developed engines in a hurry there and then left blue and it felt like it was getting too slow for them. And now they are, I mean, these guys fired a 15 chambered rocket engine instead of four from the Soviet and we're talking 15 chambers, single turbo pump, 70 times in the month of October. Wow. That's impressive. Wow. And that's like, that was on, you know, if you think about like days off, time off, you know, parts changing. Yes. Over twice a day on average of a hydrolox engine. That's insane. So I, I love them and I hope the best for them. But they're also topical right now. </s><s>[INST] They're at top of my head, so. [/INST] What about Firefly? What I like about Firefly, they've already got kind of a traditional aerospace backing. They're starting to buddy up a lot with Northrop Grumman. So they're going to be building the booster stage for Antares, which is currently flying only out of Wallops, Virginia, and is one of the only other commercial providers for the International Space Station. And Northrop Grumman is a very traditional aerospace company, you know, like lots of solid rocket boosters. And they've purchased, ironically, their current Antares is reliant on Russian engines and Ukrainian boosters, two things that I don't think you're going to be able to get your hands on too much anymore. Yes. They're going to some US propulsion and stages. So they actually are partnering with, with Firefly and their new Antares rocket will be a first stage built entirely by Firefly. So I'm, I'm excited that Firefly already has the propulsion technology. And they actually developed, ironically, their tap off cycle engine was developed in partnership with Ukraine, with Ukrainian engineers who developed the whole turbo pump system. So it's like, it's this cool meddling of, of these worlds. Their former CEO, Tom Rekusik was like, I have an interview with him and he's, anyone that can just spout nuances and facts, I just love, I just soaked that guy's information up as best I could, because he is brilliant. </s><s>[INST] Literally a doctor, a rocket doctor, you know, it's so. Yeah, I mean, that's what, like you said, the fascinating thing about these folks, they're, they're legit, they're, they're such great engineers that people, they bring these rockets to life. And then there's all this stuff that we know and don't know about in China and other parts and other nations that are putting stuff into orbit. One of the sad things also is like, you know, with Lockheed and Boeing is just military applications in general, there's so much technology that's currently being developed that we probably know nothing about. Yeah, and that makes me a little bit sad, of course. Yeah. For several reasons. One is that the use of that technology is, has really much, like, it's not, it's not that inspiring. It's like a very military focus. Yeah, it's to kill someone. It's to kill someone, yeah. There's not even like a, a side application. Right. And the big one is that the secret, it's shrouded in secrecy as opposed to being a source of inspiration. Yeah, 100%. But that's the way of the world. Like, what was that one plane that you covered that was like, we know nothing about? [/INST] Oh, the X-37B. The X-37B. Yeah, orbited for over 900 days and returned. Like, yeah, I wanna know about that thing. I don't know. That's what's, it's so frustrating. We know when it launches, people, you know, amateurs track and know, they even will be like, oh, it changed orbit, you know, it raised and lowered its orbit, blah, blah, blah. We generally have just almost no idea what it's doing up there. And it just saddens me. Cause I wanna know. And it's awesome. It's a great vehicle. </s>

<s>[INST] But anyway, whatever the future holds the past is certainly beautiful. Thanks to people like martin Scorsese and you can learn all about it with master class, this holiday give one annual membership and get one free, they have a big deal for you. If you go to masterclass dot com slash lex, that's masterclass dot com slash lex. For the amazing holiday deal terms apply. This is a lex Friedman podcast and here is my conversation with Ray Dalia. When you look at the history of the world, as you have done in your new book, principles of dealing with the changing world order, what is more important, more impactful money or power? [/INST] They go hand in hand, they support each other and they compete with each other. Those who have money have power, a certain type of power um that power has to do with all that they can buy, but it also has the ability to influence those with political power. So um and and so you see this throughout history, this symbiotic relationship, you know, um for example, between the royal families and the nobility and the church. So you see that group of people supporting each other in various ways and then wrestling around with each other for the money and power among that group. So the dynamic that's quite classic is you could look at the party is in power back in, you know that 16th, 17th centuries, you would look at royal families, nobles and um and the church, if you're in europe, and then you would look at um agricultural land and there was a certain dynamic and that varies over time. It changes um, as those people get thrown out in, technology changes. So when we evolved, when the society evolved so that it would produce goods and services and you have something like the Industrial Revolution, the first Industrial Revolution, you have machines and you have the talent of people that are off the farms. Um and then you have a struggling for power. You see that power mix change. And so we don't see that same power mix anymore. But you still see the same dynamic. You see those with money, um dealing with those who have political power around those assets that are considered most valuable to particularly the productive assets that produce money </s><s>[INST] and political powers usually centered around the nation state. So the major locus is of power is the nation's [/INST] Yes and 16 68 After 30 years of war, there was the development of nation states. Before then we had, I was really the development of countries as we know it, that there were borders and that within those borders, uh, those who had control got to control it and there was not to be intrusions in those borders and that's how we established the nation state. And then of course, within each country, there is levels, there is a central level and then there is a, typically a province or state level down below and then there's a municipal level. So they each have different levels of power and it's the coordination of those. Um that determines how the country is run. </s><s>[INST] You write that the quote archetypal big cycle governs the rising and declining empires and influences everything about them, including their currencies and markets. The most important three cycles are the ones you mentioned in the introduction, the long term debt and capital market cycle. The internal order and disorder cycle and the external order and disorder cycle. Can you describe this big cycle? [/INST] There are two orders. There is an order by order. I mean a system how the system works. So there's an internal order that is the internal governance system and it's usually set out in a constitution or some agreements. Um And then there's the world order, How the world system. So for example, in 1945, um at the end of a war which is basically a fight for determining the world order. Um The winners of the fight got together and created the world system, as we now know it 44, they created the Bretton Woods monetary system that determined the money pretty much. And because the United States one and had 80% of the world's money. Gold was money then and has 80% of the world's gold and it was half the world's economy and it was the great military power. The order was built around an american world order so that the United Nations was in new york, The world bank and the I. M. F. Were in Washington. And they built that new order. So classically you have a war to determine whose rules we're following. And then you have um the new water being constructed after that there is usually period of extended peace and prosperity piece suits prosperity. Um And so there's not, and the reason you have the piece is because no one wants to fight with the um dominant power, You know, you're gonna lose, you've surrendered, there's been the surrendering, they carve up the world, they say what it's gonna look like, who's going to control what and then um you come into that period of peace and then prosperity where then there's uh um are working together. Usually at that point, you've wiped out a lot of the debt, You've wiped out a lot of the um issues, the class warfare and so on. And then there's good working together. So those great periods Um such as the Industrial Revolution in the late 1800s, or what we've experienced in the post World War Two period are peaceful and prosperous periods, led largely by the dominant world power Over a period of time since really 1500 And maybe before, but really 1500. Um when the Dutch invented capitalism and what I mean by that, they invest, invented the first capital markets, the first stocks and the first stock market. Um That since then capitalism has been an effective tool for building wealth because it got resources into the hands of inventive people. Um That's when we moved from agriculture to the importance of inventiveness of people. They got resources and then they built that prosperity in that process of doing that. They create wealth gaps naturally. Those who make a lot of money um make a lot more and than those that don't. And it also produces opportunity gaps because for example, of those who earn a lot of money have that wealth. They have the power to educate their Children in a way that others don't. And the gaps grow and also the debts grow at that time when they go around the world with their competitiveness, they earn a lot of money. So for example, the dutch, the dutch empire learned how to build ships. Um that would go around the world. A key ingredient of this improvement in this cycle is the improvement of education and by education. I mean the skills that come from education but also civility, the ability to deal with well together. So the dutch invented ships that could go around the world and they had military power, but they were also very inventive society. 25% of the world's new inventions came from the Dutch. And so as they went around the world, they also brought their currency and they brought their military, they needed their currency, They paid for things. Um and that currency and then the more that happens, the more that becomes a reserve currency and then they have their military so they need their military strength. And so you see it evolved in all of those ways but over a period of time as they become more successful and more expensive they you know they become more expensive um than and newer countries come along like um the UK um then be learned to build ships from a lot the dutch and could do that less expensively. And also when they become more expensive, so less competitive and also the work ethic begins to change. They believe that since they're richer they can enjoy life more, they don't have to work quite as hard. And so you start to see the tilt now you started to see the development of the top and when you have a world reserve currency that allows you to borrow a lot of money because those who want to save, I want to hold your money and that means that they'll lend you money. And so those countries get deeper into debt. So you see that they gradually lose their competitiveness and they and they get themselves into financial circumstances which are not good and they have large wealth gaps which set the stage for downturns and when they have downturns um the first question is do they have enough money and traditionally you know money is resources. Um So you classically see that the coffers are bare that they're spending more money than their then they are earning and they run out of money in the coffers and their granaries are empty rather than stocked so that they give them the buffer and as that deteriorates that that worsens conditions and if they have a rival power that's also challenging them, they see greater internal conflict over wealth and then they have the problems internally and the problems externally which usually results in an internal war or an external war that leads to the change in the to the new world order </s>

<s>[INST] Yeah, it's a fascinating thing how to have heart of a friend that doesn't like sometimes like friends quote unquote or like you know those people who have when the guy and the guy universe when you have a girlfriend that uh you get the girlfriend and then the guy stops hanging out with all of his friends. So like obviously the relationship with a girlfriend is fulfilling or whatever, but like you also wanted to be what she like makes it more enriching to hang out with the guy friends or whatever. Was there any way that that's a that's a that's a that's a fundamental problem in choosing the right mate and probably the fundamental problem in creating the right ai system. Right? What uh let me ask the sexy hot thing on the process right now is GPT three got released with open Ai it's the latest language model. They have kind of an api where you can create a lot of fun applications. I think it's, as people have said, it's probably more hype than intelligence, but there's a lot of really cool Things ideas there with increasing size? You can have better and better performance on language. What are your thoughts about gpt three in connection to your work with the open domain dialogue. But in general like this, learning in an unsupervised way from the internet to generate one character at a time, creating pretty cool text. [/INST] So we partnered up before for the Api launch. So we started working with them when they decided to put together this API and we tried it without fine tuning that. We tried it with fine tuning on our data and we work closely to actually optimize the model for um some of our datasets. It's kind of cool because I think we're we're this polygon polygon for this kind of experimentation space for experimental space for for these models uh to see how they actually work with people because there's no products publicly available to do that. They're focused on open domain conversation. So we can, you know, tests house facebook blender doing or how GPT three doing. So gpt three, we managed to improve by a few percentage points. Like three or four pretty meaningful amount of percentage points. Our main metric, which is the ratio of conversations that make people feel better. And every other metric across across the field got a little boost. Right now, I'd say one out of five responses from replica comes from 53. So our own blender mixes up like a bunch of candidates from different </s><s>[INST] blood you said? [/INST] Well, yeah, just the model that looks at looks at top candidates from different models and picks the most The best one. So right now one of five will come from 53. That's really great. I mean </s><s>[INST] what's the do you have hope for like do you think there's a ceiling to this kind of approach? [/INST] So we've had for a very long time we've used um it's in the very beginning we most it was most of replica was scripted and then a little bit of this fall back part of replica was using a retrieval model. Um and then there's retrieval models started getting better and better and better, which transform is a God a lot better and we're seeing great results. And then with deputy to finally generative models that originally were not very good. And we're the very, very fallback option for most of our conversations. We wouldn't even put them in production. Finally we could use some generative models as well along. And you know, next to our retrieval models and then I will do GPT three, they're almost in part um so that's pretty exciting. I think just seeing how from the very beginning of, You know, from 2015 where the first model started to pop up here and there like segments segments the first papers on that from my observer's standpoint person. It's not, you know, it doesn't really is not really building it, but it's only testing it on people basically and I'm in my product to see how all of a sudden we can use generative dialogue models in production and they're better than others and they're better than scripted content. So we can't really get our scripted, hard coded content anymore to be as good as our anti models exciting. They're much better. Yeah. To your question whether that's the right way to go, I'm again, I'm in the service seat, I'm just um watching this very exciting movie. Um so far it's been stupid to bet against deep learning. So whether increasing the size even more with 100 trillion parameters will finally get us to the right answer, whether that's the way or whether there should be, there has to be some other Again, I'm definitely not an expert anyway, I think, and that's purely my instincts saying that there should be something else as well from memory </s><s>[INST] I know for sure. But the question is, I wonder, I mean, Yeah, and then the argument is for reasoning or for memory it might emerge with more parameters and my larger [/INST] But might emerge, you know, I would never think that to be honest, like maybe in 2017 well, we've been just experimenting with all with all the research that has been coming that was coming out then I felt like there's like we're hitting a wall that there should be something completely different. But then transformer models and then just bigger models and then all of a sudden size matters at that point, it felt like something dramatic needs to happen, but it didn't, and just the size, you know, gave us these results that, to me, are you no clear indication that we can solve this problem pretty soon. </s><s>[INST] Did fine tuning help quite a [/INST] bit? Oh yeah, without it, we it wasn't as good. </s>

<s>[INST] if if I [/INST] can do that again, maybe I never become, you know, high level black belt competitor, but you know, I had four of the things I was doing my life. I also have a family, I have this, I have that, you know what that person was able to accomplish in the martial arts relative to what they were able to put in this phenomenal, you know, other times someone could be a very successful black belt in my mind, be a bump because they could have been a lot more and you know, they could have done more, they could have focused more and there's no shame in deciding that you don't want to do that, but whatever it is that you're, you're invested in, I remember the uh take it uneasy podcast and that I love because you know, I'll just chill out like resting. It's like vacation who wants to go on vacation, I'll go on vacation for a day or two. You want to spend three weeks on vacation. Like I kill myself like get me out of here. Like this is part of this is I'm a waste of life. I'm not doing anything useful. Your </s><s>[INST] technical application right now. [/INST] Right, Well, this is fun though, but it's like a one day vacation. Exactly. But you know, if you had, I'm sure you're thinking about jumping off of the building right now. But if you had, if you had to talk to me for, you know, like three days, I'm sure you do. You probably shut me off the building. I don't blame you, I'll be done, but five hours in. But yeah, but you know, it's like you want to be pushing towards something because otherwise, what's the purpose of being here? You know, it's not just a college, it's doing something useful building growing as a person, helping others do the same if that's within your power at any given time. But I think that's kind of the neat thing about martial arts is it can be many, many different things to many different people, you know, I finally, for instance, was able to get a college degree this this year. That, which I mean, it's not a big deal for most people, but for me it was a big deal because I </s><s>[INST] was going back and finish. [/INST] Yeah. And I never envisioned ever going back and </s><s>[INST] that's a hard step to go back and finish. That's uh it weighs heavy on you if you don't, it's interesting. [/INST] I was just, I was more proud of that than most things I've ever done if I'm honest, you know, and it was neat and I really enjoyed it and it was the process of doing it, but you know, are my academic credentials impressive, like not in the least, but for me it's like it was a big deal for me personally to take that step and to go back and do that and I was proud of the direction and because it would have been easy, like do I need to do it, like no, you know, business, I'll do, okay, I'll try, I'll keep fighting, but I was happy to take the time in between fights when I was when I was wasn't booked for an opponent to do something productive rather than just, I'll just hang out, you know, like I can still train every single day, but I can also train and go to school. People go to the olympics while going to school. I can, I can do martial arts and and go to school. </s><s>[INST] Oh one thing I gotta ask is uh you know, a bunch of women listen to this podcast if they haven't done the jets who think it be kind of intimidating to step on the mat with a bunch of browse uh that like enjoy somehow killing each other. Like how do you succeed in that environment to where you can learn this art, learned how to beat all those people up. [/INST] Um Oh gosh, </s><s>[INST] is there any advice? I mean another way to ask that is like if, if uh any women listening to this are interested in starting digits. Like is there advice for that journey? [/INST] Honestly, I think it's just walking in the door and starting sometimes. I don't know how to respond because I'm not, I don't view myself as typically anxious particularly um uh interactions with other people or new people. Shai Shai is not a word that has been used for me, but if you ask my family and um they joke because our son talks a lot, he's advanced family and they're always like, oh well we know where he gets that from, like because he just doesn't stop talking. He narrates everything he does. Um and so they always tease because that's like I'm known for kind of talking a lot. Um but so I haven't been, typically, I'm not, I don't consider myself a shy person, so for me going into um a new room, a new group of people is, you know, there's always that you don't really know who they are, how they're going to treat you, but but I, I don't have a lot of anxiety with that, so I don't know if that's something that's going to put something up, I don't really know how to address that particular feeling. Um but in terms of all of the rooms I've been and I have popped into jesse James before, I knew Ryan in floor, like I traveled for my job, so in Germany in florida and California in places where I don't know anyone, they don't know me and I have never once had anyone be anything other than, than kind and solicitous and helpful um and long before, when I was a white belt and a bluebell and didn't know anything and I didn't know anyone and um I just think that is a community of people that it's so cool that no matter where you go in the world, um I walked into a gym in Prague one time where only two people spoke english and it was just, yeah, it's weird, </s>

<s>[INST] The following is a conversation with Andrew Huberman, a neuroscientist at stanford working to understand how the brain works, how it can change through experience and how to repair brain circuits damaged by injury or disease. He has a great instagram account at Huberman [/INST] lab </s><s>[INST] where he teaches the world about the brain and the human mind also. He's a friend and an inspiration in that he shows that you can be humble, giving and still succeed in the science world. Quick mention of the sponsor, followed by some thoughts related to the episode eight sleep a mattress that calls itself and gives me yet another reason to enjoy sleep. ECM Rush, the most advanced SEO optimization tool I've ever come across and cash app. The app I used to send money to friends. Please check out these sponsors in the description to get a discount and to support this podcast. As a side note, let me say that I heard from a lot of people about the previous conversation I had with Iran Brooke about objectivism, some people loved it, some people hated it. I misspoke in some parts, was more critical on occasion that I meant to be didn't push on certain points that I should have was under educated or completely unaware about some major things that happened in the past or major ideas out there. I bring all that up to say that if we are to have difficult conversations, we have to give each other space to make mistakes to learn to grow. Taking one or two statements from a three hour podcast and suggesting that they encapsulate who I am I was or ever will be is a standard that we can't hold each other to I don't think anyone can live up to that kind of standard at least I know I can't. The conversation with Iran is mild relative to some conversations that I will likely have in the coming year. Please continue to challenge me but please try to do so with love and with patients. I promise to work my ass off to improve whether I'm successful with that or not. We shall see if you enjoy this thing. Subscribe on youtube, review it with five stars on apple podcast. Follow on Spotify, support on Patreon or connect with me on twitter. A lex Friedman as usual. I'll do a few minutes of ads now and no ads in the middle. I try to make these interesting but I give you time stamps So if you skip, please still check out the sponsors by clicking the links in the description. It's the best way to support this podcast. This show is sponsored by eight sleep and its pod pro mattress. It controls temperature with an app is packed with sensors and can cool down to as low as 55° on east side of the bed separately. It's been a game changer for me with a sleep. I just enjoy sleep and power naps more. I feel like I fall asleep faster and get more restful sleep. Something about a combination of a cool bed and a warm blanket is just amazing. Now if you love your current mattress But still are looking for temperature control, eight sleeps. New pod pro cover as dynamic cooling and heating capabilities onto your current mattress. It can cool down to 55° or heat up to 110° and do so on each side of the bed separately. Also it can track a bunch of metrics like heart rate variability, but cooling alone is honestly worth the money. Go to a sleep dot com slash lex And when you buy stuff there during the holidays you get special savings as listeners of this podcast. That's eight sleep dot com slash lex. This show is sponsored by ECM Rush, which if you look around seems to be one of if not the most respected digital marketing tool out there. It does a lot of stuff including sc optimization of keywords, back links, content creation, social media posts and so on. They have over 45 tools and are trusted by over six million marketers worldwide. I personally don't like Ceo type numbers but that's basically because I'm an idiot with that stuff. I speak from the heart and data be damned. But somebody needs to pay attention to numbers because otherwise you can't make optimal decisions. I believe Heart comes first. Data second, but both are necessary. I started using them just for fun to explore non numeric things like what kind of titles or words connect with people as a writer. That information helps me in moderation of course the amount of data that they put your fingertips is amazing. So if you want to optimize your online presence, check them out at ECM rush dot com slash partner slash flex to get a free month of guru level membership. That's ECM rush dot com slash partner slash lex. This show is presented by cash app, the number one finance app in the app store when you get it used collects podcast Cash up, lets you send money to friends, buy Bitcoin and invest in the stock market with as little as $1. I'm thinking of doing more conversations with folks who work in and around the Cryptocurrency space similar to a I there are a lot of charlatans in the space but there are also a lot of free thinkers and technical geniuses that are worth exploring ideas with in depth and with care if I make mistakes in guest selection or specific details within the conversations, I'll keep trying to improve, correct where I can and also just keep following my curiosity wherever it takes me. So again, if you get cash app from the app store, google play and use the code lex podcast, you get $10 in cash app will also donate $10 to first and the organization that is helping to advance robotics and stem education for young minds around the world and now here's my conversation with Andrew Huberman, you've mentioned that in your lab at stanford, you induce stress by putting people into a virtual reality and having them go through one of a set of experiences. I think you mentioned is on Rogan or with Whitney that scare them. So just uh on a practical psychological level and maybe on a philosophical level, what are people afraid of? What are the fears? What are these fear experiences that you find to be effective? [/INST] Yeah, so it depends on the person obviously. Um and we should probably define fear right? Because you can uh without going too far down the rabbit hole of of defining these things. Um you know, you can't really have fear without stress, but you could have stress without fear and you can't really have trauma without fear and stress, but you could have fear and stress without trauma. So you know, we can start playing the word game and that actually is one of the motivations for even having a laboratory that studies these things is that we really need better physiological Euro, scientific and operational definitions of what these things are in the field of understanding um emotions and states, which is mainly what I'm interested in is very complicated but we can um we can do away with a lot of complicated debate and say in our laboratory what we're looking for to assign it a value of fear is a big inflection and autonomic arousal, So increases in heart rate increases in breathing perspiration, pupil dilation, all the hallmark signature features of the stress response. Uh And in some cases we have the benefit of getting neurosurgery patients where we've got electrodes in their amygdala and insula and the orbitofrontal cortex um down beneath the skull. So these are chronically implanted electrodes were getting multi unit signals and we can start seeing some central features of meaning within the brain. And what's interesting is that as trivial as it might seem in listening to it, almost everybody responds to heights and falling from a hi virtual place with a very strong stress if not fear response. And that's because the visual vestibular apparatus, I write the optic flow and how it links to the, you know, balance semicircular canals of the inner ear is all this technical stuff. But really all of that pulls all your physiology, the feeling that your stomach is dropping, the feeling that you're suddenly you're sweating, even though you're not afraid of falling off this virtual platform. But you feel as if you're following falling, Excuse me because of the optic flow. That one is universal. So we've got a dive with great white sharks experience where you actually exit the cage. We went out this in the real world and brought back 360 video that's built out pretty. So this </s><s>[INST] Is actually 3, 63 60 video [/INST] and this was important to us, right? So when we decided to set up this platform, a lot of the motivation was that a lot of the studies of of these things in laboratories, I don't want to call them lame because I want to be respectful of the people that do this stuff before, but they study fear about, you know, showing subjects a picture of a bloody arm or a snake or something like that, and it just unless you have a snake phobia, it just wasn't creating a real enough experience. So we need to do something where people are going to get injured, but where we can tap into the physiology and that thing of presence of people momentarily, not the whole time, but momentarily for getting there in a laboratory. And so heights will always do it. And I if people want to challenge me on this, I like to point to that movie free Solo, which was wild because you know, it's incredible movie, but I think a lot of its popularity can be explained by a puzzle, which is you knew he was going to live when you walked in the theater or you watched it on at home, you knew before that he he survived. And yet it was still scary that people somehow were able to put themselves into that experience or into Alex's experience enough that they were concerned or worried or afraid at some level. So heights always does it. If we get people who have generalized anxiety, these are people who walk wake up and move through life at a generally higher state of autonomic arousal and anxiety. Then we can tip them a little bit more easily with things that don't necessarily get everyone afraid. Things like claustrophobia, public speaking, that's going to vary from person to person. Um and then if you're afraid of sharks, like my sister friends is afraid of sharks, don't even come to my laboratory because there's, there's a thing about sharks in it that's how terrified some people are of these specific stimulate, but heights gets them every time </s><s>[INST] and I'm terrified of heights. [/INST] It's, you know, when we have, you step off a platform, virtual platform and it's a flat floor in my lab. But we you're up there where </s><s>[INST] you actually allow them the possibility in the virtual world to actually take the leap of faith. [/INST] Yeah, maybe I should describe a little bit of the experiment. So, um without giving away too much in case someone wants to be a subject in one of these experiments, we have them playing a cognitive game. It's a simple lights out kind of game where you're pointing a cursor and turning out lights on a grid, but it gets increasingly complex and it speeds up on them. And um, you know, there's a failure point for everybody where they just can't make the motor commands fast enough and then we surprise people essentially by placing them virtually all of a sudden there's just there on a narrow platform between two buildings and then we encourage them or we cue them with talking them through microphone to continue across that platform to continue the game. And you know, some people, they just won't, they actually will hold get down on the ground and hold onto a virtual beam that doesn't even exist on a flat floor. And so what this really tells us is the power of the brain to enter these virtual states as if they were real. And we really think that anchoring the visual and the vestibular, the balance components of the nervous system are what bring people into that presence so quickly. There's also the potential and we haven't done this yet to bring in 3 60 sound. So the reason we did 3 60 video is that when we started all this back in 2016, a lot of the VR was pretty lame, frankly, it was C. G. I. It just wasn't real enough. But with 3 60 video, we knew that we could get people into this presence where they think they're in a real experience more quickly. And our friend Michael Mueller, who was introduced to because of the project. I reached out to some friends. Michael Mueller is very famous um portrait photographer in Hollywood, but he dives with great white sharks and he leaves the cage and so we worked with him to build a 3 60 video apparatus that we could swim under water with went out to Guadalupe Island Mexico and actually got the experience. It was a lot of fun. It was there was some interesting moments out there of danger, but it would came back with that video and built that for the sharks. And then we realized we need to do this for everything. We need to do it for heights, we need to do it for public speaking for claustrophobia. And what what's missing still is 3 60 sound where 3 60 sound would be, for instance, um, if I were to turn around and there was a, like a giant attack dog there the moment I would turn around and see it, the dog would growl. But if I turn back towards you then it would, it would be silent. So, and that brings a very real element to one's own behavior where you don't know what's going to happen if you turn a corner. Whereas if there's a dog growling behind me and, and I turn around and then I turned back to you and it's still growling, that might seem like more of an impending threat. But um, and sustained threat. But actually it's when you start linking your own body movements to the experience. So when it's closed loop where my movements and choices are starting to influence things and they're getting scarier and scarier. That's when you can really drive people's nervous systems down these paths of high states of stress and fear. We don't want to traumatize people obviously. But we also, we also study a number of tools to that allow them to calm themselves in these environments. So the short answer is heights, </s><s>[INST] heights. [/INST] Well, from </s><s>[INST] a psychology from neuroscience perspective, this whole construction that we've developed is fascinating. We did this a little bit with autonomous vehicles. So to try to understand the decision making process of a pedestrian when they crossed the road and trying to create an experience of a car, you know, that can run you over. So there's a danger there. I was so surprised how real that whole world was. And the graphics that we built wasn't ultra realistic or anything, but I was still afraid of being hit by a car. Everybody we tested were really afraid of being hit by that car. [/INST] Even though it was all a simulation, it </s><s>[INST] was all a simulation. It was uh it was kind of a boxy actually. I mean, it wasn't like ultra realistic simulation and fascinating [/INST] looms and heights. So any kind of depth. We're just programmed to um to not necessarily recoil, but to be cautious about that edge. And that depth and then looms things coming at us that are getting larger. There are looming sensing neurons even in the retina very, very early stage of visual processing and incidentally, uh, the way Mueller and you know folks learned how to not get eaten by great white sharks when you're swimming outside the cage is as they start lumbering in, you swim toward them and they get very confused when you loom on them because clearly you're smaller, clearly they could eat you if they wanted to. But there's something about forward movement toward uh any creature that that creature questions whether or not it would be a good idea to generate forward movement toward you. And so that's actually the survival tool of these cage exit white shark divers. </s><s>[INST] Are you playing around with like one of the critical things for the autonomous vehicle research is you couldn't do 360 video because the there's a game theoretic, there's an interactive element that's really necessary. So Maybe people realize is maybe they don't. But 360 video you obviously well it's actually not that obvious to people, but you can't change the reality that you're watching. That's right. So, uh but you find that that's like is there something fundamental about fear and stress that the interactive element is essential for? Or do you find you can you can arouse people with just the video? [/INST] Great question. Um it works best to use mixed reality. So we have a snake stimulus. I personally don't like snakes at all. I don't mind spiders. We also have a spider stimulus, but like snakes, I just don't like them. There's something about the slithering and that it just creates a visceral response for me. Um some people not so much and they have lower levels of stress and fear in there, but one way that we can get them to feel more of that is to use mixed reality where we have an actual physical bat and they have to stomp out the snake as opposed to just um walk to a little safe corner, which then makes the snake disappear. That tends to be not as stressful as if they have a physical weapon. And so you've got people in there, you know, banging on the floor against this thing and there's something about engaging that makes it more of a more of a threat. Now, I should also mention that we always get the subjective report from the subject of what they experience because we never want to project our own ideas about what they were feeling, but that's the beauty of working with humans as you can ask them how they feel and humans aren't great at explaining how they feel. Um but it's a lot easier to understand what they're saying than a mouse or a Macaque monkey is saying. Um so it's the best we can do is language plus these physiological and neuro physiological signals, Is </s><s>[INST] there something you've learned about yourself about your deepest fears? Like you said snakes? Is there something that, like, if I were to torture you, I'm some Russian, so, you know, I always kind of think how can I murder this people, that this person entered the room, but also how can I torture you to get some information out of you, Whoa! Where would I go [/INST] with? It's interesting you should say that I never considered myself claustrophobic, but um because I don't mind small environments provided there well ventilated. But before Covid I started going to this Russian Banya. Um you know, in which you and I have never been to a banya. So you know the whole experience of really, really hot sauna and what do they call it? The plot? So they're hitting you with the leaves and and it gets really hot and humid in there. And there were a couple of times where I thought, okay, this thing is below ground, it's in a city where there are a lot of earthquakes, like if this place crumbled and we were stuck in here and I'd start getting a little panicky and I realized, I'm like I don't like small confined spaces with poor ventilation. So I realized, I think I have some claustrophobia and I wasn't aware of that before. So I put myself into our own claustrophobia stimulus which involves getting into an elevator. Um and with a bunch of people, virtual people and the elevator gets stalled and at first you're fine, you feel fine. But then as we start modulating the environment and we actually can control levels of oxygen in the environment if we want to. Uh it is really uncomfortable for me and I never would have thought, you know, I fly, I'm comfortable on planes, but it is really uncomfortable. And so I think I've unhatched a bit of a claustrophobia. </s><s>[INST] Yeah. For me as well, probably that one that one's pretty bad. The heights I tried to overcome. So I went skydiving to try to overcome the fear of heights. Well, they didn't help. [/INST] Did you jump out? Yeah, </s><s>[INST] yeah, jumped out. But it was, it was, it was fundamentally different experience and I guess there could be a lot of different flavors of fear of heights maybe. But the one I have didn't seem to be connected to jumping out of a plane is a very different because like once you accept that you're going to jump, then it's it's a different thing. I think what I'm afraid of is the moments before it is the, is the scariest part. And I don't think that's emphasized in the skydiving experience as much. And also just the acceptance of the fact that it's going to happen. So once you accept that it's going to happen, it's not a scary, it's the fact that it's not supposed to happen and it might that's the scary part, I guess I'm not being eloquent in the description, but there's something about skydiving that was actually philosophically liberating. I was, I was like, wow, it was the possibility that you can walk on the surface and then at a certain point there's no surface anymore to walk on and it's also in the world becomes three dimensional and there's this freedom of floating that the concept of like of earth disappears for a brief few seconds. I don't know those, those [/INST] wild, </s><s>[INST] that was wild, but I'm still terrified of heights. So I mean 11 thing I want to ask, just unfair because it's so fascinating is have you learned anything about what it takes to overcome fears? [/INST] Yes. And that comes from to from a, you know, research study standpoint to parallel tracks of research. One was done actually mice because we have a mouse lab also where we can program in different brain areas and trying to figure out what interesting brain areas we might want to probe around in humans and graduate school in my lab, she's now at caltech. Um, lindsey Soleil Published a paper back in 2018 showing that. Well, what at first might seem a little bit obvious, but the mechanisms are not, which is that they're really three responses to fear. You can pause, you can freeze essentially, um you can retreat, you can back up or you can go forward and there's a single hub of neurons in the midbrain. It's actually not the midbrain, but it's in the middle of the thalamus, which is a four brain structure. Uh and depending on which neurons are active there, there's a much higher probability that a mouse or it turns out, or a human will advance in the face of fear or will pause or will retreat. Now that just assigns a neural structure to a behavioral phenomenon. But what's interesting is that it turns out that the lowest level of stress or autonomic arousal is actually associated with the pausing and freezing response. Then as the threat becomes more impending. And we used visual looms in this case the retreat response has a slightly higher level of autonomic arousal and stress. So think about playing hide and go seek. You're trying to stay quiet in a closet that you're hiding if you're very calm, it's easy to stay quiet and still as your level of stress goes up, it's harder to maintain that level of quiet and stillness. You see this also in animals that are stalking a cat will chatter, its teeth. That's actually sort of top down inhibition and trying to restrain behavior. So the freeze response is actually an active response, but it's fairly low stress. And what was interesting to us is that the highest level of autonomic arousal was associated with the forward movement toward the threat. So in your case, um jumping out of the plane. However, the forward movement in the face of threat was linked to the activation of what we call collateral, which means just a side connection, literally a wire in the brain that connects to the dopamine circuit for reward. And so when one safely and adaptive lee meaning you survive moves through a threat or toward a threat. It's rewarded as a positive experience. And so the key, it actually maps very well the cognitive behavioral therapy and a lot of the existing treatments for trauma is that you have to confront the thing that makes you afraid so otherwise you exist in this very low level of river batory circuit activity where the circuits for autonomic arousal are humming and they're coming more and more and more. And we have to remember that that stress and fear and threat were designed to agitate us so that we actually move. So the reason I mention this is I think a lot of times people think that the maximum you know stress response and the response is to freeze in to lock up but that's actually not the maximum stress response. The maximum stress response is to advance but it's associated with reward. It has positive valence. So so there's this kind of, everyone always thinks about the bell, you know the sort of hump shape curve for you know at low levels of arousal, performance is low and as it increases performance goes higher and then it drops off as you get really stressed. But there's another bump further out the distribution where you perform very well under very high levels of stress. And so we've been spending a lot of time in humans and in animals exploring what it takes to get people comfortable to go to that place and also to let them experience how their heightened states of cognition. There there's um changes in time perception that allow you to evaluate your environment in fast at a faster frame rate, essentially this is the matrix as a lot of people think of it. Um but we tend to think about fear is all the low level stuff where things aren't worked out, but there are many um there are a lot of different features to the fear response and so we think about it quantitatively and we think about it from a circuit perspective in terms of outcomes and we try and weigh that against the threat. So we never want people to put themselves in unnecessary risk. But that's where the VR is fun because you can push people hard without risk of physically injuring them. </s><s>[INST] And that's uh like you said the little bump that that seems to be a very small fraction of the human experience. Right? So it's kind of fascinating to study it because most of us move through life without ever experienced in that kind of focus. [/INST] Well, everything is in a peak state there. I really think that's where optimal performance lies. </s><s>[INST] There's so many interesting words here, but what's performance and what's optimal performance? We're talking about mental ability to what to perceive the environment quickly, to make actions quickly. What's optimal performance. [/INST] Yeah, well it's very subjective and it varies depending on um task and environment. So one way we can make it a little bit more operational and concrete is to say um there is a sweet spot if you will where the level of internal autonomic arousal, ak stress or alertness, whatever you wanna call it is ideally matched to the speed of whatever challenge you have to be facing in the outside world. So we all have um perception of the outside world is extra reception and the perception of our internal real estate interception. And when those two think when interception and Xterra reception are matched along a couple dimensions, performance uh tends to increase, it tends to be in a optimal range. So for instance, if you're, I don't play guitar but I know you play guitar. So let's say you're trying to learn something new on the guitar, I'm not saying that being in these super high states of activation are the best place for you to be in order to learn. It may be that you your internal arousal needs to be at a level where your analysis of space and time has to be well matched to the information coming in and what you're trying to do in terms of performance in terms of playing chords and notes and so forth. Now in these cases of high threat where things are coming in quickly and animals and humans need to react very quickly, the higher your state of autonomic arousal the better because you're slicing time more. Finally, just because of the way the economic system works, it, you know, the people dilation for instance and movement of the lens essentially changes your optics that's obvious. But in with the change in optics is a change in how you been time and slice time, which allows you to get more frames per second readout with the guitar learning for instance. It might actually be that you want to be almost sleepy almost in a uh kind of drowsy state to be able to and I don't play music so I can I'm guessing here but send some of the nuance in the chords or the ways that you're to be relaxed enough that your fingers can follow an external cues so matching the movement of your fingers to something that's pure extra reception. And so there is no perfect autonomic state for performance. This is why I don't favor terms like flow because they're not well operationally defined enough. But I do believe that optimal or peak performance is going to rise when internal state is ideally matched to the spacetime features of the external demands. So </s><s>[INST] there's some slicing of time that happens and then you're you're able to adjust the slice time more finally, or more les finally, in order to adjust to the stimulus, the dynamics of the stimulus. What about the the realm of ideas. So like, you know, I'm a big believer, this guy named cal Newport's who wrote a book about [/INST] deep work </s><s>[INST] that Yeah, he's great. Uh he I mean one of the nice things, I've always practice deep work, but he it's always nice to have words put to the concepts that you've practiced it somehow makes them more concrete and allows you to uh to get better. It turns it into a skill you can get better at. But you know, I also value deep thinking where you I think it's almost meditative. You think about a particular concept for long periods of time, the programming you have to do that kind of thing for. Yes, you have to hold this concept. Thank you. Like you you hold it and then you take steps with it, you take further steps and you're you're holding relatively complicated things in your mind as you're thinking about them. And there's a lot of I mean the hardest part is there's a frustrating things like you take a step and it turns out to be the wrong directions. You have to calmly turn around and take a step back and then it's your kind of like exploring through the space of ideas. Is there something about your study of optimal performance that can be applied to the act of thinking as opposed to action? [/INST] Well, we haven't done too much work there. But what um but I think I can comment on it from a neuroscience which is really all I do is uh Well I mean we do experiments in the lab but um looking at things through the lens of neuroscience. So what you're describing um can be mapped fairly well to working memory. Just keeping things online and updating them as they change in information is coming back into your brain. Uh Jack Feldman who I'm a huge fan of and um fortunate to be friends with is a professor at U. C. L. A. Works on respiration and breathing. But he has a physics background and um and so he thinks about respiration and breathing in terms of ground states and how they modulate other states. Very very interesting. I think important work. Jack has an answer to your question So I'm not going to get this exactly right because this is lifted from a coffee conversation that we had about a month ago. But so um apologies in advance for the but I think I can get mostly right. So we were talking about this about how the brain updates cognitive states depending on demands and thinking in particular. And he used an interesting example, I'd be curious to know if you agree or disagree. Uh He said you know most great mathematics is done by people in there late teens and twenties even you could say early twenties sometimes into the late twenties but not much further on. Maybe I just insulted some mathematicians. </s><s>[INST] That's true. [/INST] And I think that it demands his argument was um there's a tremendous demand on working memory to work out theorems in math and to keep a number of plates spinning so to speak mentally and run back and forth between them, updating them in physics Jack said and I mean I think this makes sense to me too that there's a reliance on working memory but an increased reliance on some sort of deep deep memory and deep memory stores probably stuff that's moved out of the hippocampus and for draining into the cortex and is um more some episodic and declarative stuff, but really so you're you're pulling from your library basically, it's not all RAM, it's not all working memory and then in biology that and physicists tend to have very active careers into their thirties and forties and fifties and so forth. Um sometimes later and that in biology you see careers that have a much longer arc kind of these protracted careers, often people still in their sixties and seventies doing doing really terrific work, not always doing it with their own hands because the people in the labs are doing them of course, but um and that work does tend to rely on insights gained from having a very deep knowledge base where you can remember a paper and or maybe a figure in a paper you could go look it up if you wanted to but it's very different than the working memory of the mathematician and so when you're talking about coding or being in that tunnel of thought and trying to iterate and keeping a lot of plates spinning. It speaks directly to working memory. My lab hasn't done too much of that working memory. But we are pushing working memory when we have people do things like these simple lights out tasks while they're under. We can increase the cognitive load by increasing the level of autonomic arousal to the point where they start doing less well. And you know, everyone has a cliff. This is what's kind of fun. We've had, you know, seal team operators come to the lab, we have people from other units in the military very, you know, we've had a range of of intellects and backgrounds and all sorts of things and everyone has a cliff and those cliffs uh sometimes show up as a function of the demands of speed of processing or how many things you need to keep online. I mean we're all limited at some point in the number of things we can keep online. So what you're describing is very interesting because I think it has to do with how narrow or broad the information set is because and I don't program, I'm not an active programmer. So this is a regime I don't really fully know. So I don't want to comment about it in any way. Uh you know, doesn't suggest that, but I think that what you're talking about is top down control. So this is prefrontal cortex keeping every bit of reflexive circuitry at bay. The one that makes you want to get up and use the restroom the one that makes you want to check your phone all of that. But also running these interior thalamus to prefrontal cortex loops which we know are very important for working memory. </s><s>[INST] Yeah. Let me try to think that it's a little bit so reducing the process of thinking to working memory access. It's tricky. He's probably ultimately correct. But if I were to say some of the most challenging things that an engineer has to do and the scientists, scientific thinker, I would say it's kind of depressing to think that we do that best in our twenties but is uh this kind of first principles thinking step of saying [/INST] you you're </s><s>[INST] accessing the things that you know and then saying, well let me how do I do this differently than I've done it before this this weird like stepping back like is this right? Let's try it this other way that that's the most mentally taxing step is like you've gotten quite good at this particular pattern of how you solve this particular problem. So there's a there's a pattern recognition first, you're like okay, I know how to build the thing that solves this particular problem in programming, say. And then the question is but can I do it much better and I don't know if that's I don't know what the hell that is. I don't know if that's accessing working memory that's that's almost access, maybe it is accessing memory in a sense. It's trying to find similar patterns in a totally different place that could be projected onto this, but you're it's you're not quarrying a fact. You're currying like functional things like yes, patterns. I mean, you're running out your [/INST] your testing algorithms. Yeah, right? You're testing algorithms. Ice. So I want to just um because I know uh some of the people listening to this and you how have basis and you know, scientific training and have scientific training. So I want to be clear. I think we can be correct about some things like The role of working memory in these kinds of processes without being exhaustive. We're not saying that the only thing, you know, we could be correct but not assume that that's the only thing involved. And I mean, neuroscience, let's face it is still in its infancy. I mean, we probably know 1% of what there is to know about the brain. Um you know, we've learned so much and yet there may be global states that underlie this that make prefrontal circuitry work differently than it would in a in a different regime even time of day. I mean there's a lot of mysteries about this, but so I just want to make sure that we we sort of are we're aiming for precision and accuracy, but but we're not gonna we're not gonna be exhausted. So there's a difference there and I think uh you know, sometimes in the vastness of the internet that gets forgotten. Um So the other is that um you know we think about um you know we think about these operations uh at you know really focused keeping a lot of things online. But what you're describing is actually um it it speaks to the very real possibility. Probably the with certainty there's another element to all this which is when you're trying out lots of things in particular, lots of different algorithms. You don't want to be in a in a state of very high autonomic arousal. That's not what you want because the higher level of autonomic arousal and stress in the system, the more rigidly you're going to analyze space and time and what you're talking about is playing with spacetime dimensionality. And I want to be very clear. I mean I'm the son of a physicist. I am not a physicist. When I talk about space and time, I'm literally talking about visual space and </s><s>[INST] how long it takes for my finger to [/INST] move from this point to this point you </s><s>[INST] are facing a tiger and trying to figure out how to avoid being eaten by the tiger. [/INST] And that's primarily facing retirement by the visual system in humans. We don't walk through space for instance like a cent town would and look at three dimensional sent plumes. You know when a scent hound goes out in the environment, they have depth to their odor trail the odor trails they're following and they don't think about them. We don't think about odor trails, you might say, oh well the smell is getting more intense, but they actually have three dimensional odor trail. So they see a cone of odor c of course with their nose or their olfactory cortex. We do that with our visual system and we parse time often subconsciously with, mainly with our visual system also with our auditory system. And this shows up for the musicians out there. Metro gnomes are a great way to play with this. Um you know, bass drumming, when the frequency of bass drumming changes, your perception of time changes quite a lot. So in any event, space and time are linked in through the sensory apparatus through the eyes and ears and nose and um probably through taste to and through touch um for us, but mainly through vision. So when you drop into some coding or iterating through a creative process or trying to solve something hard, you can't really do that well if you're in a rigid um high level of autonomic arousal because you're plugging in algorithms that are in this space regime, This time regime matches its space time matched. Whereas creativity always think the lava lamp is actually a pretty good example, even though it has these counterculture new agey connotations because you actually don't know which direction things are going to change. And so in drowsy states, sleeping and drowsy states, space and time become dislodged from one another somewhat and they're very fluid and I think that's why a lot of solutions come to people after sleep and naps. And this could even take us into a discussion if you like about psychedelics and what we now know for instance That people thought that psychedelics work by just creating spontaneous bursting of neurons and hallucinations but that the five h. To see A. And two C. And two A receptors which are the main sites for things like LSD and psilocybin and some of the other. Um Hello. Use the ones that create hallucinations. The drugs that create hallucinations. The most of those receptors are actually in the um collection of neurons that in case the thalamus which is where all the sensory information goes into a structure called the thalamic particular nucleus. Um and it's an inhibitory structure to make sure that when we're sitting here talking that I mainly focused on whatever I'm seeing visually that I'm essentially eliminating a lot of sensory information under conditions where people take psychedelics and these particular serotonin receptors are activated. That inhibitory shell, it's literally shaped like a shell starts losing its ability to inhibit the passage of sensory information. But mostly The effects of psychedelics are because the lateral connectivity in layer five of cortex across cortical areas is increased. And what that does is that means that the space time relationship for vision like moving my finger from here to here, very rigid space time relationship right If I slow it down it's slower obviously but there's a prediction that can be made based on the neurons in the retina, in the cortex on psychedelics. This could be very strange experience. But the auditory system has one that's slightly different space time and they're matched to one another, deeper circuits in the brain. The olfactory system has a different space time relationship to it. So under conditions of of these increased activation of these serotonin receptors, space and time across sensory area starts being fluid. So I'm no longer running the algorithm for moving my finger from here to here and making a prediction based on vision alone. I'm now this is where people talk about um hearing mhm sites right? You start linking this might actually make a sound in a psychedelic state. Now I'm not suggesting people run out and do psychedelics because it's very disorganized. But essentially what you're doing is you're mixing the algorithms and so when you talk about being able to access new solutions, you don't need to rely on psychedelics if people choose to do that, that's their business. But in drowsy states this lateral connectivity has increased as well. The shell of the thalamus shuts down and what these are through these so called ponds, articulate occipital waves and what's happening is you're getting whole brain activation at a level that you start mixing algorithms and so sometimes I think solutions come not from being in that narrow tunnel of spacetime and strong activation of working memory in china. Well iterate if this then this very strong deductive and inductive thinking and working from first principles, but also from states where something that was an algorithm that never you never had in existence before suddenly gets lumped with another algorithm and all of a sudden a new possibility comes to mind. And so space and time need to be fluid and space and time need to be rigid in order to come up with something meaningful. And I realize I'm riffing long on this. But this is why I think, you know, there was so much interest a few years ago with Michael Pollan's book and and other things happening about psychedelics as a pathway to exploration and all this kind of thing. But the real question is what you export back from those experiences because dreams are amazing. But if you can't bring anything back from them, they're just amazing. </s><s>[INST] I wonder how to experiment with the mind without without any medical assistance first. Like, you know, I I pushed my mind all kinds of directions. I definitely want to I did uh shrooms a couple of times. I definitely want to uh figure out how I can experiment with with psychedelics. I'm talking to rick Dublin, I think Dublin uh soon I went back and forth. So he does all these studies and psychedelics and he keeps ignoring the parts of my email that asks like how do I participate in these studies? [/INST] There are some legality issues. I mean conversation, I would be very clear. I'm not saying anyone can do psychedelics. I think that drowsy states and sleep states are are super interesting for accessing some of these more creative states of mind. Hypnosis is something that my colleague, David Spiegel, associate chair of psychiatry at stanford works on where? Also again, it's a unique state because you have narrow context. So this is very um kind of tunnel vision and yet deeply relax. Excuse me, deeply relaxed where new algorithms, if you will can start to surface um strong state for inducing neural plasticity. And I think, you know, so if I had a part of a group um that it's called the liminal collective as a group of people that get together and talk about um just wild ideas, but they try and implement. Um and it's a it's a really interesting group, some people from the military, from larger tech and some other backgrounds, academic backgrounds. And I was asked, you know, what would be um if you could create a tool, you just had a tool like your magic wand wish for the day, what would it be? I thought it would be really interesting if someone could develop psychedelics that have on off switches. So you could go into a psychedelic state Very deeply for 10 minutes. But you could launch yourself out of that state and place yourself into a linear real world state very quickly so that you could extract whatever it was that happened in that experience and then go back in if you want it. Because the problem with psychedelic states and dream states is that first of all, a lot of the reason people do them is they're lying. They say they want plasticity and they want all this stuff. They want a peak experience inside of an amplified experience. So they're kind of seeking something unusual. And I think we should just be honest about that because a lot of times they're not trying to make their brain better. They're just trying to experience something really amazing. But the problem is space and time are so unlocked in these states. Just like they are in dreams that you can really end up with a whole lot of nothing. You can have an amazing amplified experience housed in an amplified experience and come out of that thinking you had a meaningful experience when you didn't bring anything back. </s><s>[INST] You didn't bring anything back. All you have is a fuzzy memory of having a transformational experience. But you don't actually have you had tools to bring back just sorry, actually, actually, concrete ideas to bring back. Yeah, it's interesting. You should yeah, I wonder if it's possible to do that with the mind to be able to hop back and forth. Like I think that's [/INST] where the real power of, you know, adjusting states is going to be. It probably will be with devices. Um, I mean maybe it will be done through pharmacology. It's just that it's hard to do on off switches in human pharmacology that we have them for animals. I mean we, we have, you know, cree flip combinations and we have um, you know, channel options and halo route options and um, all these kinds of things. But to do that work in humans is tricky. But I think you could do it with virtual reality, augmented reality and other devices that bring more of the somatic experience into it. </s><s>[INST] You're of course the scientists are studying humans as a collective. I tend to be just one person's scientist of just looking at myself and you know, I play when these deep thinking deep work sessions, I'm very cognizant like in the morning that there's times in my mind is so like eloquent at being able to jump around from ideas and hold them all together. And I, I'm almost like I stepped back from a third person perspective and enjoy that whatever that mine is doing. I do not waste those moments. I, and I'm very conscious of this like little creature that woke up that's only awake for if we're being honest, maybe a couple hours a day early part of the day for the early part of the day, not always early part of the day for me is a very fluid concept. You're one of the yeah, yeah, you're being single one of the problems single and no meetings. I don't schedule any meetings. I I will, I've been living like a 28 hour day so I like uh drifts so it's all over the place, but after a traditionally defined full night's sleep, uh whatever the heck that means, I find that like in in those moments there's a clarity of mind that's just that everything is effortless and it's the, it's the deepest dives intellectually that I make and I I'm cognizant of it and I try to bring that to the other parts of the day. They don't have it And treasure them even more in those moments because they only last like five or 10 minutes. Because of course in those moments you want to do all kinds of stupid stuff that are completely, is worthless. Like check social media or something like that. But those are the most precious things in, in, in an intellectual life is those mental moments of clarity and I wonder I'm learning how to control them. I think caffeine is somehow involved. I'm not sure exactly. Well, [/INST] because if you learn how to titrate caffeine, everyone's slightly different with us what they need. But if you learn to titrate caffeine with time of day and the kind of work that you're trying to do, You can bring that autonomic arousal state into the close to perfect place and then you can tune it in with, you know, sometimes people want a little bit of background music, sometimes they want less these kinds of things. The early part of the day is interesting because the one thing that's not often discussed is this transition out of sleep. So there's a book um I think it's called Winston Churchill's nap and it's about naps and and the transition between wake and sleep as a valuable period. Um I've a long time ago um someone who I respect a lot, I was mentoring me, said be very careful about bringing in someone else's sensory experience early in the day. So when I wake up I'm very drowsy, I sleep well but I don't emerge from that very quickly. I need a lot of caffeine to wake up and whatnot. But there's this concept of getting the download from sleep, which is, you know, and sleep. Your you were essentially expunging the things that you don't need. The stuff that is meaningless from the previous day, but you were also running variations on these algorithms of whatever it is you're trying to work out in life on short time scales like the previous day and long time scales like your whole life and those lateral connections in layer five of the of the neocortex are very robustly um active and across sensory areas and you're running a an algorithm or, you know, a brain, it's a brain state, they will be useless in waking, you wouldn't get anything done, you'd be the person talking yourself in the hallway or something about something that no one else can see. But in those states you do that. The theory is that you arrive at certain solutions and those solutions will reveal themselves in the early part of the day unless you interfere with them by bringing in social media is a good example of you immediately enter somebody else's spacetime sensory relationship. Someone is the conductor of your thoughts in that case. And so many people have written about this um what I'm saying isn't entirely new, but allowing the download to occur in the early part of the day and and asking the question, am I more in my head or extra um I'm more of an introspective or extra receptive mode and depending on the kind of work you need to do, it sounds like for you, it's very introspective in the end. Very you've got a lot of thinking going on, a lot of computing going on, allowing yourself to transition out of that sleep state and arrive with those solutions from sleep and plug into the work really deeply and then and only then allowing things like music news, social media doesn't mean you should talk to loved ones and see faces and things like that, but some people have taken this to the extreme when I was a graduate student at Berkeley, there was a guy. Um there is a professor, brilliant, odd, but brilliant. Um who was so fixated on this concept that he wouldn't look at faces in the early part of the day because he just didn't want to be anything else to impact him. Now, he didn't have the most um rounded life I suppose. But if you're talking about cognitive performance, this could actually be very beneficial. </s><s>[INST] You said so many brilliant things. So one if you read books that described the habits of brilliant people, like writers, they do control that sensory experience in the in the in the hours after a week. Like many writers, you know, they have a particular habit of several hours early in the morning of actual writing. They do, I'm doing anything else for the rest of the day, but they control, they're very sensitive to noises and so on. I think they make it very difficult to live with them. I tried to I'm definitely like that. Like I can I love to control the sensory uh how much information is coming in. There's something about the peaceful, just everything being peaceful at the same time. And we were talking a mutual friend of Whitney cummings who uh has a has a mansion, a castle on top of a cliff and in the middle of nowhere, she actually purchased her own island. Uh she wants silence. She wants to control how much uh sound is coming in. [/INST] She's very sensitive to sound an environment, beautiful home and environment, but like clearly puts a lot of attention into just into details. Yeah. </s><s>[INST] And very creative. Yeah. And then, yeah, that allows for creativity to flourish. I'm also, I don't like that feels like a slippery slope. So I enjoy introducing uh noises and signals and training my mind to be able to tune them out. Because I feel like you can't always control the environment self perfectly because uh because your mind gets comfortable with that. I think it's a skill that you want to learn to be able to shut it off. Like I often go to like back before Covid to a coffee shop. It really annoys me when their sounds and voices and so on. But I feel like I can train my mind to to block them out. So it's a balance, I think. [/INST] Yeah. And I think, um, you know, two things come to mind. Um as you're saying this. Um first of all, yeah, I mean, we're talking about What's best for work is not always what's best for, you know, completeness of life. I mean, you know, autism is probably many things like when autism just like feet, they're probably 50 ways to get a fever. There. Probably 50 ways to that. The brain can create what looks like autism or what people call autism. There's an interesting set of studies that have come out of David guarantees lab at Harvard med. Um looking at these are mouse mutants where, um, these are models for autism where nothing is disrupted in the brain proper and in the central nervous system, but the sensory app the sensory neurons, the ones that innovate the skin and the ears and everything are are hypersensitive. And this maps to a mutation in certain forms of human autism. So this means that the the overload of sensory information and sensory experience that a lot of autistics feel like they can't tolerate things and then they get the stereotype behaviors, the rocking and the kind of the shouting. It. You know, we always thought of that as a brain problem in some cases it might be, but in many cases it's because they just can't, they seem to have a it's like turning the volume up on every sense. And so they're overwhelmed and none of us want to become like that. I think it's very hard for them and it's hard for their parents and so forth. So I like the coffee shop example because um the way I think about trying to build up resilience, you know, physically or mentally or otherwise is one of um I guess we could call it like, I like to call it limbic friction. That's not a real scientific term and I acknowledge that I'm making it up now because I think it captures the concept, which is that, you know, we always hear about resilience. It makes it sound like, oh, you know, under stress, where everything's coming at, you, you're gonna stay calm, but there's another, you know, so limbic the limbic system wants to pull you in some direction. Typically in the direction of reflexive behavior. And the prefrontal cortex through top down mechanisms has to suppress that and say no we're not going to respond to the banging of the coffee cups behind me or I'm going to keep focusing. That's pure top down control. So the limbic friction is high in that environment. You put yourself into a high limbic friction environment means that the prefrontal cortex has to work really hard. But there's another side to limbic friction to which is when you're very sleepy, there's nothing incoming. It can be completely silent and it's hard to engage in focus because you're drifting off, you're getting sleepy so their limbic friction is high, but for the opposite reason autonomic arousal is too low. So they're turning on netflix in the background or looping a song might boost your level of alertness that will allow top down control to be in the exactly the sweet spot you want it. So that is why earlier I was saying it's all about how we feel inside relative to what's going on on the outside. We're constantly in this. I guess one way you can envision it spatially, especially if people are listening to this. Just an audio is I like to think about it kind of like a glass bar bell where one sphere of perception and attention can be on what's going on with me. And one sphere of attention can be on what's going on with you or something else in the room or in my environment. But those this barbell isn't rigid, it's not really glass would plasma work here. I don't know anything about plasma. Sorry, I don't know. Okay, But so imagine that this thing can contort the size of the globes at the end of this barbell can get bigger or smaller. So let's say, I close my eyes and I bring all my experience into what's going on in through interception internally. Now it's as if I've got two orbs of perception just on my internal state, but I can also do the opposite and bring both orbs of perception outside me. I'm not thinking about my heart rate or my breathing, I'm just thinking about something I see and what you're start to realize as we kind of use this spatial model is that two things. 1 is that it's very dynamic and that the more relaxed we are, the more these two orbs of attention, the two ends of the barbell can move around freely. The more alert, we are, the more rigid they're going to be tethered in place. And that was designed so that if I have a threatened my environment, it's tethered to that threat, I'm not going to be if something's coming to attack me, I'm not gonna be like, oh my breathing cadence is a little bit quick, That's not how it works. Why? Because both orbs are linked to that to that threat. And so my behavior is now actually being driven by something external even though I think it's internal. And so I don't want to get too abstract here because I'm a neuroscientist. I'm not a theorist, but when you start thinking about models of how the brain works, I mean brain works. Excuse me, There are only really three things that neurons do there either sensory neurons, their motor neurons or their modulating things and The models of attention and perception that we have now 2020 tell us that we've got into reception and extra reception. They're strongly modulated by levels of autonomic arousal and that if we want to form the optimal relationship to some task or some pressure or something, whether or not it's sleep an impending threat or coding, we need to adjust our internal space time relationship with the external spacetime relationship and I realize I'm repeating what I said earlier, but we can actually assign circuitry to this stuff. It mostly has to do with how much limbic friction there is how much you're being pulled to some source. That source could be internal. If I have pain, physical pain in my body, I'm going to be much more introspective than I am. Extra receptive, You could be talking to me and I'm just gonna be thinking about the pain. It's very hard. And the other thing that we can link it to is top down control, meaning anything in our environment that has a lot of salience will tend to bring us into more Xterra reception than into reception. And again I don't want to litter the conversation with just a bunch of terms. But um what I think it can be useful for people is to do essentially you've done flexes to start developing an awareness when I wake up and I mostly in a mode of interception where extra reception when I work well. Is that what is working well look like from the perspective of autonomic arousal, how alert or calm I am. I what kind of balance between internal focus and external focus is there and to sort of watch this process throughout the day. Can </s><s>[INST] you just briefly, because you used this term a lot and be nice to try to get a little more color to it, which is interception and extra reception. Uh what are we exactly talking about? So like what's included in each category? And how much overlap is there [/INST] into reception? Would be uh an awareness of anything that's within the confines or on the surface of my skin that I'm sensing </s><s>[INST] literally physiological, [/INST] physiological, like within the boundaries of my skin. And probably touch to the skin as well. Extra reception would be perception of anything that's X. Beyond the reach of my skin. So that that bottle of water um a sent um a sound although and this can change dramatically actually if you have headphones in you tend to hear things in your head if as opposed to a speaker in the room, there's actually the basis of ventriloquism. So there are beautiful experiments done by Greg reckon zone of at UC Davis looking at how auditory and visual cues are matched and we have an array of speakers and you can this will become obvious as I say it. But you know obviously the ventriloquist doesn't throw their voice, what they do is they direct your vision to a particular location and you think the sound is coming from that location. And there are beautiful experiments that Greg and his colleagues have done where they suddenly introduce an auditory visual mismatch. And it freaks people out because you can actually make it seem from a perception standpoint as if the sound arrived from the corner of the room and hit you like physically and people will recoil. And so sounds aren't getting thrown across the room, they're still coming from the defined location, an array of speakers. But this is the way the brain creates these internal representations. And again, not to I don't want to go down a rabbit hole, but um I think as much as I'm sure the listeners appreciate this but you know, everything in the brain is an abstraction, right? I mean they're they're the sensory apparatus, there are the eyes and ears and nose and skin and tasting all that are taking information and with interception. Taking information from sensors inside the body. The enteric nervous system for the gut. I've got sensory neurons that integrate my liver um etcetera. Taking all that. And the brain is abstracting that in the same way that if I took a picture of your face and I handed it to you and I'd say that's you you'd say yeah that's me. But if I were an abstract artist I'd be doing a little bit more of what the brain does. Where if I took a pad and paper, maybe I could do this because I'm a terrible artist and I could just mix it up. And let's say I would make your eyes like water bottles but I flip them upside down and I'd start assigning fruits and objects to the different features of your face. And I showed you I say lex that's you say well that's not me. And I'd say no but that's my abstraction of you. But that's what the brain does. The space time relationship of the neurons that fire that encode your face have no resemblance to your face. Right? And I think people don't really I don't know if people have fully internalized that but the day that I and I'm not sure I fully internalize that because it's weird to think about. But all neurons can do is fire in space and in time. Different neurons in different sequences perhaps with different intensities. It's not clear the action potential is all or none. Although people, neuroscientists don't like to talk about that, even though it's been published in nature a couple times, the action potential for a given neuron doesn't always have the exact same waveform people. It's in all the textbooks, but you can modify that way for </s><s>[INST] Well, I mean, there's a lot of fascinating stuff with with neuroscience about the fuzziness of all the uh of the transfer of information from neuron to neuron. I mean, we certainly touch upon it every time. We had all tried to think about the difference between artificial neural networks and biological neural networks. But can we maybe linger a little bit on this uh on the circuitry that you're getting at? So the brain is just a bunch of stuff firing and it forms abstractions that are fascinating and beautiful, like layers upon, layers upon layers of abstraction. And I think it just like when you're programming, I'm programming in python, it's uh it's all inspiring to think that underneath it all, it ends up being zeros and ones and the computer doesn't know about no stupid python or windows or Linux exit? It only knows about the zeros and ones in the same way with the brain. Is there something interesting to you are fundamental to you about the circuitry of the brain that allows for the magic that's in our mind to emerge? How much do we understand? I mean, maybe even focusing on the vision system. Is there something specific about the structure of the vision system, the circuitry of it that uh allows for the complexity of the vision system to emerge? Or is it all just a complete chaotic mess that we don't understand? [/INST] It's definitely not all a chaotic mess that we don't understand if we're talking about vision and that's not because I'm a vision scientist, ballistic division division. Well, because in the beauty of the visual system, the reason David Huebel and torn some weasel on the Nobel prize. It was because they were brilliant and forward thinking and adventurous and all that good stuff. But the reason that the visual system is such a great model for addressing these kinds of questions and other systems are hard is we can control the stimulate, we can adjust spatial frequency. How fine the grading czar, thick grading thin greetings, we can just temporal frequency, how fast things are moving. We can use cohen isolating stimuli we can use. There's so many things that you can do in a controlled way. Whereas if we're talking about cognitive encoding, like they, you know, encoding the space of concepts or something, you know, I you know, I like you if I may our am drawn to the big questions in neuroscience but I confess in part because of some good advice I got early in my career and in part because I'm not perhaps smart enough to go after the really high level stuff. I also like to address things that are tractable and I want, you know, we need to we need to address what we can stand to make some ground on at a given </s><s>[INST] time. You can construct brilliant controlled experiments just to study to really literally answer questions about. Yeah, [/INST] Yeah. I mean I'm happy to have a talk about consciousness, but it's a it's a scary talk and I think most people don't want to hear what I have to say, which you know, which is uh we can save that for later. Perhaps </s><s>[INST] it's an interesting question of uh we talk about psychedelics, we can talk about consciousness, we can talk about cognition. Can experiments in neuroscience be constructed [/INST] to shed any </s><s>[INST] kind of light on these questions. So I mean it's cool that vision, I mean to me vision is probably one of the most beautiful things about human beings. Also from the ai side computer vision has the is some of the most exciting applications of neural networks is in computer vision, but it feels like that's a that's a neighbor of cognition and consciousness is just that we maybe haven't come up with experiments to study those yet. [/INST] The visual system is amazing. Were mostly visual animals to navigate survive humans mainly rely on vision, not smell or something else. But um it's a filter for cognition and it's a it's a strong driver of cognition, maybe just because it came up and then we're moving to higher level concepts just the way the visual system works can be summarized in it. Um in a few relatively succinct statement, unlike most of what I said, which has not been. So I think that all the retina </s><s>[INST] involved, [/INST] so the retina is this three layers of neuron structure at the back of your eyes. But I think as a credit card, it is a piece of your brain and sometimes people think I'm kind of wriggling out of a reality by saying that it's absolutely a piece of the brain. It's it's a four brain structure that in the first trimester there's a genetic program that made sure that that neural retina, which is part of your central nervous system, was squeezed out into what's called the embryonic eye cups and that the bone formed with a little hole where the optic nerve is going to connect it to the rest of the brain. And those that window into the world is the only window into the world for a mammal which has a thick skull. Birds have a thin skull so their pineal gland sits and lizards to and snakes actually have a whole, so that light can make it down into the pineal directly and entrained melatonin rhythms for time of day and time of year. Humans have to do all that through the eyes. So three layers of neurons that are a piece of your brain, their central nervous system and the optic nerve connects to the rest of the brain, the neurons in the eye, Someone just care about luminescence, just how brighter dim it is and they inform the brain about time of day and then the central circadian clock informs every cell in your body about time of day. And make sure that all sorts of good stuff happens if you're getting light in your eyes at the right times and all sorts of bad things happen. If you are getting light randomly throughout the 24 hour cycle, we could talk about all that, but this is a good incentive for keeping a relatively normal schedule, consistent schedule. Light exposure, consistent schedule. Try and keep a consistent schedule when you're young. It's easy to go off schedule and recover as you get older, it gets harder. But you see everything from outcomes in cancer patients to um Diabetes um you know, improves when people are getting light at a particular time of day and getting darkness at a particular phase of the 24 hour cycle. We were just designed to um get light and dark at different times of the of the Circadian cycle. That's all being all that information is coming in through a specialist type of neuron and the retina called the melon. Ops and intrinsically photosensitive ganglion cell discovered by David Berson at Brown University. That's not spatial information. It's subconscious you don't think, oh it's daytime, even if you're looking at the sun, it doesn't matter. It's a photon counter, it's literally counting photons and it's saying, oh, even though it's a cloudy day, lots of photons coming in at this winter in boston, it must be winter and your system is a little depressed, it's spring. You feel alert. That's not a coincidence that these melon ops and cells signaling the circadian clock. They're a bunch of other neurons in the eye that signal to the brain. And they mainly signal the presence of things that are lighter than background or darker than background. So a black object would darken the background, a light object lighter than background. And that all comes. It's mainly it's looking at pixels mainly they look at circles and those neurons have receptive fields which not everyone will understand. But those neurons respond best to little circles of dark light or little circles of bright light, little circles of red light versus little circles of green light or blue light. And so it sounds very basic. It's like red, green blue and circles brighter or dimmer than what's next to it. But that's basically the only information that sent down the optic nerve. And when we say information, we can be very precise. I don't mean little bits of red traveling down the optic nerve. I mean spikes, neural action potentials in space and time, which for you is like makes total sense. But I think for a lot of people, it's actually beautiful to think about all that information in the outside world is converted into a language that's very simple. It's just like a few syllables if you will and those syllables are being shouted down. The optic nerve converted into a totally different language like morse code B, b b b B goes into the brain. And then the thalamus essentially responds in the same way that the retina does accept. The thalamus is also waiting things that's saying, you know what that thing was moving faster than everything else or it's brighter than everything else. So that signal, I'm going to get up and allow up to cortex or that signal is much redder than it is green. So I'm gonna let that signal go through. That signal is much it's kind of more like the red next to it, throw that out. The information just doesn't get up into your cortex. And then in cortex of course is where perceptions happen. And in V. One if you will, visual area one, but also some neighboring areas, you start getting representations of things like oriented lines. So there is a neuron that responds to this angle of my hand versus vertical right there is the defining work of human measles nobel and it's a very systematic map of orientation, line, orientation, direction of movement and so forth. And that's pretty much any color and that's how the visual system is organized all the way up to the cortex. So it's hierarchical. You don't build. I want to be clear. It's hierarchical because you don't build up that line by suddenly having a neuron that responds to lines in some random way. It responds to lines by taking all the dots that are aligned in a vertical stack and they all converge on one neuron and then that neuron responds to vertical lines. So it's not random. There's no abstraction at that point. In fact, in fact, if I showed you a black line, I could be sure that if I were imaging V1 that I would see a representation of that black line as a vertical line somewhere in your cortex. So at that point uh it's absolutely concrete. It's not abstract. But then things get really mysterious. Some of that information travels further up into the cortex. So that goes from one visual area to the next to the next to the next. So that by time you get into an area that nancy can wisher at MIT has studied her much of her career, the fuse to form face area, you start finding single neurons that respond only to your father's face or to joe Rogan's face regardless of the orientation of his face. I'm sure if you saw joe because you know him well from across the room and you just saw his profile. Oh that's joe walk over and say hello, the orientation of his faces and there. You won't even see his eyes necessarily, but he's represented in some abstract way by a neuron that actually would be called the joe Rogan neuron. The neurons </s><s>[INST] you might have limits like I might not recognize him if he was upside down or something like that would be fascinating to to see what the limits of that joe Rogan concept is. [/INST] So nancy's lab has done that because early on she was challenged by people that said there aren't face neurons, there are neurons that they only respond to space and time, shapes and things like that, moving in particular directions and orientations and turns out nancy was right um they used the stimulate called gribble stimulate which any computer programmer would appreciate, which kind of more suffices to something gradually that eventually just looks like this like alien thing they call the gribble and the neurons don't respond to Gruebel's in most cases they only respond to faces and familiar faces. Anyway, I'm summarizing a lot of literature and forgive me nancy and for those of the gribble people, if there are something like you don't come after me with pitchforks, actually, you know what come out pitch for, you know what I'm trying to do. So the point is that in the visual system it's very concrete up until about visual area for which has color pinwheels and seems to respond to pinwheels of colors and um and so the stimulus become more and more elaborate, but at some point you depart that concrete representation and you start getting abstract representations that can't be explained by simple point to point wiring and to take a leap out of the visual system to the higher level concepts. When we talked about in the visual system, maps to the auditory system where you're encoding what frequency of tone sweeps. So this is gonna sound weird to do, but you know, like a Doppler, like hearing something car passing by for instance. But at some point you get into motifs of music that can't be mapped to just a what they call a tona topic map of frequency. You start abstracting and if you start thinking about concepts of creativity and love and memory, like what is the map of memory space? Well, your memories are very different than mine, but presumably there's enough structure at the early stages of memory processing or at the early stages of emotional processing or at the earlier stages of creative processing that you have the building blocks your zeros and ones if you will, but you depart from that eventually. Now, the exception to this, and I want to be really clear because I was just mainly talking about neocortex, the six layered structure on the outside of the brain. That explains a lot of human abilities, other animals have them too is that sub cortical structures are a lot more like machines. It's more plug and chug and what I'm talking about is the machinery that controls heart rate and breathing and receptive fields. Um You know, neurons that respond to things like temperature on the top of my left hand. And one of the, you know, I came into the neuroscience from more of a perspective initially of psychology, but one of the reasons I forced upon myself to learn some electrophysiology, not a ton, but enough and some molecular biology and about circuitry is that one of the most beautiful experiences you can have in life, I'm convinced, is to lower an electrode into the cortex and to show a person or an animal. You do this ethically of course, uh stimulus like an oriented line or a face. And you can convert the recordings coming off with the electrode into an audio signal and audio monitor. And you can hear what they call hash. It's not the hashish smoke, it's the hash you here and it's, it sounds like it just sounds like noise. And in the cortex eventually you find a stimulus that gets the neuron to spike and fire action potentials are converted into an auditory stimulus that are very concrete crack, crack, crack sounds like a bat cracking, you know, like home runs, you know, or or outfield balls when you drop electrodes deeper into the thalamus or into the hypothalamus or into the brain stem areas that control breathing. It's like a machine. You never hear hash, you drop the electrode down, This could be like a like a grungy old tungsten electrode not high fidelity electrode. As long as it's got a little bit of insulation on it, plug it into an audio monitor, it's picking up electricity. And if it's a visual neuron and it's in the thalamus of the retina and you walk in front of that animal or person that that neuron goes and then you walk away and it stops and you put your hand in front of the eye again and it goes and you could do that for two days and that neuron will just, every time there's a stimulus it fires. So whereas before it's a question of how much information is getting up to cortex and then these abstractions happening, where you're creating these ideas when you go sub cortical, everything is, </s><s>[INST] there's no abstraction, it's [/INST] two plus two equals four. There's no abstractions. And this is why I um you know, I know we have some common friends at neural link and I love the demonstration they did recently. I'm a huge fan of what they're doing and where they're headed. And no, I don't get paid to say that and I have no business relationship to them. I'm just a huge fan of the people in the mission. But my question was to some of them, you know when you're going to go sub cortical because if you want to control an animal. You don't do it in the cortex. The cortex is like the abstract painting. I made of your face moving, removing one piece or changing something may or may not matter for the abstraction but when you're in the sub cortical areas of the brain, a stimulating electrode can evoke an entire behavior or an entire state. And so the brain, if we can have a discussion about the brain and how the brain works, we need to really be clear which brain because everyone loves neocortex, it's like oh canonical circuits and cortex we're gonna get the cortical connect um and sure necessary but not sufficient not to be able to plug in patterns of electrical stimulation and get behavior eventually we'll get there. But if you're talking sub cortical circuits and that's where the action is. That's where you could potentially cure Parkinson's by stimulating the sub thalamic nucleus because we know that it gates motor activation patterns in very predictable ways. So I think for those that are interested in neuroscience it pays to pay attention to like is this a circuit that abstracts the sensory information or is it just one that builds up hierarchical models in a very predictable way and there's a huge chasm in neuroscience right now because there's no conceptual leadership, no one knows which way to go. And this is why I think neuralink has captured an amazing opportunity which was okay, well while all you academic research labs are figuring all this stuff out. We're going to pick a very specific goal and make the goal the endpoint and some academic laboratories do that. But I think that's a beautiful way to attack this whole thing about the brain because it's very concrete, Let's restore motion to the park and Sony inpatient academic labs do that want to do that too? Of course, let's restore um speech to the stroke patient, but there's nothing abstract about that, that's about figuring out the solution to a particular problem. So, anyway, those are my and I and I admit I've mixed in a lot of opinion there, But having spent some time like 25 years digging around in the brain and listening to neurons firing and looking at them anatomically, I think given its 2020 we need to ask the right, you know the way to get better answers, ask better questions and the really high level stuff is fun. It makes for good conversation and it has um brought enormous interest, but I think the questions about consciousness and dreaming and stuff, they're fascinating. But I don't know that we're there yet. So, </s><s>[INST] You're saying there might be a chasm in the two views mm of uh the power of the brain arising from the from the circuitry that forms abstractions or the power of the brain arising from the majority of the circuitry. That's just doing very uh brute force dumb things that are like, they don't have any fancy kind of stuff going on. That's really interesting to think about [/INST] and which one to go after first and and and here I'm I'm poaching badly from someone I've never met but whose you know, work I I follow, which is and it was actually on your podcast, I think Elon musk said you know, basically the brain is a always a monkey brain with a supercomputer on top and I thought that's actually probably the best description of the brain I've ever heard because it captures a lot of important features like limbic friction, right? People think of like, oh, you know when we're making plans, were using the prefrontal cortex and we're executive function and all this kind of stuff, but think about the drug addict who's driven to go pursue heroin or cocaine, they make plans so clearly they use their frontal cortex. It's just that it's been hijacked by the limbic system and all the monkey, the monkey brain as you referred to. It's really not fair to monkeys though, Ellen because actually monkeys can make plans. They just don't make plans as sophisticated as us. I spent a lot of time with monkeys, but I've also been a long time with you anyway, I'm </s><s>[INST] you're putting you're saying like there's a lot of value to focusing on the monkey brain or whatever the heck you call it like [/INST] I do because let's say I had an ability to place a chip anywhere I wanted in the brain today and activate it or inhibit that area. I'm not sure I would put that chip in neocortex except maybe to just kind of have some fun and see what happens. The reason is it's an abstraction machine and especially if I wanted to make a mass production tool tool and mass production that I could give to a lot of people because it's quite possible that your abstractions are different enough than mine that I wouldn't know what patterns of firing to induce. But if I want, let's say I want to increase my level of focus and creativity. Well then I would love to be able to for instance, control my level of limbic friction. I would love to be able to wake up and go, you know, I have an eight o'clock appointment. I wake up slowly so between 78, but I want to do a lot of linear thinking. You know what? I'm gonna just I'm going to turn down the limbic friction and or ramp up prefrontal cortex activation. So there's a lot of stuff that can happen in the thalamus with sensory gating. Um for instance, you could shut down that shell around the thalamus and allow more creative thinking by allowing more lateral connections. These would be some of those would be the experiments I'd want to do. So they're in the sub cortical quote unquote monkey brain. But you could then look at what sorts of abstract thoughts and behaviors would arise from that rather than. And here, I'm not pointing a finger at neuralink at all. But there's this obsession with neocortex, but I'm gonna, well I might lose a few friends, but I'll hopefully gain a few. And and also, um, one of the reasons people spend so much time in neocortex, I have a fact in an opinion. One fact is that you can image there and you can record their right now. The two photon and one photon microscopy methods that allow you to image deep into the brain still don't allow you to image down really deep unless you're jamming prisms in there and endoscopes and then in the endoscopes are very narrow and you're getting very, you know, it's like looking at the bottom of the ocean through a, through a spotlight and so much easier to look at </s><s>[INST] the waves up on top. [/INST] Right, so let's face it folks, a lot of the reasons why there's so many recordings and layer 23 of cortex with all this advanced microscopy is because it's very hard to image deeper. Now, the microscopes are getting better. And thanks to amazing work mainly of engineers and chemists and physicists, let's face it, they're the ones who brought this revolution in neuroscience in the last 10 years or so. You can image deeper, but we don't really, that's why you see so many reports on layer 23, the other thing, which is purely opinion and I'm not going after anybody here, but is that as long as there's no clear right answer, it becomes a little easier to do creative work in a structure where No one really knows how it works. So it's fun to probe around because anything you see as novel if you're going to work in the thalamus or the pulver in our or the hypothalamus, so these structures have been known about since the 60s and 70s and really since the centuries ago you are dealing with existing, you have to combat existing models and whereas in cortex no one knows how the thing works, neocortex six later cortex. And so everything is there room for discovery. There's a lot more room for discovery and I'm not calling anyone out. I love cortex, we've published some papers on cortex, it's super interesting, but I think with the tools that are available nowadays and where people are trying to head of not just reading from the brain monitoring activity, but right into the brain. I think we really have to be careful and we need to be thoughtful about what are we trying to write? What script are we trying to write? Because there are many brain structures for which we already know what scripts they write. And I think there's tremendous value there. I don't think it's boring. The fact that they act like machines makes them predictable. Those are your zeros and ones, let's start there. But what there, what sort of happening in this field of writing to the brain is there's this idea and again, I want to be clear. I'm not pointing at neuralink I'm mainly pointing at the neo cortical jockeys out there that you go in, you observe patterns and then you think replaying those patterns is going to give rise to something interesting. I I should call out one experiment or two experiments which were done by satsuma Tonegawa Nobel prize winner from M. I. T. Done important work in memory and immunology of course is where you guys Nobel as well as Mark Medford's lab at U. C. San Diego. They did an experiment where they monitored a bunch of neurons while an animal learned something. Then they captured those neurons through some molecular tricks so they could replay the neurons. So now there's like perfect case scenario, it's like okay, you monitor the neurons in your brain then I say okay neurons one through 100 were played in the particular sequence. So you know the space time, you know the keys on the piano that were played that gave rise to the song which was the behavior and then you go back and you reactivate those neurons except you reactivate them all at once, like slamming on all the keys once on the piano and you get the exact same behavior. So the space time code may be meaningless for some structures. Now that's freaky, that's a scary thing because what that means is that all the space time firing in cortex the space part may matter more than the time part. So rate codes and space time codes we don't know and you know I'd rather have I'd rather deliver more answers in this discussion questions but I think it's an important consideration. </s><s>[INST] You're saying some of the magic is in the early stages of what the closer to the raw information, high brain interesting I believe so um [/INST] you know the stimulus you know there are only encodes that stimulus. So you know the transformation when I say this for those who don't think about sensory transformations it's like I can show you a red um circle and then I look at how many times the neuron fires in response to that red circle and then I can show the red circle a bunch of times green circles, see if it changes and then essentially the number of times that is the transformation you've converted red circle into like three action potentials, you know B. P. P. Or whatever you want to call it, you know for those that think in sound space. So that's what you've created. You know the transformation and you march up the it's called the neuro axis as you go from the periphery up into the cortex and we know that and I know lisa um Feldman Barrett or visit Barrett, Feldman Barrett film. Excuse me. Um lisa um talked a lot about this that you know birds can do sophisticated things and whatnot as well but humans there's a strong, what we call civilization, A lot of the processing has moved up into the cortex and out of these sub cortical areas but it happens nonetheless. And so as long as you know, the transformations you are in a perfect place to build machines or add machines to the brain that exactly mimic what the brain wants to do, which is take events in the environment and turn them into internal firing of neurons. So the </s><s>[INST] mastery of the brain can happen at their early, you know, another perspective of it is uh you're saying this means that humans aren't that special. Uh if you look at the evolutionary time scale the leap to intelligence is not that special. So like the extra layers of abstraction isn't where most of the magic happens of intelligence, which gives me hope that maybe if that's true, that means the evolution of intelligence is not that rare of an event. [/INST] I certainly hope not. Um, </s><s>[INST] you you hope there's [/INST] I hope there are other forms of intelligence. I mean I think what humans are really good at. Um and here I want to be clear that this is not a formal model, but what humans are really good at is taking that um plasma Barbell that we were talking about earlier and not just using it for analysis of space, like the your immediate environment, but also using historical information. Like I can read a book today about the history of medicine. I happen to be doing that lately for some stuff I'm researching and I can take that information and if I want I can inject it into my plans for the future. Other animals don't seem to do that over the same time scales that we do now. It may be that the chipmunks are all hiding little like notebooks everywhere in the form of like little dirt castles are something that we don't understand. I mean the waggle dance of the bee is in the most famous example, bees come back to the hive, they orient relative to the honey comb and they waggle. There's a guy down in Australia named Srinivasan who studied this. It's really interesting. No one really understands it. So he understands it best. The bee waggles at a couple of ways relative to the orientation of the honeycomb and then all the other bees see that it's visual and they go out and they know the exact coordinate system to get to the, to the source of whatever it was the food and bring it back. And he's done it where they isolate the bees, He's changed the visual flight environment all this stuff they are communicating and they're communicating something about something they saw recently, but it doesn't extend over very long periods of time. The same way that you and I can both read a book or you can recommend something to me and then we could converge on a set of ideas later. And uh in fairness because she was the one that said it and I didn't and I hadn't even thought of it. Um when you talk to lisa on your podcast, she brought up something beautiful which is that I had never really occurred to me and I was sort of embarrassed that it hadn't, but it's really beautiful and brilliant. Which is that you know, we don't just encode senses in the form of color and light and sound waves and tastes. But ideas become a form of sensory mapping. And that's where the you know, the really, really cool and exciting stuff is. But we just don't understand what the receptive fields are for ideas. What's an idea of receptive field </s><s>[INST] and how they are communicated between humans because we seem to be uh to be able to encode those ideas in some kind of way. Yes, it's taking all the raw information and the internal physical states. The that sensor information put into this concept blob that would come in the store and they were able to communicate that [/INST] your abstractions are different than mine. I actually think the comment section, you know on on social media is a beautiful example of where the abstractions are different for different people. So much of the misunderstanding of the world is because of these these idea receptive fields. They're not the same. Whereas I can look at a photo receptor neurons are olfactory neuron or a V. One neuron and I am certain I would bet my life that yours look and respond exactly the same way that lisa's due in mind you but once you get beyond there it gets tricky. And so when you say something where I say something and somebody gets upset about it or even happy about it, their concept of that might be quite a bit different. They don't really know what you mean, they only know what it means to them. </s><s>[INST] Yeah. So from a neural link perspective it makes sense to optimize the control and the augmentation of the of the more primitive uh circuitry. So like the stuff that is closer to the raw sensor [/INST] information go deeper </s><s>[INST] if they think by the way so [/INST] deeper into the brain and to and to be fair so matt McDougall who is the neurosurgeon neuralink and clinical nursing. A great guy, brilliant. They have amazing people. I have to give it to them. They have been very cryptic in recent years. Their website was just like a like no like nothing there. You know, they know they really know how to do things with style and and they've upset a lot of people but that's good too. Um matt is there. I know matt he actually came up through my lab at stanford although he you know he was a neurosurgery resident. He spent time in our lab. He actually came out on the shark dive and did great white shark diving with my lab to collect the VR. That we use in our fear stuff. I've talked to matt and I think you know, he and other folks that are hungry for the deeper brain structures. The problem is that damn vasculature, all that blood supply. It's not trivial to get through and down into the brain without damaging the vasculature in the neocortex, which is on the outer crust. But once you start getting into the thalamus and closer to some of the main arterial sources, you really risk getting massive bleeds. And so it's it's a it's an issue that can be worked out. It just is hard. </s><s>[INST] This is maybe be nice to educate. I'm showing my ignorance. So the smart stuff is on the surface. So I didn't realize that. I didn't quite realize because you keep saying deep. [/INST] Yeah. So </s><s>[INST] so like the early stages are deep. Yeah. So in actually physically in the brain. [/INST] Yeah. So the the way that um you know, of course, you've got your your deep brain structures involved in breathing and heart rate and kind of lizard brain stuff. And then on top of that this is the the the model of the brain that no one really subscribes to anymore but anatomically it works. And then on top in mammals. And then on top of that you have the limbic structures which gate sensory information and decide whether you're going to listen to something more than you're gonna look at it or you're gonna split your attention to both kind of sense reallocation stuff and then the neocortex is on the outside. Um and that is where you get a lot of this abstraction stuff. And now not all cortical areas are doing abstraction, some like visual area one auditory area one, they're just doing concrete representations. But um as you get into the higher </s><s>[INST] order stuff [/INST] that when you start hearing names like infra parietal cortex and you know when you start hearing multiple names in the same, then you're talking about higher order areas. But actually there's an important experiment that um that drives a lot of what people want to do with brain machine interface. And that's the work of Bill Newsom who is at stanford and Tony Matson who runs the center for neuroscience at N. Y. U. This is a wild experiment and I think it might freak a few people out if they really think about it too deeply. But um anyway, here he goes, there's an area called Mt in the cortex. And if I showed you a bunch of dots all moving up and this is what they this is what Tony and Bill and some of the other people um in that lab did way back when as they show a bunch of dots moving up. Somewhere in Mt there's some neurons that respond, they fire when the neurons move up and then what they do is they start varying the coherence of that motion. So they made it. So only 50% of the dots moved up and the rest move random and that neuron fires a little less and eventually it's random in that neuron stops firing because it's just kind of dots moving everywhere. It's awesome. And there's a systematic map so that other neurons are responding to things moving down and other things are responding left and other things. Right, Okay. So there's a map of direction space. Okay, well that's great. You could lesion Mt animals lose the ability to do these kind of coherent discrimination or direction discrimination. But the amazing experiment, the one that just it's kind of eerie is that they lowered a stimulating electrode into empty, found a neuron that responds to when dots go up but then they silence that neuron. And sure enough the animal doesn't recognize. The neurons are going up and then they move the dots down. They stimulate the neuron that responds to things moving up and the animal responds because they can't speak. It responds by doing a lever press which says the dots are moving up. So in other words the sensory, the dots are moving down. In reality on the computer screen, they're stimulating the neuron that responds to dots moving up and the perception of the animal is that dots are moving up. Which tells you that your perception of external reality absolutely has to be a neuronal abstraction. It is not tacked to the movement of the dots in any absolute way. Your perception of the outside world depends entirely on the activation patterns of neurons in the brain. And you can hear that and say, well duh, because if I stimulate, you know, the stretch reflex and you kick or something or whatever, you know, the knee reflex and you kick, of course there's a neuron that triggers that, but it didn't have to be that way because a the animal had prior experience be your way up in this, you know, higher order cortical areas. What this means is that and I generally try to avoid conversations about this kind of thing, but what this means is that we are constructing our reality with the space time, firing the zeros and ones and it doesn't have to have anything to do with the actual reality. And the animal or person can be absolutely convinced that that's what's happening. </s><s>[INST] Are you familiar with the work of Donald Hoffman? So he's uh uh so he makes an evolutionary argument that's not important of that? We our brains are completely detached from reality in the sense that he makes a radical case that we have no idea what physical reality is and in fact is drastically different than what we think it is. So he goes, that's scary. So he doesn't say like there's just because you're kind of implying there's a there's a gap, there might be a gap of constructing an illusion and then maybe using uh communication to maybe uh create a consistency that's sufficient for human collaboration, whatever or mammal, you know, just maybe even just life forms are constructing a consistent reality that may be detached. I mean, that's really cool that neurons are constructing that like that you can prove this is when your science at its best vision science. But he says that like our brain is actually just lost its ship and on the on the on the path of evolution to where we're normal. We're just playing games with each other in constructing realities that allow our survival, but it's completely detached from physical reality. We're missing a lot. We're missing. Like most of it, if not [/INST] all of it. Well, this was um it's it's fascinating because I just saw the Oliver sacks documentary, there's a new documentary out about his life and there's this one part where he's like, I've spent part of my life trying to imagine what it would like to be to be a bat or something, to see the world through the like, you know, the sensory apparatus of a bat and he did this with his these patients that were locked into these horrible syndromes, that to pull out some of the beauty of their experience as well, not just communicate the suffering, although the suffering too. And as I was listening to him talk about this, I start to realize it's like well, what you know like they're these mantis shrimps that can see 60 shades of pink or something and they see this stuff all the time and animals can see UV light every time I learn about an animal that can sense other things in the environment that I can't like heat sensing, why not? I don't crave that experience the same way Sachs talked about craving that experience, but it does throw another penny in the jar for what you're saying, which is that it could be that most if not all of what I perceive and believe is just a neural fabrication and that for better for worse. We all agree on enough of the same neural fabrications in the same time in place that we're able to function. </s><s>[INST] Not only that, but we agree with the things that are trying to eat us uh enough to where we don't they don't eat us meeting like that, it's not just us humans, you know, because it's interactive, it's interactive. So like so like uh now I think it's a really uh nice thought experiment, I think because uh Donald really frames in a scientific, like he makes a hard like as hard as our discussion has been now, he makes a hard scientific case that we don't know shit about reality. Uh I think that's a little bit uh hardcore, but I think it's I think [/INST] it's our </s><s>[INST] I think it's a good thought experiment that kind of cleanses the palate of the confidence we might have about uh about because we are operating in this abstraction space, you know, and uh you know the sensory space, it might be something very different and and it's kind of interesting to think about if you start to go into around with neuralink or start to talk about just everything that you've been talking about with dream states and psychedelics and stuff like that, which part of the which layer can we control and play around with and maybe look into a different slice of reality. You know, you got to do the experiment. The key [/INST] is to just do the experiment in the most ethical way possible. You just I mean that's the beauty of experiments, this is why um you know, there's there's there's wonderful theoretical neuroscience happening now to make predictions and but but that's why experimental science is so wonderful. You can go into the laboratory and poke around in there and be a brain explorer and and listen to and right to neurons and when you do that, you get answers, you don't always get the answers you want, but that's the you know, that's the beauty of it. I think when you were saying um this thing about reality and the Donald Hoffman model, I was thinking about Children, you know um like when I have an older sister, she's very sane um uh but when she was a kid, she had an imaginary friend </s><s>[INST] and she would play with this [/INST] imaginary friend and it had there was this whole there was a consistency. This friend was like it was Larry lived in a purple house. Larry was a girl. It was like all this stuff that a child a young child wouldn't have any issue with. And then one day she announced that Larry had died, right and it wasn't dramatic or traumatic and that was it and she just stopped and I always wonder what that um neuro developmental event was that um a kept her out of a a psychiatric ward had she got you know, kept that imaginary friend. But but it's also there was something kind of sad to it, I think the way it was told to me because I'm the younger brother, I didn't, I wasn't around for that, but my my dad told me that you know there was a kind of a sadness because it was this beautiful reality that had been constructed. So we kind of won I wonder as you're telling me this whether or not, you know, as adults, we try and create as much reality for Children as we can so that they can make predictions and feel safe because the ability to make predictions as a lot of what keeps our autonomic arousal in check. I mean we go to sleep every night and we give up total control and that should frighten us deeply. But you know, unfortunately autonomic arousal, the yanks us down under and we don't negotiate too much, so you sleep sooner or later. Um I don't know. Um I was a little worried. We get into discussions about the nature of reality because it's interesting in the laboratory. I'm a very much like what's the experiment? What would you know, what's the analysis going to look like? What mutant mouse or we're going to use what what what experience we're gonna put someone through. But I think it's wonderful that in 2020 we can finally have discussions about this stuff and look kind of peek around the corner and say, well no link and people, others who are doing similar things are going to figure it out. They're going to the the answers will show up and we just have to be open to interpretation. </s><s>[INST] Do you think there could be an experiment uh centered around consciousness? I mean, you're plugged into the neuroscience community. I think for the longest time, the quote unquote Seaward was totally not uh was almost anti scientific. But now more and more people are talking about consciousness. Ellen is talking about consciousness ai folks is talking about consciousness. It's it's still nobody knows anything, but it feels like a legitimate domain of inquiry that's hungry for a real experiment. [/INST] So I fortunately three short answers to this. Um The first one is a years later, I'm not, I'm not particularly something. I I agree that the joke I always tell us, um, two things you never want to say to a scientist, one is what do you do? And the second one is um, take as much time as you need and you definitely want to say them in the same sentence. Um, you have three short answers to it. So there's a, there's a cynical answer kind of, uh, it's not what I enjoy giving, which is that um, if you look Into the 70s and a back at the 1970s and 1980s And even into the early 2000s, there were some very dynamic, very impressive speakers who are very smart in the field of neuroscience and related fields who thought hard about the consciousness problem and fell in love with the problem but overlooked the fact that the technology wasn't there. So, um, I admire them for falling in love with the problem. But they gleaned tremendous taxpayer resources essentially for nothing. And these people know who they are, some of them are alive. Someone remarked, I'm not referring to Francis crick, who was brilliant by the way. And I thought the classroom was involved in consciousness, which I think is a great idea. It's this obscure structure that no one's really studied. People are now trying to study it. So I think Francis was brilliant and wonderful, but there, you know, there were books written about it. It makes for great television stuff and I thought around the table or after a couple glasses of wine or whatever. It's an important problem nonetheless. And so I think I do think the consciousness, the issue is not operationally defined, right, that psychologists are much smarter than um a lot of hard scientists in that for the following reason, they put operational definitions, they know that psychology, if we're talking about motivation, for instance, they know they need to put operational definitions on that so that two laboratories can know they're studying the same thing. The problem with consciousness is no one can agree on what that is, and this was a problem for attention When I was coming up. So in the early 2000's, people would argue what is the tension is a spatial attention, auditory attention, is it? And finally people are like, you know what we agree </s><s>[INST] With, they agreed on that one. [/INST] So remember sort I remember hearing people scream a lot of tension, right? They couldn't even agree on attention. So I was coming up as a young graduate student, I'm thinking like I'm definitely not gonna work on attention and I'm definitely not gonna work on consciousness. And I wanted something that I could solve or figure out, I want to be able to see the circuit or the neurons, I want to be able to hear it on the audio, I want to record from it, and then I want to do gain a function and loss of function, take it away, see something change, Put it back. See something change in a systematic way. And that takes you down into the depths of some stuff that's pretty um plug and chug, you know, but you know, I'll borrow from something in the military because I'm fortunate to do some work with units from special operations and they have beautiful language around things because their world is not abstract and they talk about three m targets, 10 m targets, 100 m targets. And it's not an issue of picking the 100 m target because it's more beautiful or because it's more interesting if you don't take down the three m targets and the 10 m targets first, you're dead. So that's a, I think scientists could pay to, you know, adopt a more kind of military thinking in that in that sense. The other thing that is really important is that just because somebody conceived of something and can talk about it beautifully and can glean a lot of um resources for it doesn't mean that it's lead anywhere. So this isn't just true of the consciousness issue. And I don't want to sound cynical, but I could pull up some names of molecules that occupied hundreds of articles in the very premier journals that then were later discovered to be totally moot for that process. And biotech companies folded everyone. And the lab pivots and starts doing something different with that molecule and nobody talks about it because as long as you're in the game, we have this thing called anonymous peer review, you can't afford to piss off anybody too much unless you have some other funding stream. And I have avoided battles most of my career, but I pay attention to all of it and I've watched this and I don't think it's ego driven. I think it's that people fall in love with an idea. I don't think there's any, there's not enough money in science for people to sit back there rubbing their hands together. You know, the beauty of what neuralink and Elon and team because obviously he's very impressive. But the team as a whole is really what gives me great confidence in their mission is that he's already got enough money. So it can't be about that he doesn't seem to need it at a level of, I don't know him, but it doesn't he doesn't seem to need it at a kind of an ego level or something I think is driven by genuine curiosity and the team that he's assembled include people that are very kind of abstract neuro neocortex, space time coding people there. People like matt who is a neurosurgeon you can't, I mean, you know, you can't be as neurosurgery, failures in neurosurgery are not tolerated. So you have to be very good to exceptional to even get through the gate and he's exceptional and then they've got people like dan Adams who was at UCSF for a long time is a good friend and a known him for years, um, who is very concrete, studied the vasculature in the eye and how it maps to the vasculature and cortex. When you get a team like that together, you're going to have dissenters, you're going to have people that are high level thinkers, people that are coders when you get a team like that, it no longer looks like an academic laboratory or even a field in science. And so I think they're going to solve some really hard problems and again, I'm not here. Uh, they don't, you know, I've nothing at stake with them. But I think that's the solution. You need a bunch of people who don't need first author papers who don't need to complete their PhD who aren't relying on outside funding, who have a clear mission and you have a bunch of people who are basically will adapt </s><s>[INST] To solve the problem. I like to acknowledge that the three m target and the 100 m target. So the folks in neuralink are basically many of them are some of the best people in the world at the three m target. Like you mentioned that your surgery, like they're solving real problems. There's no Bs philosophy, philosophical, uh, smoke some weed and look back and look at the stars. But yeah, uh, so both on Ellen and because I think like this, I think it's really important to think about the 100 m and the 100 m is not even, not even 100 m but like, like the stuff behind the hill that's that's that's too too far away. Which is which is where consciousness I'm Uh huh. Maybe I tend to believe that consciousness could be engineered any part of the reason part of the business I want to build leverages that idea that consciousness is a lot simpler that we've been we've been talking about. [/INST] Well if someone can simplify the problem, that will be wonderful. I mean the reason we can talk about something as abstract as face representations, inclusive form, face area is because nancy can we sure had the brilliance to tie it to the um kind of lower level um statistics of visual scenes. It wasn't because she was like, oh I bet it's there. That would have been interesting. So people like her understand how to bridge that gap and they put a tractable definition. So so I just that's what I'm begging for in science. It's a tractable definition. This is what I want people </s><s>[INST] to sit in a I want people who are really uncomfortable with woo like consciousness like high level stuff to sit in that topic and sit uncomfortably because it forces them to then try to ground and simplify it into something that's concrete because too many people are just uncomfortable to sit in the consciousness room because there's no definitions. It's like attention or or intelligence and artificial intelligence community, but the reality is it's easy to avoid that room altogether, which is what I mean. There's analogies to everything you've said with the artificial intelligence community, with Minsky and even alan turing that talked about intelligence a lot and then they drew a lot of funding and then it crashed because they really didn't do anything with it and it was a lot of force of personality and [/INST] so on. But that </s><s>[INST] doesn't mean the topic of the touring test. And intelligence isn't something we should sit on and think like, I think like what is first of all Touring actually attempted this with the turing test. He tried to make concrete this very question of intelligence. It doesn't mean that we shouldn't linger on it and we shouldn't forget that ultimately, that is what our efforts are all about the artificial intelligence community and in the people where there's neuroscience or whatever bigger umbrella you want to use for understanding the mind. The goal is not just about understanding layer two or three of the vision, it's it's to understand consciousness and intelligence and maybe created or just all the possible biggest questions of our universe. That's that's ultimately the dream. [/INST] Absolutely. And I think what I really appreciate about appreciate about what you're saying is that everybody whether or not they're working on a kind of low level synapse that's like a reflex in the musculature or something very high level abstract can benefit from looking at those who prefer three, you know, everyone's going after a three m 10 m 100 m targets in some sense, but to be able to tolerate the discomfort of being in a conversation where there are real answers where the zeros and ones are known zeros and ones those are the equivalent of them in the nervous system. And also as you said, for the people that are very much like, oh I can only trust what I can see and touch those people need to put themselves into the discomfort of the high level conversation because what missing is conversation and conceptualization of things at multiple levels. I think one of the, this is um I don't gripe about my life has been fortunate, we've been funded from the start and we've been happy um in that in that regard and lucky and we're grateful for that. But I think one of the challenges of research being so expensive is that there isn't a lot of time, especially nowadays for people to just convene around a topic Because there's so much emphasis on productivity. Um and so they're actually believe or not, there aren't that many concepts, formal concepts in neuroscience right now, the last 10 years has been this huge influx of tools and some people in neural circuits and probing around and connect homes, it's been wonderful. But You know, 10, 20 years ago when the consciousness stuff was more prominent? The C-word as you said. Um what was good about that time is that people would go to meetings and actually discuss ideas and models now. It's sort of like demonstrate, it's sort of like demonstration day at the school science fair where everyone's got their thing and some stuff is cooler than others. But um I think we're going to see a shift. I'm grateful that we have so many computer scientists and theoreticians and um are theorists, I think they call themselves. Um somebody tell me what the difference is someday. Um and you know, psychology and even, dare I say, philosophy, you know, these things are starting to converge, you know, neuroscience that the name neuroscience, there wasn't even such a thing when I started graduate school as a post doc. It was neurophysiology or you're on your own ademas or what now it's sort of everybody's invited and that's beautiful. That means that something useful is going to come of all this. And there's also a tremendous work of course happening for the treatment of disease and we shouldn't overlook that. That's where, you know, ending, eliminating reducing suffering is also a huge initiative on neuroscience. So there's a lot of beauty in the field, but the consciousness thing continues to be a it's like an exotic bird. It's like no one really quite knows how to handle it and it dies very easily. Well </s><s>[INST] yeah, I think also from the ai perspective I so I view the brain is less sacred I think from a neuroscience perspective, you're a little bit more sensitive to bs like bs narratives about the brain or whatever. I'm a little bit more uh comfortable with just poetic bs about the brain as long as it helps engineer intelligence systems. [/INST] Well, you know what I mean? Well and I have to you know, I confess um ignorance when it comes to, you know, most things about coding and I have some quantitative ability, but I don't have strong quantitative leanings. And so I know my limitations too. And so I think the next generation coming up, you know, a lot of the students at stanford are really interested in quantitative models and theory and ai and I remember when I was coming up Um a lot of people who are doing work ahead of me, I kind of rolled my eyes at some of the stuff they were doing, um including some of their personalities, although I have many great senior colleagues uh every of the world, so it's the way of the world, so nobody knows what it's like to be uh you know, a young graduate student in 2020, except the young graduate students. So I know what I'm. I know there's a lot of things I don't know and um in addition to what we do, a lot of public education, increased scientific literacy and no scientific thinking etcetera. A big goal of mine is to try and at least pave the way so that these really brilliant and forward thinking. Um younger scientists can make the biggest possible dent and make what will eventually be all of us old guys and gals look stupid. I mean that's that's what we were all trying to do. That's what we were trying to do. So yeah, </s><s>[INST] so from the highest possible topic of consciousness to the to the lowest level uh topic of David Goggins, uh I [/INST] don't know if it's low, low level, he's high performance, high performance, </s><s>[INST] but like, like I don't think David has a has any time for philosophy. Let's just put it this way. [/INST] Uh Well it's, I mean I think we can tack it to what we were just saying in a in a meaningful way, which is whatever goes on in that abstraction part of the brain. He's figured, you know, he's figured out how to dig down in whatever the limbic friction, he's figured out how to grab ahold of that scruff it and send it in the direction that he's decided it needs to go. And what's wild is that he's, what we're talking about is him doing that to himself, right? He's, it's like he's scruff in himself and directing himself in a particular direction and sending himself down that trajectory. And he, what's beautiful is that he acknowledges that that process is not pretty, it doesn't feel good, it's kind of horrible at every level, but he's created this rewarding element to it. And I think that's what's so, it's so admirable and it's what so many people crave, which is regulation of the self at that level. </s><s>[INST] And he practices. I mean, there's a ritual to it. There's uh every single day, like no exceptions, there's a practice aspect to the suffering that he goes through principles, principles, suffering. And I mean, I just, I mean, I admire all aspects of it, including him and his girlfriend slash wife. I'm not sure she'll probably know this fiance, [/INST] wonderful person. No, no, we've only community. Uh we've only we've only communicated with her um by a text about some stuff. Um I was asking David, but yeah, they they clearly have formed a powerful team. Um and it's a beautiful thing to to see people working in that kind of synergy, </s><s>[INST] it's inspiring to me. Same as with Dylan, that a guy like David Goggins can find love uh that that you find a thing that works, which gives me hope that, like, whatever, whatever flavor of crazy I am, you can always find another thing that works with that. But I've had uh maybe let's trade Goggin stories, uh you're from a neuroscience perspective me from a uh self inflicted pain perspective. I somehow found myself in communication with David about some challenges that was undergoing. Um one of which is we were communicating every single day, email phone About the particular 30 day challenges I did that stretched for longer of push ups and pull ups. [/INST] You made a call </s><s>[INST] out on social, [/INST] social media actually. I think that was what was the point I knew of you before, but that's where I started tracking some of what you were doing with these physical challenges. And I um </s><s>[INST] the hell is wrong with that? [/INST] Well, no, I think I actually, I don't often comment on people's stuff, but I think I commented something like neural plasticity loves a non negotiable rule. I know is a non negotiable contract because at the point where, yeah, neural neural plasticity really loves a non negotiable contract because, you know, and I've said this before, so forgive me, but you know, the brain is doing analysis of duration path and outcome and that's a lot of work for the brain and the more that it can pass off duration path and outcome to just reflex the more energy and it can allocate to other things. So if you decide, there's no negotiation about how many pushups, how far I'm gonna run, how many days, how many pull ups etcetera. You actually have more energy for push ups running and pull up. And when </s><s>[INST] you say in your classes, you mean like the brain wants, the decision is made, it will start rewiring stuff to to make sure that this, we can actually make this [/INST] happen. That's right. I mean, so much of what we do is reflexive at the level of just course circuitry breathing, heart rate, all that boring stuff, digestion, but then there's a lot of reflexive stuff like how you drink out of a mug of coffee that's reflective too, but that you had to learn at some point in your life earlier when you were very little analyzing duration path and outcome and that involves a lot of top down processing with the prefrontal cortex, but through plasticity mechanisms, you now do it. So when you take on a challenge, provided that you understand the core mechanics of how to, you know, run push ups and pull ups and whatever else you decided to do. Once you set the number and the duration and all that, then you all you have to do is just go, but people get caught in that tide pool of just, well, do I really have to do it? How do I not do that? What if I get injured? What, you know, can I sneak at this so that you know, and that's work and to some extent I look, I I'm not David Goggins obviously, um nor nor do I claim to understand his process uh partially, you know, um but maybe a little bit which is that it's clear that by making the decision, there's more resources to devote to the effort of the actual execution. </s><s>[INST] Well that's a really like what you're saying was not a lesson that was obvious to me is still not obvious, something I really work at, which is there is always an option to quit. And I mean that's something I really struggle with. I mean I've quit some things in my life. Check stupid stuff. Yeah. And one lesson I've learned is if you quit once it opens the door that, but like it's really valuable to trick your brain into thinking that you're, you're going to have to die before you quit. Like it's actually really convenient. So actually what you're saying is very profound, [/INST] but </s><s>[INST] you shouldn't intellectualize it. Like it took me time to develop like psychologically in ways that I think it would be another conversation because I'm not sure how to put it into words, but it's really tough. I'm me to to do certain parts of that [/INST] challenge, which is a huge, is a huge output. The number that </s><s>[INST] I was, I thought it would be the number would be hard, but it's not, it's the entirety of it, especially in the early days was just spending, I'm kind of embarrassed to say how many hours this took. So I didn't say publicly how many hours because people I knew people would be like, don't you aren't you supposed to do other stuff? [/INST] Well, it's how are you doing? And I don't want to speculate too much about. But occasionally David has said this publicly where people will be like, don't you sleep or something and his process used to just be that he would just block the lead, you know, like gone. But it's, it's actually, um, it's a super interesting topic and because self control and directing our actions and the role of emotion and quitting. These are, these are vital to the human experience and they're vital to performing well at anything. And obviously at a super high level being able to understand this about the self is crucial. Um, so I have a friend who was also in the team's, his name is Pat Dossett did nine years in the seal teams. Um, and in a similar way that there's, there's a lower about him among team guys because of a kind of funny challenge he gave himself, which was, so he and I swim together, although he swims further up front than I do. Um, and he's very patient. Um, but you know, he was on a, he was assigned when he was in the teams to a position that gave him a little more time behind a desk than he wanted and it's not as much time out and deployments, although he did deployments. Um, so he didn't know what to do at that time, but he thought about it and he asked himself, what does he hate the most And it turns out the thing that he hated doing the most to his bear crawls walking on your hands and knees. So he decided to bear crawl for a mile for time. So it was a bear crawling a mile a day, right? And I thought that was an interesting example they gave because you know, like why pick the thing you hate the most? And I think it maps right back to olympic friction. It's the thing that creates the most limbic friction. And so if you can overcome that then there's carry over. And I think the notion of carryover has been talked about psychologically and kind of in the self help space like oh if you run a marathon it's gonna help you in other areas of life. But will it really? Well, it well, I think it depends on whether or not there's a lot of limbic friction because if there is what you're exercising is not a circuit for bear crawls or a circuit for pull ups, what you're doing is you're exercising a circuit for top down control. And that circuit was not designed to be for bear crawls or pull ups or coding or waking up in the middle of the night to do something hard. That circuit was designed to override limbic friction. And so neural circuits were designed to generalize, right? The stress response to an incoming threat that's a physical threat was designed to feel the same way and be the same response internally as the threat to an impending exam or divorce or marriage or whatever it is that's stressing somebody out. And so neural circuits are not designed to be for one particular action or purpose. So if you can as you did, if you can train up top down control under conditions of the highest limbic friction that when the desire to quit is at its utmost either because of fatigue or hyper arousal being too stressed or too tired, you're, you're learning how to engage a circuit and that circuit is forever with you. And if you don't engage it, you it sits there, but it's atrophied, it's not, it's like a plant that doesn't get any water. And a lot of this has been discussed in self help and growth mindset and all these kinds of ideas that circle the internet and social media. But when you start to think about how they map the neural circuits, I think there's some utility because what it means is that the limbic friction that you'll experience in, I don't know, maybe some future relationship to something or someone, it will, it's a category of neural processing that should immediately click into place. It's just like the limbic friction you experienced, trying to engage in, God knows how many push ups, pull ups in and running, uh runs, you were doing 20 </s><s>[INST] 5000 twist counting. [/INST] So folks, if if lex does this again, more comments more like, well, there's a problem with you getting more followers </s><s>[INST] as you're gonna get [/INST] more. Actually, I should say that's the benefit, I don't know, maybe it's not politically correct for me to ask, but like, there is this uh stereotype about Russians being, you know, like, like like like like uh like being really um you know durable and you know, I I started going to that Russian banya that way back before Covid, and um they could tolerate a lot of heat, you know, and and they would sit very stoic, you know, no one was going on, it's hot in here, they were just kind of like, ease into it. Um So maybe there's something that </s><s>[INST] might be something there, but it could be also just personal. I just have some, I found myself everyone's different, but I've found myself to be able to do something unpleasant for very long periods of time, like I'm able to shut off the mind and I don't think that's been fully tested. [/INST] And the monkey mind or the supercomputer. Uh </s><s>[INST] Well, it's interesting, I mean, which mind is which mind tells you to quit exactly, [/INST] limbic limbic friction tells </s><s>[INST] you what the limbic friction is, the source of that, but who are you talking with? Exactly. [/INST] So there's a um we can put something very concrete to that. There's a paper publishing sell super top tier journal uh two years ago, um looking at effort and this was in a visual environment of trying to swim forward toward a target and a reward. And it was a really cool experiment cause they manipulated virtually the visual environment. So um the same amount of effort was being expended every time, but sometimes the perception was you're making forward progress and sometimes the perception was making no progress because stuff wasn't drifting by meant no progress. So you can be swimming and swimming and not making progress. And it turns out that with each bout of effort there's a epinephrine and norepinephrine is being released in the brain stem. And glia traditionally were thought of as support cells for the neurons. But they do a lot of things actively to our measuring the amount of epinephrine and norepinephrine in that circuit. And when it exceeds a certain threshold, the glia send inhibitory signals that shut down top down control. They literally, it's the quit. You stop, there's no no more you quit enduring. It can be rescued, endurance can be rescued with dopamine. Uh so that's where the subjective part really comes into play. So you quit because you've learned how to turn that off, or you've learned how to write. Some people will reward the pain process so much that friction becomes the reward. And you know, when you talk about people like Goggins and other people I know from special operations and people have gone through cancer treatments three times you hear about, you know, just when you hear about people, the victor frankel's stories, I mean, you hear about nelson Mandela, you hear about these stories, I'm sure the same processes involved again, this speaks to the generalized ability of these processes as opposed to a neural circuit for a particular action or cognitive function. So I think, um you have to learn to subjectively self reward in a way that replenishes you, uh Goggins talks about eating souls. It's very dramatic example in his mind. Apparently that's a form of reward, but it's not just a form of reward where you're it's like, uh you're picking up a trophy or something. It's it's actually it gives the energy, it's a reward that gives more neural energy. And I'm defining that as more dopamine to suppress the nor adrenaline, adrenaline circuits in the brain stem to </s><s>[INST] ultimately maps of that. Yeah, he creates enemies. He's always fighting enemies. I never I think I have enemies, but they're usually just versions of me inside my head. Uh so I've thought about through that 30 day challenge, I tried to come up with, like fake enemies. It wasn't working the only enemy I came up with, this, David. [/INST] Well, now you you certainly have a form, a formidable adversary in this one. I </s><s>[INST] don't care. I'm David, I'm willing to die on this one. So let's go there, but like, [/INST] well, let's hope you both uh both survive this. Um But </s><s>[INST] my problem is the physical, there's uh everything we've been talking about in the mind, there's a physical aspect that's just practically difficult, which is like, I can't like, you know, when you injure yourself at a certain point, like, you just can't function [/INST] or you're doing more damage. Talking about taking yourself out of running for Yeah, for </s><s>[INST] the rest of your life potentially, or like you know, for years. So you know, I'd love to avoid that, [/INST] right? </s><s>[INST] There's just like stupid physical stuff that you just want to avoid. He want to keep it purely in the mental and if it's purely in the mental, that's when the race is interesting. But yeah, the problem with these physical challenges as as David has experienced, I mean it has a toll on your body. I tend to think of, the mind is limitless and the body is kind of unfortunately quite limited. [/INST] Well, I think the key is to dynamically control your output and that can be done by reducing effort, which doesn't work for throughout, but also by um restoring through these uh subjective reward processes and and we don't want to go down the rabbit hole of why this all works. But these are ancient pathways that were designed to bring resources to an animal or to a person through foraging for hunting or mates or water all these things and they work so well because they're down in those uh circuits where we know the zeros and ones and that's great because it can be subjective at the level of oh, I reached this one milestone, this one horizon, this 13 m target. But if you don't reward it, you it's just effort. If you do self reward it, It's effort -1 in terms of the adrenaline output. </s><s>[INST] I have to uh ask you about this. You're one of the great communicators and science. Some really big family was enjoying in terms of like the educational stuff you're putting into on neuroscience. What's the uh do you have a philosophy behind? It is just uh an instinct, [/INST] my </s><s>[INST] unstoppable force. Do you have uh like what's your thinking? Because it's rare and it's exciting. I'm I'm excited that you know uh somebody from stanford. So I okay I'm in multiple places in the sense of like where my interests lie and one you know politically speaking academic institutions are under fire. Uh you know for many reasons we don't need to get into get into into a lot of other places. But I believe in uh in places like stanford in places like M. I. T. As uh one of the most magical institutions for inspiring people to dream people to build the future. I mean it's I believe that it is a really special, these universities are really special places. And so it's always exciting to me when uh somebody as inspiring as you represents those places. So it makes me proud that somebody from stanford is is like somebody like you is representing stanford. So uh maybe you can speak to what how did you come to be who you are and being a [/INST] communicator. Well first of all thanks for the kind words especially coming from you I think um stanford is an amazing place as is M. I. T. And it's such a </s><s>[INST] mighty is better by the [/INST] book. </s><s>[INST] I'll let it out anything you say [/INST] many friends at M. I. T. Yeah you know high it's smarter friends. Ed Boyden is </s><s>[INST] is [/INST] is uh best you know among the best in class. There's some people not me that can hold hold a candle to but not many maybe one or two. But I think the the great benefit of being in a place like MIT or stanford um is that when you look around you know that the the averages very high like that you have many best in class among the you know one or two or three best in the world at what they do and um it's a wonderful privilege to be there. And one thing that I think also makes them and other universities like them very special is that there's an emphasis on what gets exported out of the university. What you know, not keeping it ivory tower and really trying to keep an eye on what's needed in the world and trying to do something useful. Um And I think the proximity to industry in Silicon Valley and in the boston area in Cambridge also lends itself well to that and there are other institutions of too of course. So um the reason I got involved in educating on social media was actually because of a pat Dossett the bear mild barry called. I was at the turn of 2018 to 2019. We had formed a Good friendship and we were talked me into doing these early morning um cold water swims. I was learning a lot about pain and suffering, but also the beauty of cold water swims. And and we were talking one morning, he said, so what are you gonna do to serve the world in 2019? Like that's the way that like a Texan former seal talks, like we're just literally, what are you going to serve the world in 2019? Like, well, I run my lab is like, no, no, what are you gonna do? That's new. And he wasn't forceful in it. But I was like, that's interesting question. I said, well, um, find my way, I would just, you know, teach people everyone about the brain because I think it's amazing. He goes, we'll do it all right, you guys shake on it. So we did it, you know, And so I started putting out these posts and it's grown into um to include a variety of things. But you asked about a governing philosophy. So I want to increase interest in the brain and the nervous system and in biology generally that's one major goal. I'd like to increase scientific literacy, which can't be rammed down people's throats of talking about how to look at a graph and statistics and you know, Z scores and p values and genetics. It has to be done gradually in my opinion. Um I want to put valuable tools into the world, mainly tools that map to things that we're doing in our lab. So these will be tools centered around how to understand the direct one states of mind and body. So reduce stress, raise one stress threshold. So it's not always just about being calm. Sometimes. It's about learning how to tolerate not being not calm. Um raise awareness for mental health. There's a ton of micro missions in this, but it all really maps back to, you know, like the eight and 10 year old version of me which is I used to spend my weekends when I was a kid reading about weird animals and had this obsession with like medieval weapons and stuff like catapults and and then I used to come into school on monday and I would ask if I could talk about it to the class and teach and I just it's really I I promise and some people might not believe me, but it's really I don't really like being the point of focus. I just get so excited about these gems that I find in the world in books and in experiments and in discussions with colleagues and discussions with people like you and and around the universe. And I can't just compulsively I got to tell people about it. So I try and package it into a form that people can access. You know, I think if I've uh I think the reception has been really wonderful stanford has been very supportive um thankfully. Um I've done some podcasts even with them and they reposted some stuff on social media. It's a precarious place to put yourself out there as a research academic. I think some of my colleagues, both locally and elsewhere, probably wonder if I'm still serious about research, which I absolutely am. And I also acknowledge that um you know, their research and the research coming out of the field needs to be talked about, and not all scientists are good at translating into a language that people can access, and I don't like the phrase, dumb it down. What I like to do is take a concept that I think people will find interesting and useful and offer it sort of like a um you would offer food to somebody visiting your home, you're not gonna cram frog ra in their face, you're gonna say like, do you want a cracker? Like, and they say, yeah, like do you want something on that cracker? Like do you like cheese? Like yeah, like do you want swiss cheese or you want that? Really like, stinky, like, french, I don't like cheese much, but um or do you want frog rod, like what's that like? So you're trying the best information prompts more questions of interest, not questions of confusion, but questions of interest, and so I feel like one door opens than another door opens than another door opens and pretty soon. Um the image in my mind is you create a bunch of neuroscientists who are thinking about themselves neuro scientifically and I don't begin to think that I have all the answers at all. Um I cast a neuroscience sometimes a little bit of a psychology lens onto what I think are interesting topics and you know, I uh you know someday I'm gonna go into the ground or the ocean or wherever it is I end up and um uh I'm very comfortable with the fact that not everyone's gonna be happy with how I deliver the information but I would hope that people would feel um like some of it was useful and meaningful and got them to think a little bit harder </s><s>[INST] since you mentioned going into the ground and uh victor franco man's search for meaning. I read that I reread that book uh quite often. What let me ask the big ridiculous question about life. Uh what do you think is the meaning of it all? Like maybe why do you, do you mention that book from a psychologist perspective which victor franco was? Or do you do you ever think about the bigger philosophical questions it raises about meaning? What's in the meaning of it all? [/INST] Mm one of the great challenges in assigning a good, you know, giving a good answer to the question of like what's the meaning of life is um I think illustrated best by the victor frankel example although there are other examples too, which is that our sense of meaning is very elastic in time and space. And I'm I'm we talked a little bit about this earlier, but it's amazing to me that somebody locked in a cell or concentration camp can bring the horizon and close enough that they can then micro slice their environment so that they can find rewards and meaning and power and beauty even in a little square box or or a horrible situation. And I think this is really speaks to one of the most important features of the human mind, which is, we could do, let's take two opposite extremes. One would be, let's say the alarm went off right now in this building and the building started shaking our vision, our hearing, everything will be tuned to this space time bubble for those moments and everything that we would process all that would matter. The only meaning would be get out of here safe figure out what's going on. Contact loved ones, et cetera. If we were to sit back totally relaxed, we could do the, you know, as I think it's called pale blue dot thing or whatever where we could imagine ourselves in this room and then they were in the United States and this continent in the earth and then disappearing down us and all of a sudden you get back can seem so big that all of a sudden it's meaningless, right? If you see yourself as just one brief glimmer in all of time and all of space you go to, I don't matter. And if you go to, oh, every little thing that happens in this text thread or this, you know, comment section on Youtube or instagram your space time bubble is tiny, then everything seems inflated and the brain will contract and dilate it. </s><s>[INST] Spacetime [/INST] vision and time, but also a sense of meaning and that's beautiful and it's what allows us to be so dynamic in different environments and we can pull from the past and the present and future. Um, that's why examples like nelson Mandela and victor frankel had to include. It makes sense that it wasn't just about grinding it out. They had to find those dopamine rewards even in those little boxes they were forced into. So I'm not trying to dodge an answer. But for me personally, and I think about this a lot because I have this um complicated history in science where my undergraduate graduate advisor and post doctoral advisor all died young. So, you know, and they were wonderful people and had immense importance in my life. But what I realized is that because we can get so fixated on the thing that we're experiencing holding tremendous meaning, but it only holds that meaning for as long as we're in that space time regime and this is important because what really gives meaning is the understanding that you can move between these different spacetime dimensionality is and I'm not trying to sound like a theoretical physicists or anyone that thinks about the cosmos in saying that it's really the fact that sometimes we say and do and think things and it feels so important. And then two days later, like what? What happened? Well you had a different brain processing algorithm entirely, you were in a completely different state. And so what I want to do in this lifetime as I want to, if I want to engage in as many different levels of contraction and dilation of meaning as possible, I want to go to the micro I sometimes think about this, I'm like if I just pulled over the side of the road, I bet you there's an ant hill there and their whole world is fascinating. You can't stay there and you also can't stay staring up at the clouds and just think about how we're just these little beings and it doesn't matter. The key is the journey back and forth, up and down that staircase, back and forth and back and forth. And my goal is to get as many trips up and down that staircase as I can before the reaper comes for me. Beautiful. </s><s>[INST] So the the the dance of dilation and contraction between the different space zoom in, zoom out and get as many steps in on that staircase, [/INST] that's that's my goal anyway. And I've watched people die. I watched my postdoc advisor die wither away my graduate. It was tragic. But they found beauty in these closing moments because their bubble was their kids in one case or like one of them was a giants fan and I got to see a giants game, you know, in her last moments and like, and you just realize like it's a giants game, but not in that moment because time is closing and so those time bins feel huge because she's slicing things so differently. So I think um, learning how to do that better and more fluidly recognizing where one is and not getting too tacked to the idea that there's one correct answer. Like that's what brings meaning, that's my goal anyway. </s><s>[INST] I don't think there's a better way to end. Andrew. I really appreciate that you would uh come down and contract your space time and focus on this conversation for a few hours uh, is a huge honor. I'm a huge fan of yours as I told you, I hope you keep going and educating the world about the, the human mind. Thanks for talking. [/INST] Thank you. I really appreciate the invitation to be here and people might think that I'm saying it just because I'm here, but I'm a huge fan of yours. I send your podcast of my colleagues and other people and I think what you're doing is isn't just amazing. It's important and so thank you. </s><s>[INST] Thanks for listening to this conversation with Andrew Huberman and thank you to our sponsors, a sleep a mattress that cools itself and gives me yet another reason to enjoy sleep. ECM Rush, the most advanced SEO optimization tool I've ever come across and cash app, the app I used to send money to friends. Please check out these sponsors in the description to get a discount and to support this podcast. If you enjoy this thing, subscribe on youtube, review it with five stars on apple podcast. Follow on Spotify, support on Patreon. I'll connect with me on twitter Alexis Friedman and now let me leave you with some words from Carl Young. I am not what happened to me, I am what I choose to become. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] This episode is sponsored by eight sleep and its pod pro mattress. It controls temperature with an app and it's packed with sensors and can cool down to as low as 65° on each side of the bed, separately in the texas summer heat. Even with air conditioning, I can't tell you how awesome it feels to take a nap in a cold bed or to get a really nice full night's sleep in a cold bed. I tend to enjoy a lot of things about life but this is definitely up there. Great sleep on a cold bed and then maybe with a bit of a warm blanket. That's heaven. Anyway, they have a pot pro cover so you can just add that to your mattress without having to buy theirs but their mattress, I have it. It's pretty nice if you want to know my opinion, it can track a bunch of metrics like heart rate variability but cooling alone is worth the money. Go to eight sleep dot com slash legs to get special savings. That's eight sleep dot com slash lex. This is the lux Friedman podcast And here is my conversation with you on me Park. Can you tell your story from north Korea to today as you describe in your 2015 book and with the extra perspective on life, love and freedom you've gained since then, [/INST] wow, that's a long story. So I was born in the northern part of North Korea initially and my father was a party member and my mom was housewife, I had one older sister and I remember born in that country, I never thought I was you know, unusual country now I'm thinking about is I literally called the Heart of the kingdom, but I thought I believed that I was living in the best country on earth. It was a socialist paradise and everybody in the rest of the world were shipped. My dear leader and there was nothing to envy for me. So I had this enormous pride in my heart and great for to be in that country. </s><s>[INST] So his love for the leader, not fear [/INST] for me a list. It was love. Yeah, it was on the migration and gratitude. It changed lately, but for me was pure pure like love. </s><s>[INST] Was there any like looking back with the perspective you have now, would you describe some of those moments growing up as full of happiness or was that delusion at the time? So not knowing the alternative, will you still be able to be happy? [/INST] The fact that I did not know like in North Korea this is the only country in this 21st century has no internet And they don't even know the existence of Internet, not only that we don't even have this 20 like, like 24-hour electricity. So not knowing definitely helped. I think to be seen </s><s>[INST] as a human being, you're still able to find moments of happiness. [/INST] I think my happiness was from family. Nothing else. Even though those days keep telling me that they were our source of meaning and happiness. I don't think I ever gotten happy by that. Maybe they're here and they're in schools and like when I was learning propaganda, you know the proud feeling, right? I mean the greatest nation here and there. But actually true happiness came from laughing with my family. My friends. </s><s>[INST] Are there any childhood memories, pleasant or painful ones? Let's stand out to you now. [/INST] I mean like, you know whenever I think about my north Korea, the interesting is there's no color. I mean one is because north korean country has no color, right? Most of things are unpaved and trees are cut down, we have no fear. So people cut down trees to make food. So but only that like even what we are wearing was like no color. So it's uh interesting. X memory to look back. What </s><s>[INST] about fashion? I've noticed from sort of uh you now you're you have quite an incredible sense of fashion. So contrast that with your time in North Korea, how do you remember fashion just or ways that people can express themselves visually, Was it all bland? [/INST] There was no word for fashion in North Korea. We didn't even know. It was not even our dictionary. So of course I did not know what victory A secret mothers. I didn't even know what mothers were. So when I came out I learned the mother was a job and like what is that? And I'm so confused. So there's so many jobs that we have here doesn't exist in North Korea. </s><s>[INST] What was life like in North Korea is compared to the rest of the world. So maybe he said there's no internet. Uh 24 hour electricity is a luxury you do not have. What about food? What about water? What about basic human rights? I [/INST] think that's the thing. Like when people were asking me, can you tell me about like life in north Korea and in the past I was like, I cannot describe it to you. And initially I thought because of my english that I cannot find the words. It's not that it's a different planet. The common sense that we have doesn't exist there. That people literally do not know the concept of romantic love or human rights or liberty. So when I'm thinking back to my country, it's uh you know, like as you cannot imagine your life on mars right now, it's like that kind of difference. I grew up never seen the map of the world. I never knew that. I was asian, like the regime told me that I was kim your song the first kim race and then our calendar doesn't begin when jesus christ was born. Our candor begins where kim's was born. So we and history was forgotten to us. They didn't teach us about of course Christianity or like the big Bang, like our history began when kIM was born. So everything was forgotten to us. And it was a different meaning. I mean, feeling of existence, you know, it's not even like the same life. I literally think that was almost like my past life and this is like a new life that I began. </s><s>[INST] You're, you're almost like a different human being now. [/INST] Absolutely, yeah. </s><s>[INST] So you've, I have to say, I often say that my favorite book is Animal Farm by George Orwell. I've read it. I don't know how many times and so I was really happy to hear that. That was of the many books. Excellent books they will hopefully talk about. You mentioned that animal, Animal Farm had a big impact on you. It was the book that kind of uh, led to a kind of awakening for you. Maybe can you describe what impact it had? [/INST] So after going through what I went through, I and I arrived in south Korea after many years of journey they were saying. So kim's or dictators and South Korea is not colonized by american investors and americans. First of all, not bastards. They're good people. And then there's everything that you believe in North Korea was a line, it was a propaganda Then at 15 I was thinking, so if everything that I believe was a lie, how do I know what you're telling is not lie. That was so hard. How do I trust ever again? And I just, it was chaos and belief, right? I did not know what was true anymore. And that's the moment. Years later I read this book like Animal Farm just by mistake. It was a very short book in the library. I was like, okay I can finish that quickly and been there ending that like the last chapter right? They could not see between the pigs and humans anymore. Like that sentence. I understand everything what happened, I just made everything sense to me, what happened to me, my people and to my country. </s><s>[INST] Yeah, that there is uh there's so many things that could say about that book, but uh yeah, there's a haunting nature to the young and I guess spoiler alert, but you should have read this already, you're listening to this. Um at the end the animals were looking to the humans and to the pigs and they couldn't see the difference. And then there was this kind of gradual transition from the initial revolutionary steps of animals fighting for their freedom too. Slowly the pigs Gaining control went from four legs, good two legs bad to uh four legs, good two legs better, [/INST] even better at things like that. They were </s><s>[INST] so gradually transitioning the ideology under which the farm operates and I think the gradual nature of that where basically you have generations born not knowing how things were in the past and that's that's what makes the most kind of for me haunting transition from freedom to slavery to suffering to injustice, all those things and the animals don't know they're part of that. And also for me personally, I always kind of found a kinship with boxer the horse because I just kind of an idiot, I just worked really hard and I just love the idea of working hard for an ideal and the tragic nature of to the end that horse boxer working his ass off for for the pride uh for others. But yeah, for the pride of the farm, you know, and then the pigs giving him sort of using that, but then just setting them to the slaughterhouse anyway, when he was no longer useful. I mean there's so many tragic elements that echo everything I've seen in the soviet union and many of the elements that you see even harsher, more drastic way in North Korea. If there's something hopeful you pull from that book, like within the suffering, within the gradual decline, the taking away the freedom. There were still moments of beauty, it seemed like [/INST] it can be, but I think for me was when I was and then the last page of the book until that point I was angry towards the dictator. Why do you do this as a human being? I was so angry, dreaming of killing him by revenging my father the people that he cared but when I was ending the la chapter actually everybody was responsible to create this dystopia in my country that that animal's unusual animals when they're scared when they received the first execution and then they were not doing their jobs, picking out and keep questioning. They had a question and then the essence they feel fear they silence because of that. Like that's when I was like my grandma knew life could be different. I think the one thing about north Koreans are unique is that they don't know they're oppressed. They don't know that they are slaves to the dictator and the fact that other people know they're oppressed. Like in America a lot of people think they are oppressed. Like you are not oppressed, you don't even know the definition of oppression. And like that's like when the new animals came the new animals didn't even know what the life could be like. There's no alternative for them to compare even. And I was like my grandmother knew why didn't they not do anything about it and they were just scared, they kept silent and everybody was responsible. </s><s>[INST] So the people who knew we're too afraid to say and then there's people that just didn't even know and I don't know what's more terrifying about human nature, looking at this group of people who are afraid to say that things could be otherwise. And then the group of people that I don't even know it could be better. It's uh I don't know that that's the reason I returned to that book often. Because it's such maybe because it's interesting using animals to represent ideas that were very human. It almost allows you to explore the darkness of human nature without sort of being broken by it. So, you mentioned anger. When I watch your interviews, you're really calm and collected, not just your interviews, you know, instagram the way you present yourself. You Yeah, I don't know. It seems like you're almost at peace with the world. Is there in private times when you're just angry, do you feel fear? Do you go to dark places, Depression, all those kinds of things? Are you are you able to put that world that you are in behind you? [/INST] It's a joke because I talk about North Korea every single day and I still rescue people like from china and Russia and other countries, right? And sometimes I'll rescue mission fears and they get capturing them back. I should have people in North Korea report to me. So like when I talk to my sister who chose to not be in this life activist life, she forgot most of things. And like for the other hand, I remember everything. So sometimes it's a it's a blessing to keep reminded of how because it's you know, they say happiness relative thing, it is sometimes, I mean I think it's also people say because nobody was fooling when you're growing up, everybody was suffering, you should have been okay, right? But no, like if you are suffering that degree, no matter even if there is no comparison if you're in nazi Germany in the holocaust, right in the concentration camp, I'm sure nobody was better than them. I'm sure they were suffering the same thing I suffered. But now because I'm in this place I can compare easily right? Getting that perspective. But it is true. Like I still have days that I cannot get out of bed and I'm really hoping like that was the llamas talking about downloading your brain blah blah. Like if maybe technology develops that I can download some part of my memory and then I can erase your statement was deleted and that was so much better. </s><s>[INST] This is a sorry for the tough question. But if I came to you, if Ellen I came to you and said we can erase that part of your memory. Would you do it? [/INST] Some days? I would do it for sure. And my mom would do 100%. My sister would do it or other defectors know day 300% for me, I hesitated because I'm a witness. So if I delete that part, I don't know how you're that can be but it is painful. Like after I talked to give a speech right? I mean I'm fine, but some of them depressed sometimes if the top was very intense, I'm not depressed for three weeks, it takes a while for me to be recharged. But I don't know why it is, you know. Yeah, I just don't know. Well </s><s>[INST] there's also the and uh this guy named victor franco who wrote the book Man's Search for meaning and there's some aspect. So he talks about the holocaust and you can, in the, in those moments of suffering still discover meaning, still discover happiness in the simplest of joys. Like while starving, you know, a little piece of bread could be a source of incredible joy. And there's some aspect in which that experience gives you a clarity about the world. Like somehow experiencing suffering allows you to deep, deeply experienced joy and love and also empathize with the suffering of others and like it's almost like brings you closer to other humans. So this double edged sword that um that the highest of joys sometimes are catalyzed by suffering. It's hard to know what to do with that. You see that with World War Two of the stories of soldiers that have suffered, but some of the closest bonds um of brotherhood of just pure love was experienced by them and it's it sucks that our brains are like this uh you know, the love requires hardship, I don't know why that is. [/INST] Yeah, that's like that's thing of course in my journey I learned how to survive right when to not trust and meant to run. But I think most of us keep learning what it means to be a human being. I think that was like the ultimate thing I was kept learning and I still don't know fully what it means, but I do think it seems like suffering as an accessory to stay for people to be grateful and even be joyful to sometimes. </s><s>[INST] So I talked about love quite a bit and you mentioned that romantic love. Uh I'm fascinated about love in many aspects. Um but you mentioned romantic love was forbidden in North Korea. What do you think about love now that you've kind of discovered it? What's the role of love in life? Why was it so why do you think it was forbidden? And North Korea? [/INST] So the tragic thing about North Korea is not only just banning Shakespeare, like we don't even know what Romeo and Juliet is right. Our movies is never about love stories but then also they banned the love between mother and daughter, wife and husband and you know and between your friends, they denied you being a human. So only love that I knew was when I describe my feeling tourist the leader and in a written form, that was the only love that people know in North Korea and now I'm like there are many loves, you can experience, I mean I think you definitely love science right? But imagine that if you're being denied that so there are so many loves in life but in North Korea all of those things are denied. And I think for me love what makes you tick like you know love for your child, love for your parents, love your friends love or even yourself that is denied. So I mean many people say like love is an option but like then why do you live? I think we live to love and it doesn't have to be a romantic love. It can be anything but finding love and in any person or any subject. I think that's a goal. I think that's when people find meaning in something. </s><s>[INST] Yeah, I think we love romantic love is just one sort of [/INST] one </s><s>[INST] echo of the some corps thing. Science, I love science love robots, all of those things and it sounds like deliberately or not. The north korean regime wants to channel that very deep aspect of the human spirit all towards the leader. [/INST] That's it. That's the only thing they allow us to fear. And nobody. So I remember you read 1984 by Georgia where it talks about double think and double speak who controls the language who controls thoughts And why he does talk about as they go. They like you eliminate a lot of words right now like later one or can represent 10 different things and like what fascinates me is like how many vocabulary meaning people can have and like when I literally came out remember when you went to san Francisco and someone came to me and helped me and then he was, the guy is like oh baby don't worry, I'm gay. I was like what the heck is gay? I don't know right? And then they had to go to hotel and google to get and it's like oh that's what you meant and like that. Like they deny what that is. I'm sure there are gays in north Korea. I'm sure there is but you don't know what it is and like that they eliminate words. So the fact that you know the concept that is the state's much better than and that's the thing, a lot of people like when you're born you somehow know what justice is, what liberties and it's or somebody taught you that and like that's the thing people is like oh humans are inherently know what is right, what is wrong, what is oppression and like no that's like Bs you gotta learn </s><s>[INST] that's fascinating that words give rise to ideas. So like as a child, one of the ways to learn about justice and freedom is to first line in the word and then to ask well what is it, if you don't have the word for it then you never have the kind of first spark at least you trying to be curious about it That's interesting and controlling the words and then yeah, I [/INST] mean </s><s>[INST] you control the thoughts. There's so many echoes. I mean I have uh it's a very different but perhaps a very similar experience which is the journey of my family through the soviet union. Because there is a love of country, there is a pride of the people like you are proud of your family in general. But I wonder how much of that is polluted by the propaganda [/INST] a lot. For sure. Yeah, it is here this day. I'm like my father who died in china and he was tortured and then he died, he wanted to go back before his death right? And then it's like that if you go back you're gonna be executed and like I want to be executed. </s><s>[INST] He wanted to go back to North Korea [/INST] to be executed so he can be buried in his own land. And then his last wish was if I die crimen ate me and then bring my ashes back to my country when I'm dead. I still want to be in my country and this is nationalism. This is a propaganda right? And but now like it's the same thing, like it's the same thing if I die, I somehow buried in my land and I still feel like an outsider. I'm always longing for my home is the heart of her home. Like my people say, what's your dream? Like, do you want to be a present during the run for office? Like, I just want to go home, that's my dream, right? And people here don't get it ever. </s><s>[INST] I don't know what to do with that. I love my country and I think for me, my country is um the United States and perhaps it will be for you to one day. [/INST] It is, I think it's becoming us has been a very special place in my heart. I think this is the first place I felt like I I feel like home. And I mean, I was in South Korea longer and I didn't feel that way. So </s><s>[INST] I think there were very different life stories, but I think it's almost two different people uh for me is the person that was in the soviet union and the person that's here, those are two different people. That previous person's home in the soviet union, and he's part of me. And I suppose in that same way, you're, you know, your first maybe two decades of life are somehow longing for the home that is North Korea, and your next two decades of life might be finding a home in the United States? Your your dad, can you tell the story of um of his struggle uh of his death? Mhm. Let me first. Do you miss him. Do you think about [/INST] all the time? Like, I had a son when I was 22 and I had IV. times, And I'm like as you say I'm like 80 lb. But back then I was like £75 because of my mess of your malnutrition. Somehow my body is very different. And so after three times of a V if I've ever after 23 I was really wanting family and the reason I wanted him is because I felt so guilty for my father that he never seen this world. I somehow like when you're so desperate you becoming logical like I want to believe in that I really like buddhist idea right? You come back to life and I prayed please come to me and I guess my son so I will take care of you. I come back and when I was pregnant with my son even though I planted pregnant with a girl doctor made a mistake, it became a boy. So I made his middle name. Like my father's sentencing. I think he's the only american got north korean name. It is </s><s>[INST] so he's it's a part of your father. Isn't your son. [/INST] That's how I that's how I make the sense of it and that's how I move forward. If I like as a logical human being. You know when you're dead you're done. Maybe like that's what I at least you think. But then life becomes too unbearable. And somehow that's the thing like we tell ourselves stories in order to live and that's how I came with my title of the book in order to live, I had to tell myself a lot of stories to overcome a lot of things. I think I was just part of it. </s><s>[INST] Can you tell the story of you escaping north Korea to china? [/INST] Yeah, I think it's a thing, it's amazing. Um Even though I was like 13 my like Life outside North Korea, it's almost like went by like one and my life to that point was like eternity. Remember Being in China. I arrived at the end of March at 13 and by October it was six months passed and I literally felt like I lived eternity and one day live in china For like living one year, one day I was a war like surviving through one day was so hard every night. I was like I cannot believe I got done one day today. That was the thing I was grateful for before I went to bed. Okay, I survived, I didn't get captured and I made you another day on earth. </s><s>[INST] So the experience of the minutes is what fear, fear of being captured, [/INST] Fear, loss. Everything. It's a because I mean I sought my own mom in china to survive two. So it was more than that and it's not feeling. I think that's the thing in china. I learned not to fear and after my escape was a challenging, I didn't feel anything and it was hard not feeling anything is a torture. It's a bigger torture. Can never feel like even if your sadness that's better than not feeling anything and I feel something when I had my son that's when I started healing. So he was a miracle to save me. But yeah, in china, it wasn't even fear. Like it was numb. </s><s>[INST] You're numb. It was like paralysis just overwhelming on the uncertainty of your future. Did you have a sense that your future held at the time? [/INST] What do you even even feature? I don't even know that word. I like a lot of times I was looking at myself like I left my body and like just looking at me and just not feeling it. It's not like I'm scared for her. I'm like sad for her. Just looking at me like, oh, that's interesting, wow. Not feeling anything and mean like being raped going through every emotion of life to survive, right? Like, But like, like somehow I don't know if you say so or something like looking at it just like you feel nothing. You don't feel anything for that person. </s><s>[INST] So even with your mom, like what was, was there some, I don't know, some warmth that you were able to extract from the connection with your mom? [/INST] Yeah, of course I think that made me survive. I had a very strong connection with my family And I think that's what kept me going to do all of that. I think as you said, I escaped at 13, my sister at the age of 16 escaped with their friend first and I was going to escape with her. But one day I got like really bad stomachache. And my parents took me to hospital in North Korea hospital. They don't have like three machines, like they don't even have electricity. They literally using one needle to inject everybody. And people don't die from cancer in North Korea die from infection and fever and hunger, right? It's most likely you're gonna die more by being treated by dr they're not not being treated. I think I was lucky even though they thought I had appendix, they operate on me without any pain killer. And I didn't get infection. I survived. So that's how I got delayed to escape my sister. And she left me a note in my bedside saying I follow this lady and this is another trick about human trafficking way. She sold me to china as a sexual slave and she executed for it later. And she had </s><s>[INST] was executed for that later. [/INST] She had five daughters and she sold all her Children to China. And we can now sitting here judging on like how heartless you are selling your own Children to China and as a sexual slave, they were like her Children were 17 years old. But that was the only way for her to save her Children. And if she didn't tell me that they I would be dead right now. So I'm grateful that she sold me. And I think that's the thing is a life is so crazy. You cannot judge. He's just so complex. And Yeah, that's how she changed my life by selling me. She sold my mom and myself in 2007 to China. </s><s>[INST] So you're grateful for that. You're grateful for that suffering. [/INST] Of course I am great </s><s>[INST] because the alternative is worse. [/INST] I will not be here with you. You never knew what I just said. </s><s>[INST] What do you make of the others suffering in the world today? The people there in North Korea. So that is part of the your of your life's work is helping those people. What do you think about them? What should people know about them? [/INST] I think that's when I get angry whenever I think about them. Like </s><s>[INST] I know your anger directed at uh, [/INST] the heartlessness of people, the ignorance of people. Like so when I got on North Korea and going through all of them And I went to South Korea one day I was watching television and this is like a famous korean K pop stars and crying and doing some fundraising concert. And I literally thought I was like, oh my God, something is a hollering going wrong in this country. Why are these people crying? It was cherry like campaign and then later was china was animal rights campaign to helping our cats and puppies in the shelters. Do you know anybody says they're tears like that to another human being right now. I know right? People rather give $1 million China. And I think I love in a mosque. I read this way. I love these people want to like go the moon mars and then people told them like yeah we went to the moon like I did not know in north Korea. I think that's what upsets me. Like why there is not even one single human with that kind of brilliance in their brain. They can't save so much suffering but nobody does anything. I think that's been I I feel like hard to find hope in humanity. And that's when I guess the upset because think about like even biden or trump Obama. They know what's happening in north Korea. Exactly. Right. I mean we see satellite photos, there's public executions. I mean the U. N. Says this is the holocaust happening again. And is it happening if the holocaust's happening again? How why? How are you okay doing nothing about it? But somehow humans are able to okay nothing anything. And this is like this is a hard like when people say I'm gonna change the world, I want to make a difference. Like it's hard to believe it. You know. </s><s>[INST] Yeah. That we can turn our back to human suffering at scale when it's right in front of us. I mean that makes you think about the holocaust this is just everybody was looking the other way because it was almost too hard to look at it. [/INST] No it's not it's easier thing like that's the thing. I was like here to speak at the South by southwest two years ago and like everybody's talking about like Elon musk project going to the moon, right? We're going to be more to spend like species. I was like back then not even know who he was. So if you're trying to go out of the earth, you haven't even explored our earth yet. You cannot go to north Korea right now. You haven't explored that part of our our like planet. Can we do that first and then move on to explore </s><s>[INST] the landscape of human suffering like alleviate suffering in the world. There's there's a lot of suffering happening in africa that has to do with disease and for some reason it's even though we turn our back to that kind of suffering to we still can try to do something about it. And they're still efforts uh in terms of health care in terms of medicine, in terms of by engineering, in terms of like all these efforts to help people from disease. But like that's almost like converting it into an engineering problem in trying to solve it that somehow it's easier for us humans. But when there's obvious sort of non disease related torture of humans. We look the other way. Yeah. Whether it's china or north Korea. I'm I mean that has to be changed somehow. We have to change that [/INST] somehow isn't the thing right now. Like china like they bring the Xinjiang *** right? They've They say Oh it's a vitamin take it and then it kills their sperm and they cannot reproduce their birthday, gone down something 47 to something 50% in the one year time. It's a genocide in 21st century. And they get those people and get their like organs out. Imagine if there are some people who do that cute puppies and cats, there's gonna be insane among the product, they're going to destroy everything. And this is like a human nature that I don't get why there's so much anti human sentiment in this modern world. We don't have to. The fact that I was saying like the fact that you care about animals rights is beautiful because you care about something who cannot speak for themselves. The fact that we care about animals because they cannot speak for themselves, right? They don't have it ability. And there are many people who cannot speak for themselves right now. And why do you refuse to be the voice for them because they are simply being a human and maybe it connects to us not being proud of who we are. Like maybe I don't know what it is. Why do they deny humans this way? Maybe they don't like themselves. </s><s>[INST] Yeah, it's almost uh, we would have to acknowledge some dark things about ourselves in order to start helping. Yeah. Um what's the solution? So You know, I see two solutions. One is in the military side, it's assassination or the full on invasion and then on the activism side which is figuring out ways two, like, like you said, sort of let people in north Korea understand their situation sort of from within try to reform or maybe there's others. Obviously there could be activism from the outside to build up momentum for the entirety of the world, especially the world, that it's not just the United States or europe but also as Russia and china and so on. What what what are your ideas here, how we can what we can do as individuals and as countries [/INST] then the first thing that we can do is speak about chinese raw in this sponsoring dictatorship in north Korea. Like I haven't had so much struggle talking about north Korea right? They say how north Korea is possible. Why it's like the way like this is 99% accountability going to CCP kim jong un cannot last without chinese have been one week, this is completely funded this holocaust is a funded by CCP but if you talk about in the mainstream of course they don't buy it and I think it's in a way North Korea is a lot easier to solve than even in the Middle East. There's nothing conflict like between people. There's no ideology, no religion, nothing people are peaceful, right? There's nothing one civil, any discontent among the people. All problem is there's a dictator funded by the second economic power in the world and even any military, they know if they killed him jong un, they want to get killed by chinese. Nobody can dare to stand up against kim jong because china is backing it. So somehow here in the west to be collectively acknowledging that china is the responsible person for these crimes against humanity in north Korea, then we can somehow, I don't know how to stand </s><s>[INST] up to china. Exactly were failing to do that in a way um in all kinds of avenues of life of public life because for many reasons they're probably primarily financial. But it also I'm against, I don't know, maybe you can correct me, I'm against sort of making china this evil enemy because I've seen this with Russia as well. And I don't think that leads to progress. I think you want to highlight like you basically want to help china the chinese people become the best version of themselves so speak to the chinese people and not fear not uh making the leaders of china like into these caricatures of devils. I feel like the cold war, the way it was done in Russia. I just for both sides, they were caricaturing each other through propaganda and the result was not productive at all. It did not help Russia becoming the best country could be, did not help America become the best country it could be. And the same thing with china, I feel like making them into this enemy like being afraid of china being uh making them into the thing that's going to spy on us, that's going to destroy the rest of the world. That's not going to help china like reform themselves. They're going to plant their feet. The dictators, the evil people will become more evil, the power hungry will become more like they will centralize the power more. It feels like um may be naive but it feels like it should be like again love not violence that solves this thing now of course in north Korea is like long gone, [/INST] 80 years, 80 years, </s><s>[INST] you can't love is not going to solve that problem or I mean I don't, it's very difficult. [/INST] They have tried that because on the sunshine policy which is there are two people walking down street and the sun and have been in the middle battle so who can take off that man take off jacket so we tried to blow as much air he could and then that man was like putting more like his jacket on, right, not taking off my sunshine came up. Okay, I'm going to give a lot of worms and then he took his jacket out and came out So that was the theory, Let's give North Korea as much love they want. Let's give them a lot of money, whatever they want. Let's give to them. Do they know that we are not here to attack them and North Korea? What they did was the guy who did the sunshine policy in south korean name, kim D's on one another. Peace prize for that and keep your skin. Jr used the money to build nuclear weapons. So that's how they came with the nukes. So I think that's the thing I hope you love solves problems, </s><s>[INST] But there's gotta be a way and that's the hope is with the 21st century's you can directly speak to the people somehow when there's no internet, when there's nothing like that, it's hopeless. I think China, there's a hope that China is still connected to the Internet. [/INST] I love your optimism. I have seen the actual dark side of china on the underground, I hope, I think that's the thing, people in the west, right? They said, how can you be that bad? They asked me like I walk in passing this a young teenager man and later the world with my sister, he's like intestine coming out through his bad way and even in that moment, what he wanted was please give me food. He was hungry, </s><s>[INST] his intestines hanging out of his body and he's [/INST] asking for food, you know what humans that demand when they die in North Korea, all they want is eating right and people say, oh nothing can be that bad, but people just here haven't seen an actual, through evil, </s><s>[INST] would you say that the evil comes from a tiny minority of people? Or is it permeate much larger parts of the population? Like if we look at sex trafficking, how many people Like is it 99.9% of the people are are longing to do good in the world or is there is it uh or do we all have the capacity for evil and certain kind of environment? Certain kind of governmental structures inspire a large percent of the population to do bad things. [/INST] I think humans are capable of anything, there is no exception. I don't think there's anything to bond with the morality. I think in North Korea you can say initially that there's a few guys in the top one to the power and then doing this but eventually made a society where people don't even know what compassion is. You don't know the concept of, we don't know that you need to feel bad for another human being and they're suffering. The fact that you know, compassion is in your knowledge, that's what you do that humans need to learn. It's not anything bad about human nature is just saying humans are capable, everything we are the most adaptable species on the planet. That's why we created the internet like talking this way right? No other animals have done it because we are so adaptable. That is a good thing and that's a bad thing. So in the end that situation they all can be. I mean during the holocaust right, those people, they could have been capable of good too if they were exposed to different system. And that's why when people underestimate evil, that's what scares me, evil is evil. It's a different thing. It's a complete different thing. And of course I Get your idea, we don't want to isolate 1.3 billion human beings in unearthed by Chinese. But the thing is we are talking about this regime, not the people. I love Chinese people, I speak Chinese, I loved all about the country but this system does promote evil. </s><s>[INST] Well that's an optimistic view actually because we can fix systems uh it's hard to fix people. So if we fix systems and the people are adaptable [/INST] as you said, </s><s>[INST] I mean that and then the question is uh first of all you have to talk about it just as you're doing your right now like this little flame that burns bright and it's really important for North Korea. But just keep talking about it until there's until hopefully it leads to at the highest levels of power revolutionizing the systems in the world and then uh in china and north Korea, do you see North Korea being a potential instigator of a nuclear war. [/INST] They will not start in the korean war as long as they can do whatever they want right now. Right, North Korea's army Not designed to fight the enemy. They designed to prevent their own people, the coup d'etat and the revolution of their own citizens. That is 1.6 million of North Korea with a tiny country, the fourth largest armies in the world. Mhm. So this this country designed to fight their own citizens </s><s>[INST] and the army, the fourth largest in the world, is designed to basically fight its own people, [/INST] oppress their own people. That's what North korean military is about. </s><s>[INST] Okay, let me ask you some aspects about north korean life, can you describe the song bone system of uh ascribed status used in north Korea? [/INST] Yeah, so that's a very interesting thing right? Right now there are a lot of people playing with this ideology of like democratic socialism, socialism, comments or whatever you call Marxism. Leninism, right? They have all like this similar features where we give collective power to a certain entity and they will make the decision for the bigger good Right? And North Korea came up with the idea that kim Il Sung, he was the leninist, he was Marxist saying I'm going to create the most equal society on human face. So it was a communist North Korea and then they came up with this humble system is like family caste system, three big categories, warrior wavering and hostile And that in building three classes, they divide into 50 different classes. So a lot of people don't even know which exactly as you belong to. That's a secret government document and that's how they decide your future. So, you know where in North Korea before you're born, your life is determined for you. And this is almost joke, right? They dreamed of creating the most equal society. The N. D. W. Became most unequal society in the face of humanity. So there are 50 different classes and where the one guy on the top became a got. So when this animal farmers, we keep saying like there's so many all the animals are equal and some of any more than others. Exactly. But it's not only it's just more equal one guy in order to became a god. </s><s>[INST] North Korea was born out of Marxist ideals. [/INST] Can </s><s>[INST] you comment on uh douche ideology, which seems to be its own kind of socialism, but with unique aspects here, it really does ideologically says the importance of having a great leader. Um Is there some interesting similarities or differences that you can comment on between other implementations of communism throughout history? The soviet union china elsewhere? [/INST] So to taste very unique, it came around the 90s after Soviet Union collapse. So before that North Korea was very still loyal to the Max is a mechanism which the state takes care of you. We are going to give you the right education, healthcare, your livelihood everybody is going to be equal, you're gonna have in the working collective farm collective workplace, everybody collectively do things together and let's work for the paradise. But 1991, the Soviet Union collapsed and until then North Korea's are heavily subsidized by Soviet Union's eight and then so you didn't give them anything. So not three million people dying on the streets. The regime then came up with the idea, okay, our goal is what is successful, willing for us is keeping the 10% of population alive, which is in the capital, Pyongyang. So they designed the Hunger games, there is a capital, 13 other districts. Everybody on the countryside on purpose being starved. So those people who are starving cannot thinking about meaning of life, cannot thinking about shooting to the moon, right? They're not gonna think of anything or they don't think it's like finding this meal </s><s>[INST] all on purpose. [/INST] All of it is man made famine. International committee was begging to give north Korea food. Why now stood at the U. N. They beg to give north Korea formula medicine and food. They are begging, can you please video people and say no thank you last year like we know it's got a horrible, horrible flooding south korean president begging, can you get, can I give you police and medicines like no because he wants to be don't want provider, he doesn't want people to think of other people giving him the thing so on purpose, other people starving And the Tooty ideas. That's when you're coming from. So until that communism was about like state is being a father figure, takes care of all your knees, right? Give the power to us and you're all good. But north Korea regime says, okay now we cannot give papers Russian. So which means to me means self reliance. You need to take care of yourself while you're giving everybody to us. So now in 1990s, the region told us, Okay, we are not going to give you Russian. You cannot trade. That's illegal. But you find your own way to survive to be self reliant. That's what you say is that. And you know, but when you are God, you can do whatever you want. You don't need to make a sense. That's a difference. Being a God and being a leader, even in this religion, it's not for certifiable. You cannot challenge it. God's way is suspicious. God works in a mysterious way. So when your God, people not gonna say, oh this doesn't make sense, right? You're gonna okay. Whatever God says as a human being, we can never change his thought. It's unbelievable what regimes can do. </s><s>[INST] There's something about famine. You know that um is another is another level of evil to me. You know what Stalin did in Ukraine in the 30s. Fuck them. Mm. This is what torture is cannibalism and [/INST] North Korea to the humans Right now in 21st century Seven billion people on this earth right now. You make enough for 10 billion people. Nobody should have starving right now. It's worrisome to me. The humanity is moving forward with the technological advance Blah Blah. We're going so fast in in advancement and we are living this like 25 millions of human beings in the cage completely leaving them behind. And North Koreans living like 16th centuries and never like this morning I was taking a shower, Beautiful shower. Like one never knew what she was. I was a bathing few a few times a year going to the river. How do you even know what shampoo is? And this is how human beings in 20 persons living and it doesn't bother us and rather most of people obsess being a vegan and like how how do you reconcile this? </s><s>[INST] I think we get used to stuff very quickly when you get used to comfort. That's just the way of human life. You take the beautiful things for granted. Um So I try to appreciate everything I have. So whether it's uh like the food I have now or like the luxury to have a diet and be struggling with that or just the basic, simple moments of being alive with the people I love actually get like I think I'm on drugs all the time because I feel like just even like the smog everything on this table. It just brings me joy. But it's like filling your life with joy in the full capitalistic american way you can still at the same time uh not feel too bad about yourself and still focus on the suffering in the world. And I think there's some way that in trying to build a better world in America it has ripple effects elsewhere sort of like so I'm a fan of rockets in space, it sounds perhaps counterintuitive but sending rockets to space will help solve the north Korea ah a problem because it lets people dream and build cool stuff. So it's not the rocket, it's the other people that like are inspired by the rocket and then look to other problems in the world. I mean that's what Ellen did is like he solved problems in the world and so like what can I do to help it? And I think the north Korea one is a tough one though because that's ultimately has to do with revolutionizing government [/INST] china, that's china, that's what it takes. Chinese chinese Communist Party is impossible. That's why we couldn't solve North Korea for that many decades. </s><s>[INST] But china for now is china but it's china, it's um so it's Russia, it's certain aspects of the United States and struggling with that. One of the you know there's a bunch of technologies that are striving at this. Um for example uh I don't know what your thoughts about cryptocurrencies. [/INST] So </s><s>[INST] like there's a idea that money could be a way to destroy or to challenge the power centers of the world. So if you give, if you take away the power from fiat currency and give it to this thing that can be controlled by government, that's Cryptocurrency, whether it's Bitcoin at the and all those kinds of things. That's the way to get money into the hands of people to where the government can't take that money away. [/INST] But north Koreans don't have electricity, no internet so we can do that with china. We can do it a lot of african dictators countries right? I do think big cryptocurrencies such a fascinating technology right? I think this is an amazing experiment when that power is in our hands. I'm the huge african believer but I think nurses to behind you know, I think that's what is unique about North Korea. Is that most of things that we talk about is now please different planet literally the common law that we have is now applicable. </s><s>[INST] Oh what about kim jong un kim jong un? Is he intentionally evil or is he mindlessly mindlessly propagating an evil system created by his ancestors? What's your sense of the man? [/INST] So with kim your song I can give him more. But I feel that he was a initial true believer of communism but then as later he gained the power he realizing I think I guess back then he thought most of people are done by individual stump. So therefore I need to make the decision for all of you. That pure arrogance came from out of him, even that I can tolerate. Okay fine and kim jong un who never like yeah fine. He grew up in the system to you but kim jong un is very unique. This guy was educated in Switzerland in the heart of democracy. He knew how human beings should be treated as a child. He went when your child, your brain is very susceptible But you can change anybody. Like why the mall was so that's like changing young people's minds. Like that's every revolutionary. They do right, they go change young people's minds. 1st this guy was so obsessed with power. Him being a god, even starting in Switzerland didn't change him and that's why I think that's a pure evil, you know I can't give it more benefit out to his grandfather and father. But when it comes to kim jong un, this is like what pure evil looks like pure selfish being. That's what it looks like. </s><s>[INST] Is there, is there some sense where he's justifying everything he's doing to himself or do you think there's a psychopathic aspect to where he enjoys the suffering? [/INST] I think in his life, right? I read a lot of like north Korea, a lot of CIA documents a lot of intelligence people work there and even like worked in north korean tabulates and escaped, I can hear about them. So kim jong un when he, they're born, they treat like gods. So they never have a sense of then being a human, they're like equal with the others for them. Like we are just any kind of tool like that. We're not pulling that thing does right? Anybody is a tool like once boxer dies getting slaughtered for my cause and they do not even feel guilty about it because they don't view us that you deserve, you're worthy of it. So it's not like he even feels he doesn't even recognize that the suffering like of course you you this this is what you do serving me because I am, I am this. So I think that's like beyond that it's not like suffering. His enters his mind. He doesn't even think what we go through. </s><s>[INST] So he thinks himself as a god and then everybody else is just tools that they're disposable. Right? There was rumors several times of him dying. Do you think he is? Um Obviously his health is not good. Do you think he will die soon? What happens if kim jong un dies? [/INST] Why? When it comes to north Korea, anybody knows what they're going about, kim jong un does is lie right? Nobody knows. I'm sure C. I. A nose but they may never revealed that C. I. A. Has enough intelligence to can tear where kim jong un is what he's doing. They just don't assassinating because they don't see the means of it right now because you </s><s>[INST] think they can assassinate him, they can, they do have ability to get why the hell did they not assassinated? [/INST] Because they don't care, they don't care about the suffering of 25 million people. They got to pay the price. If they assassinate Kim Jong un, they gotta pay the price that </s><s>[INST] the financial, there'll be political price to pay it will anger [/INST] china. Absolutely. That is a huge piece for them. </s><s>[INST] And then they'll have to deal, obviously there'll be financial military consequences of having to deal with the turmoil. The uncertainty. Yeah. The revolutions that will spring up. Yeah, [/INST] that's the thing like that's why they don't want to take that risk, they don't want to do anything. The US now became very passive when they persecute these moral values to the rest of the world. </s><s>[INST] They did the same thing with the holocaust in the early days actually, [/INST] they were like just didn't care and that's what they're always policy has been, they don't care. I mean, so if kim jong dies, it's gonna be very hard for north Korea to replace anybody in his position because kim's is a brand, It's not just like a leader for us, right? Whenever we think of king who came with my mind, like who's like almost God figure that North Korea is the number 10 religions in the world, they copied the Bible. So if you believe that if there are people believe in the God and just christ. How do you not believe the north, we believe in the same thing. So keep your son's grandfather and his parents were devout christians. So kim your son was grew up this like christian like work versus so when he finds his country, he said, I love my people so much that I'm giving you my son, kim jong un his body dies. But his spirit is with us forever. Who can know how many here I have what I think and when we suffer when we go to paradise with him and when you block every single information going into country, of course people want to believe it. </s><s>[INST] So who would be the successor if he dies? [/INST] He has a son. First son born 2009 and not not old enough if he dies now. So either his sister, my rule for a short amount time. It's not like a leader, but like we play like temporary placement and then when the sun is older enough he might take it off. This is a kingdom that's most likely and china will do everything they can to maintain the status quo for the north korean regime. So north korean people have no option here. We just need some later to courageously come up and do the right thing. </s><s>[INST] So we can't just wait this out. [/INST] So we can, it's not something that I take its course and I'm going to change. Like we not even know that economic freedom does not bring political freedom, we know in china, it doesn't, that's as unique thing about freedom, you gotta fight for it otherwise you don't never get it. Freedom is something has been fought and if nobody fighting for freedom is not gonna be there, </s><s>[INST] can we talk a little bit about freedom? What does it mean to you having had, we talked about love in that same way about freedom having sort of discovered it later in life. Uh what does it mean to you? [/INST] Yeah, I think every day I get a new definition of freedom. It is, it's a never ending journey having this relationship with being free and what it means to be a freeway. I think I think uh you definitely can live life without being free and also happy life too. I saw a lot of nerves can inlets were fat and have power but didn't have freedom. Were very happy in a way happier than the people that I found in your or like investment bankers and consultants in Manhattan and 70% of them go like talk talk therapist. I was very confused. I remember writing my book in new york like my editor were saying, I mean you know like you're traumatized, you need to go talk to a therapist and like what is therapy here? What is trauma because in north Korea they don't have words for stress or trauma because how can you be stressed in the socialist paradise. So they don't let you be knowing what that is. So And then they were like, yeah, hearing people having problems, go talk to a therapist and I was like, like how much is that if you have $200 per hour and it's discounted right to say no thank </s><s>[INST] you. [/INST] You know, I was like, you know, and we know that freedom comes with responsibility and in a way it's not that easy to be free thinking for yourself constantly. Like when you, in a way I understand like let's give government every power we have, let them decide what education that I get, let them decide where I live. Like, you know, let someone figure that out for me and that's how North Korea began hoping the government's gonna represent my own interest, believing that they were good and with that benefit of doubt and good faith, you began the nightmare right? So freedom is not like gave way to be happy at all in a way you can make life a lot more complex, but then it's fun, isn't it? You start being for yourself. You start making mistakes and it's so fun to be free, even though you can be suffering way more than the people who are not free. </s><s>[INST] The thing about freedom is when you have freedom, you also have the responsibility for your actions and that could be a huge burden. Yeah, because if you succeed, it's you, but if you fail, it's you. And if you do horrible things, it's you, if you don't do something, for example, if you don't help people in North Korea, it's you. And that's a huge burden. And living with that burden is a kind of suffering. I mean, there's some aspect in which freedom is suffering [/INST] because </s><s>[INST] life is suffering and then the freedom is uh, you as an individual fully living through that. So you talk to your friends with Michael malice, he believes. And so I want to kind of ask you about government. Uh, he believes he's an anarchist and he believes kind of freedom fully implemented in in human societies, meaning that humans should, uh, all be free to choose how they, you know, transact with each other, how they live together, There shouldn't be a centralized force that tells you what to do. Do you think there's some role for government in um, in a healthy society? So if you look at North Korea, there is the most horrible implementation of government. But then if we look at what the United States strives to be at least in principle, there's uh, there's an ideal of a government that represents the people and helps the people. Like, is there, is there a place for that kind of ideal? Or is government always going to get us into trouble? [/INST] I am not. I mean, I spoke to Michael malice, I kept asking why why is anarchist, right? And he doesn't even believe military? None of it. And I was like, I was like, I don't think I want to be in the world you're describing when that's really scary. I want the law enforcement. I want like I don't in a way that so why equality makes no sense. Is that the fact that we knew and I were born, we were born in a very different capability of thinking, different intelligence, different, capable in our physics, Right? So, equality is nonsense. You can never achieve that. All right. So, to me, that's been it's very scary when the government equality on everybody. That is impossible. That </s><s>[INST] specifically equality of outcomes. Like given that we all started different places, enforce like measure in some kind of way where people stand and if they're an equal enforce equality. Yeah. And that's what leads to the kind of things that you mentioned with the class system in North Korea. [/INST] Yeah. So, I think that's why government can be bad. They can be very dumb. And another thing is that they cannot know what you want. A lot of times people don't even know what they want and individual like how the heck do you or some government gonna know what is best for you? Nobody knows. We just all do our best. I do think there's some governments against Switzerland, you know, have more give power to the different state chemical. I think I'm more, you know, like giving power to the state and the individual decide where they want to go ingredients taste like I mean why did you choose texas right? There's no income tax, right? Like there's a lot of things people find texas like you know charming and they come here so in a way that I don't want to be the one strong government makes every single thing the same way in a way I want to kind of experiment to everything. We can have anarchy state, there's no police, nothing going on. You can be whatever one you want and you can go on stage really is like abortion is bad blah blah. This is bad. All the conservative values and let the ideas compete and let them how they are being pregnant in real real life. But I think it's very scary when the U. S. Government is getting bigger and bigger and then they try to make every state under one big government. And that's like when I get really alarmed, </s><s>[INST] are there things that you see in the United States in the current culture that kind of has echoes of the same things you saw in North Korea. That that worry you that much. [/INST] Absolutely. It's it's uh it's in America now the meritocracy doesn't matter right? It's evil. The white man's idea of like talking about if you're competent enough, there's like oh if you're coming from which right family you are going to be competent. So other people don't have a chance. But look at asians who came from, nothing is competent and go like Harvard Law school and medical school. So it doesn't almost, it's like there's no incentive for you to work hard anymore in the system right now that is north Korea, there's no incentive because you're born with your class already. So no matter what you do you can never know the whole everything about North korean system is that there's nothing holding marry up. So if you're coming from other cultures that like make a market, joined the royal family and she became a lawyer, you go up by North Korea. If someone from high class gonna marry somebody down, you only go down with them. That's how they prevent class mix </s><s>[INST] right? They kind of enforces the separation because there's a huge disincentives to go to marry too integrate between classes. Yeah. What do you do about this kind of um you know, especially in universities but in, cos I'm thinking about starting a, cos I'm looking at this very carefully, there's these ideas of diversity and meritocracy that's attention. So I think there's there's a big way in which diversity broadly defined is not at all in tension with meritocracy. So having a variety of people backgrounds way of thinking all those kinds of things is a huge benefit to to any group. But the way diversity is often defined is by very crude classes of people, whether it's by skin color or agenda or or some very kind of large group way and that that actually does two things in my mind. One, it drowns out real diversity or not real, but the whole spectrum of diversity which is like within class diversity of like um, are you, are you somebody who is um, are you someone who is exceptionally good at mathematics or somebody who is exceptionally good at psychology? Are you good with people? Are you good with numbers, all that kind of stuff that I think spans or intersects in fascinating ways with these kinds of groups. So that's diversity. And then meritocracy is this thing that probably the reason I wanted to move to silicon Valley and the reason I didn't is like having a fire to change the world within you. Like meritocracy is like I want to be the best in the world at this and I will strive and work hard, not stepping on others, but like purely within yourself, be the best version of yourself. That that idea is um in some ways being not celebrated or [/INST] demonized, literally meritocracy is being demonized right now in America, working hard is a symbol of you coming from some established family. The fact that you celebrate accomplishment, hard work is a sign of your patriotic, what everything they call away and they want to abolish that they want to like stop giving kids grades, That's what they're already doing, right? They want stuff. They were like, we should abolish like S18 America they take to go to college, right? They won't even abolish that. Yeah. Some kids have no ability to do math. So why do we have to force him to learn math? And that's what comes with humans, overcome challenges. That's what makes us special. But then like because these kids coming from this family, let's find the reason why they cannot and then they don't have to do that thing but they still deserve the same job. They need to be a lawyer and doctors and that's like what in North Korea was like not, there was not a meritocracy begin with. Where did you go in the same family? The family the blood right? Like if one person does something wrong it's a collective guilt because I spoke out to the generations of my family got punished who are left behind. And then in America I see the same thing. Like if you're somehow great great grandfather owned a slave now you are privileged and you're guilty because you are white and get them. But how do you change your ancestor? How did you have a thing on it? And that is where there is no way out. There's no forgiving. There is moving forward and this current culture in America now, like I remember at Columbia like before class everybody has to go around of saying tell us what's your pronoun is and my english my daughter language I learned as adult even saying he and she I'm confused by pure mistake. And they said call me day because I'm gender fluid, basically, I can be a girl. But next hour you talk to me, I'm a boy, right? And if you don't do it right, they like look at you, why are you pick it? Right? And I see me so nervous and this is where I come to. This is a regression of civilization. We are regressing as a humanity here. Like the alignment, all of those things made us so much brighter and looking forward. Now we're going backwards. </s><s>[INST] Well, I think there's a pendulum aspect to it because it's my hope in terms of backwards. So, pendulum goes backwards too. But it just goes back and forth, I think. And then in the long arc of history, we're making progress. I think all of the discussions of diversity and inclusion and all those kinds of things. I always thought that they're healthy in a in moderation, right? There should be a small part of the conversation amongst other things. The natural asked like it seems that they kind of have this way of just consuming all conversations. It's like the meetings like diversity and inclusion meetings multiply somehow. Or it's like the only thing that you're talking about, it's very kind of absurd. And when I look at uh even at M I. T. It's it's a strangely disproportionate amount among the discussions about that. And also to me as an engineer. Those discussions are very frustrating because they don't seem to actually do anything so like they want to bully people instead of creating systems that fix defined like definitive problems and that in itself. That kind of bullying. That's the same kind of thing you saw in terms of mccarthyism in America against the communists. You certainly saw that in soviet union against everybody who's not communist. It's uh it creates hate not progress When you talked to Jordan Peterson recently and people should listen to that conversation. It was a fascinating one I think. Um he he he almost got emotional on the discussion about universities and your experience with Colombia because he like myself for perhaps different reasons. Have a hope for our academic institutions, some of the most incredible people, some of the most incredible engineering and idea development innovation happens in universities. And so we both deeply care about them. Um Is there something, So the reason he got emotional, the reason he was kind of hurt is the fact that you did not, you were not deeply inspired by your experience [/INST] that deeply It made me dumber made me scared. It made me terrified that I had to censor myself in America. Like are you seriously telling me that you don't ever sensor yourself and when you talk you can you truly say whatever you want about race or anything gender. We all censor ourselves. Let's be honest, right, we are all doing that and that's what I learned. Like I thought I was coming to country. They're never needed. Like first thing my mom told me growing up in north Korea was don't even whisper because the birds and mice could hear you. Mhm. And I thought okay now America is truly the land of the free home of the brave. You can say anything you want and then you have freedom to change your mind and evolve, right? But the people now demand you to be the perfect version. They demanded to be. You cannot change your mind and then what is the meaning of life? You cannot grow but you should be safe to talk about anything and then later okay I was wrong the night if you do that, you gotta get penalized for it. </s><s>[INST] I mean since the ship is a funny thing because you probably should not say dumb things. You should try to say things you want to say in the most eloquent. The most effective way you can. So I mean that's what editing is. Right? Yeah. So there's some level of like being careful with what you say, not because you're afraid of some overarching kind of group of bullies but you want to be the best version of yourself and you express stuff. But there's some sense where in the university setting you can put that self censorship like level down more and say stupid stuff and explain and play because you should be forgiven for that kind of uh play. Especially when you're discussing difficult aspects of human history. Whether that that include racism that that include atrocities. I'm still nevertheless sort of hopeful. But at the same time I'm surrounded by engineers so I don't get to interact with people and humanity as much and it seems like there is getting [/INST] worse. </s><s>[INST] I yeah, I don't know. Uh well I do sort of interact with psychologists but they haven't touched on those kinds of topics yet. I still in defense of psychology, I still, I wish we had more numbers but I still feel like most psychology people don't partake in this kind of stuff either. They're just doing excellent research. We're just highlighting this one. America does well you're kind of highlighting anecdotal experiences and uh making a big deal out of them. But that's good because like it's a slippery slope if those, if those things start to overtake all of academia started becoming a big problem even in the engineering field. So we should be concerned. But it is truly tragic that somebody who's exceptionally well read like you who's fire was stoked first with Orwell that fire should burn bright. Like this should not be, you should be writing many books, you know what I mean like and you'll be, you talk to jordans, you know, it's very possible depending what you want to do with your life that you will be a future Jordan Peterson right? Uh So like that and Columbia should be a place that enriches that your mind and the fact that it didn't, this is tragic. [/INST] It's like I was there four years. It wasn't like I had one class that was bad in a one semester. Yeah, that was there when DR pepper was asking is there any one class that had no sentiment of this this purchase signaling particularly right? There was none entire course. I think I took 100 and 26 credits total. Not even one class doesn't matter. We were talking about classic art and that's the thing. I literally thought, okay I push the last does the minister the core like the art and music. Right? So I thought this is gonna be the least politically correct class I can take. And then it begins with who has problem with calling this course the western civilization, Museum of Art and Music. And it was raising their hands because like why do you have to learn about this battle remotes are the bigots and all the people like, you know, everything is going to by white men and there's even music, even these paintings. And as I I didn't raise my hand, everyone's looking at me, how do you not have the problem with us? Like you used to hate was Eur asian Eur So I think that's the thing is I think the problem is very deeper than what people think and it's it's so that's what I when I learned it's like it's not that safe in America. We can go completely to the south and looking at even europe that is like I used to be very more optimism like you know but now I actually see wow this country can go to south and we might if U. S. Forces them right? This is the only country left to battle with the Communist party in china. We may lose the opportunity to be free ever again as a humanity. </s><s>[INST] So I mean that puts a lot of value on having these kinds of conversations. It is I mean I'm troubled, I'm troubled by a lot of things but like censorship on Youtube for example. Yeah it was very annoying to have to listen to donald trump all the time. Like just like create drama like news cycle was completely drowned out by donald trump but like banning him from twitter. It was like Mhm. Mhm. That that was that was scary for me because it's like that's a step towards the direction where you're going to like where does that take us? You're going to silence people then it's like Jordan Peterson is next. [/INST] That's why we need to promote freedom of thinking a speech right. And one thing that I love about Dr Peterson is human, see psychologists right talks about we we we think by talking that's why when you go to a therapist to talk and then you hear yourself and then you think and you come up to the answer, it's so important for humans to talk. So we can think. So when they say you cannot talk means you cannot think and they don't know the consequences of that. And this is why I promote I want the freedom of speech even though it hurts ridiculous. You know, sometimes it can be dangerous, but the price, the alternative is so bad that we should take the, you know, make this trade off everything as a trade off in this world and it comes with the sacrifice. Right? So I think that that's what I'm gonna see in America. But it's unfortunately like the people like you say to who decides what is hate speech, what is dangerous. That's what I've been getting scared because everybody is imperfect. How do we want to give the power to them? And they're going to decide today. They might argue with me said, okay, your speech is good, promotes good and then they might come back next year. Is that your problem must be bad. What are you doing that happens to you? </s><s>[INST] We have to almost like get ideas out and then play with them. I think what's a really important component of that is forgiving each other for like realizing the word different person day by day and certainly years later. Uh, and I think some of that is both cultural mechanisms of saying like we forgive each other for wrong ideas or not wrong ideas but for who we are, the full evolution of the human being for the steps we've taken on that evolution and also creating mechanisms that allow you to um allow us to forgive each other. Like for example on twitter is like horrible with this because one of the main viral ways that people create drama on twitter is like pulling up an old tweet that somebody said right and then saying oh this is the guy that thinks that and but that's like the opposite of the mechanisms, we need to forgive ourselves, forgive each other for the things we've said in the past. And um so part of that is the cultural part of this is as as the the technological um mechanisms you mentioned uh jordans, Jordan Peterson, you had a great conversation with him. What uh what was chatting with him? Like I'm just I'm just curious because he's deeply passionate, especially in the soviet union side about the atrocities of these kinds of systems. What was it like what did you um what did you agree with him on? What did you disagree or some things you both kind of learned from each other to that conversation, Do you think? [/INST] So here is my story, the Jordan Peterson very long one. So one day I was walking down in Chicago and they were like huge theater were sold that he said a big letter, Jordan Peterson saw that and then it was a huge theater in the middle of Chicago right? Like He doesn't like comedian like who can be selling this entire thing up like seven p.m. And then with my ex husband were walking the street and then we saw people were like selling this like tickets like and therefore very higher price right? And then you want to take it. And then he was like yeah sure we're running is packed. And then I was just gave birth or like, but I wasn't able to understand his english that much. My english was served, </s><s>[INST] didn't know who he was really, No, I was just curious. [/INST] Yeah, it was like 2008 </s><s>[INST] guy that sells out a theater. [/INST] Yes I I saw Dave Rubin came up before him and makes jokes. I still don't know who David Rubin is afterwards, I met them all but back then I had no clue what that is and then he was giving lessons but what I got from that night was not what Jordan's said, but what people did on the audience, these people like I'm not thousands of people in this big theater crying like babies and that was like whatever that guy is doing is very special right? He was like making any jokes, he had no slides, just one simple person standing in the huge agitator talk and long time too and people cry, I was like wow okay, whatever that is, I got to take it out and then I got home and then later many years then I got a book and I was started reading his book and it talks about explains so much right? Like now at Columbia learned like everything gender is like made of concept construct like the hierarchy of one man's idea of making the hierarchy and he begins with the number one. The blasters had a hierarchies, evolution of history that is ingredient us that we want to hire key right? And then chapter five about socialization child, you know, how do you raise them and all of it? And then what's why telling the truth is matters right? And there's a white like it's anti 12 lessons. I read it and I was like, I'm, I was so grateful that I'm alive, just like there was people always say if Socrates alive, how much would you pay to have lunch with them? That kinda helped me? Right. So for me it was like, okay, I'm like alive in the same contemporary or one of the greatest fingers of my entire generation and then like how much money exactly right? How much money would I pay? No, no, no limit amount. And I like reach out to Michaela on your pockets on twitter and connected and then one day she would say, do you want to go on my fathers podcast and I was like what as like of course and I was very nervous, but I didn't expect him to be like that connected because I thought he was psychologist, like he saw so much suffering in the world. He studied soviet union, he's hobbies collecting those things to remind him of the suffering of a human being. So sometimes some people here so much atrocity, they become like very, you know, not engaged. </s><s>[INST] He felt he was feeling, he was like, he was living through the experiences with you as you were talking about it. It was an amazing conversation. So Jordan is one of the great thinkers of our time, but I would say the greatest thinkers of our time is Michael malice. You've also got to talk to you. Uh well, so he wrote a book on North Korea. It's an interesting style book. I learned a lot from it, I learned a lot from Michael about it. And it's interesting that he chose North Korea as a, as a thing to study, that he of all people, this fascinating human being that is, Michael chose this darkest of aspects of of humanity to study What uh what do you think of Michael, what do you think of his book on North Korea called? Dear reader, that people should definitely check out. [/INST] Absolutely. So back then when I reach out Michael through a mutual friend in South Korea. I, my english wasn't good. So I got a copy in my hand, I tried to read and a lot of them, I didn't understand. So, but I thought it was very fascinating how he explained north through the dear leader's perspective, right? There's nobody has ever done that. And you can revere so much about the state and absurdity of entire situation and also through humor and that's what's amazing about migrants. He knows full gravity of tragedy. He knows the first offering is not just like people here in America on the buzzfeed, making fun of jim jones haircut. They don't care what people go through. Micro cares </s><s>[INST] deeply cares. And then he still does ridiculous jokes. So that that kind of reveals in a dark way the absurdity of evil. He does that masterfully do you? [/INST] He's a genius. </s><s>[INST] All right. Let's see if he watches this. Uh you know, let's not make his head too big here. But is there some aspect too? I mean there is an absurdity to the whole thing. Uh kim jong un is this? I mean it's almost like a caricature of evil. Like [/INST] joke is a joke? I don't think it's a joke. They just think like this is too too absurd. That is a big laugh. Can you imagine you laugh at holocaust this is that ridiculous? </s><s>[INST] Can you maybe psychoanalyze that a little bit because that that's where my mind goes to. Like he's so ridiculous that you can't, it's almost like hard to believe this is real. Uh is that just is that just my kind of in people's desire to escape the cruelty of reality by just kind of making a joke out of it. [/INST] I think it is a few things right? Like um to North Korea as a nation number one or number two, smartest I. Q. People in the world despite their malnutrition. So </s><s>[INST] so so there is I mean that that was that's an interesting point. So in your sense the people [/INST] are not dumb, </s><s>[INST] still carry the brilliance. There's a there's a culture there that's like hungry too become realized like the people that that were like that that are silenced by the electricity by the actually having no food. All those kinds of things. Like if you add the electricity, if you add the food, you're going to have a cultural center of the world. [/INST] Like South Korea, that's what exactly did write the exact same Korea. One became world like 11th largest economy, one became the world's most like Polish nation, right? And this is a perfect example. Like if I don't know if you read about why nation affairs the system, it's not about a culture, it is not about people, it's not a bike. What makes us two different is a system. South Korea. North Korea is a perfect example of that one is exact same capability. We were homogeneous country, same language tradition, all of that we gave them different system. One is free democracy and a dictatorship and came up with the biggest different result. And I think North Korea reviews that to us. It's not because we are great that we are living in this prosperity free market. The ideas gave us to this the system we built, our ancestor spirit gave us this privilege. It's not us, nothing is about us being special here, right? The system that we have is quite special. And North Korea proves that to us it doesn't matter even if you're smart it that's all irrelevant. And I think that's why people just keep denying that they will not feel special. Like because I'm awesome, I got all of this like no it's not. You you got this and when people say like I hate capitalism is like without capitalism, how do you came up with this thing literally, how did you come up with this? </s><s>[INST] The system's matter and they matter like way more then this individualistic society would like to imagine [/INST] it is the most important thing you can have in life choosing the right system. </s><s>[INST] Do you have a device for young people today? You've lived an incredible life and you have I hope an incredible life ahead of you. What advice would you give to young people today? High schoolers, college students how to um be successful in their career may be successful in life. [/INST] Last thing I want them to feel is guilty. It's it doesn't do anything, right? So I hate when people talk about white guilt, it's like that doesn't make even any sense, right? I think the fact that they bond with freedom is a blessing for all of us. It's not like I want them to want to do something because they are guilty. I want them to do something because they are grateful. It is true. Like we are sitting here at the fact why I have Children is suffering having kids. You don't sleep costly like so much work. Like any like logical rational mind. You should never want Children. Right? Why would you do that to yourself? Especially as a woman? Right? You don't want to do that to yourself. But I think Think about like we are sitting here today, two of us in this amazing technology in this country because somebody in 700,000 years ago, they're hunting berries and surviving cold. Every suffering that you can imagine, they fall for us. That's what we ended up here. So life is ultimately bigger than us plan. I think that's what I want them. It's not like I want them to do the right thing and be the best version of themselves. It's like I want them to feel grateful and we should be grateful for </s><s>[INST] the freedom and then take full advantage of that. I mean it starts with the freedom to experience everything in life [/INST] and for your life literally like you have my father like you know, working, dying is a lot easier than living that tastes like few minutes right next month and leaving takes forever. So when I was facing this unbelievable challenge I thought okay this most rational thing I can just killing myself right now. But the hardest thing I can choose choose to live. And my father did that even in the concentration camp even no matter what he said, life is a gift. You need to fight for it. And I think that's what's missing here that we don't think life as a gift. It's a gift like how many people had to fight for me to be here today. Think about the sacrifice they made for many, many, many generations. I don't even know what they went through. I can't even fathom what they went through. </s><s>[INST] They fought for life. [/INST] Yeah and that is my responsible enough. So it doesn't make them therefore fire was not meaningless. Right? It meant something because now I'm carrying on that fight. </s><s>[INST] You mentioned considering suicide. Do you think about your mortality now now that you're perhaps in slightly more comfortable place you still think about death? [/INST] I do because I was informed actually 21 that I was on the killing list of Kim Jong un in by South Korean intelligence and then and then I had to live with that right? But now I actually feel more because I don't know you follow Jamarcus triggers story. The Saudi journalist got chopped off in Turkey embassy right? His reason why he got killed it was he became very prominent on twitter. He had a huge voice and Saudis followed him now I became very first North korean to have this many social media followings and recently north Korea start investigation investigation team to analyze whatever I do even though it's first time for them so they don't even know what at this point like they're like this is so new. What do we do with kim jong un kim jong un the half brother kim jong un got killed in Malaysia. That is a lot of tragedy that I feel so sorry for the U. S. Government is that kim jong un was giving information to the CIA for the past like 10 years that trip when he got killed demolition airport he was meeting up with the CIA agent for two days on the northern Ireland. See I could have protected them. They didn't they let him die and killed them. North korean kim jong un care. Do you know the Malaysian that the ladies that the VX nerve asian north korean security in Malaysian air in the international land. So I mean Jamarcus sticky who was used to present and the Washington post journalist when he got killed in Saudi like a lamb. They chopped them into pieces in that most inhuman death. What was the consequences for the Saudis nothing the word is we think reliving injustice country now there is no justice, there is no accountability for killing any distance, no matter how big their names are. </s><s>[INST] So you don't think you're vast and quickly growing social media presence protects you [/INST] know it does the opposite because kim jong un injured and I spoke out, I don't know if you went through it. They did everything they could to character assassinate me saying I'm a liar, I'm a C. I a spy, I get paid. And then they reached out the penguin saying we're gonna blow up, you cannot write this book. And they did with the Sony, they had a son in studio for making that stupid movie interview. Right? And then penguin did their investigation. They met every survivor that I went through in the desert. They got the voice recordings of them because they don't want them to change their mind later. Right? People remember differently. So they got the voice recordings like the penguin regarding, got the old audience and now we are ready for the lawsuit. We're going to publish this book because we checked and verified every single thing that was going in the book and north Korea couldn't do anything anymore. </s><s>[INST] But that's character assassination by the way, that's a whole other conversation that you were able to survive. That I appreciate the kind of strength it requires to survive that because you don't know and your character being assassinated is in some ways can be as painful, has actual [/INST] assassination. It's the worst. Everybody thinks you're a liar, like everybody thinks you're a liar and now everybody like you said this nature of internet that as long as something is written internet, they think that's a fact and the stupid person can start a blog and write about you. But they think, oh because it's written on the internet is a legit, </s><s>[INST] especially negative stuff. That's the thing I was kind of trying to elaborate on. There's a viral aspect to calling somebody a fraud or a liar that nobody questions whether it's true or not, it just spreads, it's a, it's a dark side of our human nature that we want to destroy the people who are rising. [/INST] We cannot stand it. Yeah. Any change maker in this world who wasn't controversial multi to continue like nelson Mandela, he was called as a terrorist. Right? So I just did not know </s><s>[INST] the character assassination is the thing, you know, probably continue with [/INST] you forever. </s><s>[INST] You have to get stronger and stronger I think in the face of that. But actual assassination and perhaps it's me being hopeful because I have a situation with Russia that I hope I'm not under. Well, I don't care actually, but there's some aspect in which social media presence I thought protects you a little bit because just imagine the outrage uh, from an attempted assassination assassination of you. [/INST] What was outraged when johnny cash you got killed like that </s><s>[INST] was the social media presence large [/INST] over one million people, I don't have that problem. He was 1.6 million to their followers </s><s>[INST] and the outrage wasn't there? [/INST] No because Saudis spoke to amazon prime studio netflix. There were people made a documentary body but told everybody don't cannot get that deal. So there was a huge censorship on that. And people of course like I mean they can talk about it one day, some distance from Saudi got cared horrible. They move on the next cute puppy right then execute cat like that's what the nature of this new generation does. They desensitized, It doesn't affect them. They keep following this instant pleasure instant high. That's what Instagram does to you. It changes your brain like that. So I was reading this book the shallows. We became shallow and shallow and our brain changed permanently. So this is a generation, we can get them angry for like 10 minutes create hashtags for one day. But then as quick as that was, it goes down like instantly. And I think that's the, </s><s>[INST] well that means okay, so that means that there is uh, it's an effective way to get rid of opposition is by murdering them. And that means a United States. If it stands for freedom, if it stands for the freedom of exchange of ideas, should be protecting people like you. [/INST] But they don't, they be involved. They didn't even protect kim jong un who is giving information 10 years. He was in his life. That's what is so I mean working for CIA is not bad and I I don't I hope, I mean the thing is he was giving information to bring down the vision that is valuable. That is something noble about him. But then you don't go extra miles to that. That's when I lost my faith in the U. S. System as well. Like this country just cares about saving face what is most minimum cause they pay for anything. And I went to South Korea constantly every single day intelligence calling me. You like the north korean agent going this place, where are you going to the U. S. S. Came to us? Nobody this is when people say that Usa asian like I wish they called me. I really truly do. But nobody nobody does here. I'm sure they know what's going on. But the south korean agent is more like oh my gosh, we don't want you to get killed. A south korean citizen right? And now I'm trying to become a citizen. So it's in a way it's I don't know what's worse Are </s><s>[INST] you are you afraid for your life? [/INST] I was afraid for those ever 3, 4 years I was afraid and he was but I had it came terms really like in my enemy is not some crazy psychopath is a state with a nuclear power to attack the most powerful country. If kim jong un decides if I die I'm gonna die. It's not up to me right? So in a way or just liberating that you is that if you are like afraid of some maps or some like Gangsters on the street, it's almost like you have power over a little bit. You gotta be like thinking that's my fault. I went that way right? But when it comes to kim jong un I know like my enemy so much bigger than me, it's in a ways of liberation and also it was I just I live a lot. So I have seen a lot, I've seen everything. I don't have that much younger. I left here like okay I'm going to soon. You know it's like okay maybe it's time like that is a part of life. So </s><s>[INST] in some sense you're willing to accept death to keep fighting for freedom in your and at least in part to place you call home. [/INST] Yeah. Do </s><s>[INST] you hope that one day you can return to north Korea? [/INST] I hope so. I hope I bring my son and time. This is like where your ancestors from two, </s><s>[INST] it would look very different. Then the place you came from in your as you hope, do you hope that there's a democracy one day that North Korea looks like South Korea. Yeah [/INST] what they're being paradise right? That's it. But I'm a rational optimist. I'm not, I just optimistic because I have to be I think as long as there are people who have changed the word right? Like who believed in something and worked for it and like, I don't know like this. So like Alice Rose a few people holding entire list word, right? I really believe in that. I think as long as that continues, that can happen in my country, as long as people like you someday when I decided to do something on north korean, working for using your brain power to solve this puzzle. How fascinating would that be. That's why I continue to speak content recruit </s><s>[INST] to inspire millions to do something. The books you like are all the books I love. So, I have to mention this, you mentioned briefly on the with Jordan's uh Siddhartha by Hermann Hesse is uh is an incredible book there. I mean, I don't I don't know exactly what I want to ask here, but there's some, I think the book kind of through telling a story reveals that life is suffering and yet there's beauty in it. The beauty in every moment that that uses a kind of a river painting metaphor. Is there is there um is there something that you could say speak to like how that book impacted your life and the way you live life maybe the way you see life whether it's on the life is suffering side or that life is beautiful side? [/INST] I mean, he goes to the entire journey, right? He goes in the state like I'm so enlightened that I cannot deal with the people there in love and quite about it right there like that? So like primitive. Once he has his own son he actually being attached? He actually cares. It actually really does hold thing, right. That's the thing that he used to think not once his son comes find him, he looks at life differently. I think that's the thing. I did have that kind of journey where or nothing matters, right? So bitter, so, so like, so cynical and after I met so many incredible people, I was talking about that person who told him he was gay, he told me I love you. And I was like, why do you love me in the past? People when they wanted me because they want to rape me. Everybody wanted something from me. That's why they wanted me and I never understood you can love somebody unconditionally. And this gay guy, the last one was one of sleep with me, right? And he loves me. And I think I had a blessing after my journey meeting people who loved me unconditionally because I was just being a human. And I think that's what it is now for me that like him, I live for love now. I live for love any kinds of love, Love for knowledge. I like I read so many books because I love books right? I love what I do. I love my people. I love humanity. You know, even it sometimes annoys me. I love myself </s><s>[INST] and that's beautiful to the annoying parts are beautiful too. What do you let me ask the ridiculous question. What do you think is the meaning of this whole thing of what's the meaning of life? [/INST] Yeah. I think at this point I stopped question why I'm here, right? Like it doesn't matter. Someone put the atom, there are a big thing. I'm here. That's truth. Right? I'm going to accept that fully. So what? Instead of me keep asking the impossible question, what I'm here? I'm gonna let you do that. Let's do that. Right? You guys go out into space and look for the evidence I'm conducting </s><s>[INST] you Except that you're here and you're just going to enjoy your here for love. As you said, [/INST] That's the thing. I think I'm here for the process of pursuing something bigger than me. Process of doing something. It's not like more, It's not a virtue signaling thing. It just makes me happy that I fight for something bigger like that. Me, right? So how boring is that? Every day you get up? Oh my God, I'm gonna buy myself this. I'm going to get this for myself is so boring, isn't it? So in a way, I think that's what it is. I am grateful that I'm in a state. I don't have to fly for myself anymore. But more than people have to do that. And that's sometimes more than enough they have to do and I salute them. They are doing? Fighting, saving themselves every day. But now I'm I'm not there, I'm very blessed. That's why I'm very grateful. </s><s>[INST] So fighting for something much bigger than you. But do you still believe that you can change the world that you can be a thing that at least in part helps North Korea or even broader helps alleviate some suffering in the world. [/INST] So that's the thing. I was reading this book for the by randomness, right? I was like, they were here like, oh my God, you're so courageous, you're amazing. I'm like no I'm not, I'm horrible. Right? I know myself. You don't want to tell me that. It's a random why I end up here. Like why did I pick up in Greece so quickly? Why do I love books? Right. I don't know why it's random. </s><s>[INST] Why just enjoy [/INST] it. Yeah, it's random. I think, I don't know how the history will remember me. I think the only thing I have to disappoint to make sure that people after consulting a lot of security teams like nah, North Korea became a lot smarter. Like you said, they may get more disguised as a suicide and I call accent. So when I die, they don't even know I got scared and I think that's a higher chance. So I think that's the thing like people are suffering, take it or not is your choice and at least it's my responsibility for them to know what's going on. I think if you did not know and didn't do anything. You're not even beautiful thing. But once you know, then you're not doing it. Then you saw this like not right? So that's what I'm doing. Like I want people to know and then what they're gonna do is not my, my problem afterwards. Right? So my role is very small in that regard. And I just hope that or humanize north Koreans for the first time because we have been so dehumanized, right? Like we are like looking like roberts, if you look at us marching and quite like when you later dies almost seems like we don't even have the same emotions. People cannot connect us in the same level. Okay. And I think that's something is uh that's something media have done it to us </s><s>[INST] and you're you're shining a small light on this dark part of the world that I think and you make it, you're so modest. But I think I think you will have that little light just might be a big thing that changes that incredible amount of suffering that's happening on that part of the world. You know, I mean, you're you're an amazing person. Um, so fortunate to get a chance to talk with you. I can't wait what you do in the future. You're I hope you write many more books. I I do hope you continue making videos continue having conversations. You're an inspiration to me and millions of others. I really appreciate you. Talk with me today. [/INST] I'm sorry. Thank </s><s>[INST] thank you. Thank you for listening to this conversation with you and me. Park and thank you to belcampo gala games. Better help and aid sleep, check them out in the description to support this podcast and now let me leave you some words from bob Marley, Better to die fighting for freedom and be a prisoner all the days of your life. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] So you said a few things there one is IOS why IOS why? Which operating system? Which one is the sexiest thing to try to get to? Or the most impactful thing? And the other thing you mentioned is remote versus like having to actually come in physical contact with it. Is that the distinction? [/INST] So Iphone exploits have just been a government's number one priority recently. Actually the price of an android remote zero day exploit, something that can get you into android phones is actually higher. The value of that is now higher on this underground market for zero day exploits than an iphone IOS exploit. So things are changing. </s><s>[INST] There's probably more android devices. So that's why it's better. But then the iphone side, if I, so I'm an android person because I'm a man of the people, but it seems like all the elites use iphone all the people at nice dinner parties. So uh, is that is that the reason that like the more powerful people use iphones, is that why [/INST] I don't think so. I actually, so it was about two years ago that the prices flipped? It used to be that if you could craft remote zero click exploit for IOS, then that was about as good as it gets. You could sell that To a zero day broker for $2 million. The caveat is, you can never tell anyone about it Because the minute you tell someone about it, Apple learns about it, they patch it in that $2.5 million dollar investment that that zero day broker just made goes to dust. So a couple of years ago and don't quote me on the prices, but an android, zero click remote exploit for the first time topped the IOS. And actually a lot of people's read on that was that, that it might be a sign that Apple security was falling and that it might actually be easier to find an IOS zero day exploit than find an android zero day exploit. The other thing is market share. There are just more people around the world that use android and a lot of governments that are paying top dollar for zero day exploits these days, Our deep pocketed governments in the gulf that want to use these exploits to monitor their own citizens monitor their critics. And so it's not necessarily that they're trying to find elites. Uh, it's that they want to find out who these people are that are criticizing them or perhaps planning the next arab spring? </s><s>[INST] So, in your experience, are most of these attacks targeted to cover a large population or is there attacks that are targeted towards specific individuals? [/INST] So I think it's both some of the zero day exploits that have fetch top dollar That I've heard of in my reporting in the United States were highly targeted. You know, there was a potential terrorist attack. They wanted to get into this person's phone. It had to be done. In the next 24 hours, they approached hackers and say, we'll pay you x millions of dollars if you can do this. But then you look at when we have discovered IOS zero day exploits in the wild, some of them have been targeting large populations like workers. So a couple of years ago there was a a watering hole attack. Okay, it's a watering hole attack. There's a website, it was actually had information aimed at wahlburgers and you could access it all over the world. And if you visited this website, it would drop an IOS zero to exploit onto your phone. And so anyone that visited this website that was about the tiger's anywhere, I mean, tigers, tigers living abroad, basically the weaker diaspora would have gotten infected with this zero day exploits. So in that case, you know, they were targeting huge swaths of this one population or people interested in this one population basically in real time? </s><s>[INST] Who are these Attackers from the individual level to the group level? Psychologically speaking. What's their motivation? Is it purely money? Is it the challenge? Are they malevolent? Is it power? These are big philosophical human questions, I guess. [/INST] So. These are the questions I set out to answer for my book. I wanted to know, are these people that are just after money? You know, if they're just after money, how do they sleep at night not knowing whether that zero day exploit they just sold to a broker is being used to basically make someone's life a living hell. And what I found was there's kind of this long sordid history to this question. You know, it started out In the 80s and 90s when hackers, we're just finding holes and bugs and software for curiosity's sake really as a hobby. And some of them would go to the tech companies like Microsoft or Sun Microsystems at the time Oracle. And they'd say, hey, I just found this zero day in your software and I can use it to break into Nasa. And the general response at the time wasn't thank you so much for pointing out this flaw and our software will get it fixed as soon as possible. It was, don't ever poke around our software ever again or we'll stick our general counsel on you and that was really sort of the common thread for years. And so hackers who set out to do the right thing were basically told to shut up and stop doing what you're doing. And what happened next was they basically started trading this information online now when you go back and interview people from those early days, they all tell a very similar story, which is, they're curious, they're tinkers, you know, they remind me of like, the kid down the block that was constantly poking around the hood of his dad's car. You know, they just couldn't help themselves. They wanted to figure out how a system is designed and how they could potentially exploit it for some other purpose. It doesn't have to be good or bad, but they were basically kind of beat down for so long by these big tech companies that they started just silently trading them with other hackers. And that's how you got these Um really heated debates in the 90s about disclosure, should you just dump these things online because any script kiddie can pick them up and use it for all kinds of mischief, um, but you know, don't you want to just stick a middle finger to all these companies that are basically threatening you all the time. So there was this really interesting dynamic at play and what I learned in the course of doing my book was that government agencies and their contractors, sort of tapped into that frustration and that resentment. And they started quietly reaching out to hackers on these forums and they said, hey, you know that zero day you just dropped online. Could you, could you come up with something custom for me and I'll pay you six figures for it so long as you shut up and never tell anyone that we, that I paid you for this and that's what happened. So throughout the nineties there was a bunch of boutique contractors that started reaching out to hackers on these forums and saying, hey, I'll pay you six figures for that bug you were trying to get Microsoft to fix for free. And so began or so catalyzed this market where governments and their intermediaries started reaching out to these hackers and buying their bugs for free. And in those early days, I think a lot of it was just for quiet counterintelligence, traditional espionage. But as we started baking the software, Windows software Schneider Electric's demons, industrial software into our nuclear plants and our factories and our power grid and our petrochemical facilities and our pipelines. Those same zero days came to be just as valuable for sabotage and war planning. </s>

<s>[INST] So let me linger on this. So you were actually, uh, there's some people might not be even familiar with discord or slack and so on. Even these platforms have like communities associated with them. Meaning the, the big, like the meta community of people who are aware of the feature set and that you can do a thing that this is a thing. And then you could do a thing with it. Discord, like when I first realized that I think it was born out of the gaming world is like, holy shit, this is like a thing. There's a lot of people that use this. There's also a culture that's very difficult to escape that's associated with discord that spans all the different communities within discord. Reddit is the same, even though there's different subreddits, they're still because of the migration phenomenon, maybe there's still a culture to Reddit and so on. Yes. So I I'd like to sort of try to dig in and understand what's the difference between the online communities that are formed and the platforms on which those communities have formed. Sure. An important. [/INST] Yes, yes, yes, it is. It is. So for example, an office, a good design for an office is frequently you have, um, you know, the, uh, the commons, which is like the lunch room or the gathering area, then everybody else has a cave on the border that they can kind of retreat to. </s><s>[INST] Cave in the commons. I love, by the way, I was laughing internally about the heat versus work. I think that's going to stick with me. That's such an interesting way to see Twitter. Yeah. Like, is this heat or is this, is this thread? Like, cause there's a lot of stuff going on. Right. Is it just heat or are we doing some like, is it, is there a directed thing that's going to be productive at the end of the day? That's right. I love this. I'll never see it manually. The cave in the commons is, uh, is really nice. [/INST] Uh, so that has to do with the layout of an office that's going to be productive at the end of the day manually. That's effective. That's right. And, uh, so you can think of many kinds of social networks, um, as being on the cave in commons continuum, for example, Twitter is just all commons. The caves are just like individual DMS or DM threads or whatever, but it's really basically just one gigantic global public fight club for the most part. Right. And then you have four love club. </s><s>[INST] Well, some would love the mostly fight or actually it's, I love aggressively. That's all. Yeah. [/INST] I mean, the way I think, I mean, Twitter is like a cross between, uh, you know, a, a library and a civil war, you know, it's something where you can learn. And, but, but you can also fight if you choose to fight. Right. </s><s>[INST] Um, yeah. Well, I mean, it's, um, uh, because of it's the commons structure of it, it's a mechanism for virality of anything you just described the kind of things that [/INST] become viral, meaning no, no offense to librarians. It's like a library and Liberia, Liberia was racked by civil war for many years. Right. </s><s>[INST] It's a library. Just one of my favorite sets, uh, for porn. Just kidding jokes. I'm learning as that's probably crossing the line, uh, for the engineers working on this humor module, maybe take that down a notch for porn. [/INST] Yeah. Um, gosh, uh, we're just talking about, oh yeah. So continue. I'm going to continue. Right. Commons. Yeah. So Twitter is a commons. Then Facebook is like, it's got all these warrens and stuff. Facebook is very difficult to reason about, uh, like privacy on that. And the reason is I think it's easy to understand when something is completely public like Twitter or completely private like signal. And those are the only two modes I think in which one can really operate. When something is quasi private, like Facebook, you have to just kind of assume it's public because if, if it's interesting enough, it'll go outside your friend network and it'll get screenshotted or whatever and posted. And so, you know, Facebook is sort of sort of forced into default public despite its privacy settings, you know, for anybody who says something interesting, you know, if it's like, uh, you can figure out all their dials and stuff like that, but just hard to understand unless it's totally private or totally public, right? You have to basically treat it if it's totally public, if it's not totally private, okay, at least under a real name. I'll come back to soon. So you've got Twitter. That's, that's total commons. Facebook is like a Warrens, you know, like, you know, it's like rabbit Warrens or like an ant colony where you don't know where information is traveling. Then you've got Reddit, which has serve your global Reddit and then all the subreddits, that's a different model of cave in commons. </s><s>[INST] Oh yeah, so continue [/INST] yeah, I think one of the reasons it works is that you have individual moderators where something is totally off topic and acceptable, totally off topic, and unacceptable in this subreddit and totally on topic and acceptable another, that's like kind of a, you know, a precursor of the digital societies, I think that we're going to see that actually are become physical societies, like lots and lots of subreddit like things, you know, have become physical societies. Then you start going further into like discord where it's, it's more full featured than, um, you know, as you go Twitter, Facebook, Reddit, now you jump into discord and discord is a bunch of individual communities that are connected and you can easily sort of jump between them, right? And then you have slack and, you know, yes, you can use, um, slack to go between different companies, slacks, but slack historically at least, I'm not sure whether their current policy is historically, they discourage public slacks. So there it's mostly like you have your main slack for your company. And then you sometimes may jump into like, uh, let's say you've got a design consultant or something like that, you'll jump into their slack, but. Discord is, you've got way more discords usually that you jump into than then slack's. Right. </s>

<s>[INST] It's fascinating to try to such a cool economically school in that sense because you're trying to predict something that hasn't been done before, the impact, the effects of something that hasn't been done before. [/INST] Yeah. You're trying you're going out of sample and we have good reason to believe that that there are, you know, that it's non linear that there comes a point at which it doesn't work the way it has uh in the past. </s><s>[INST] So as an economist, how do you see science and technological innovation? When I took various economics courses in college, technological innovation seemed like a no brainer way of growing economy and we should invest in it aggressively. Yeah, I may be biased, but it seemed like the various ways to grow an economy. It seems like the easiest way, especially long term. Is that correct? And it's so why aren't we doing it more? [/INST] Well, that's ok. The first question is, yeah, I mean all it's pretty much overwhelming. We think we can more or less measure this, although there are some assumption involved, but it's something like 70-80% of the growth and per capita income is is basically the advance of knowledge. It's not just, it's not just a crude accumulation of capital. It is, it is the fact that we get, get smarter a lot of that by the way, is more prosaic kinds of technology. So, you know, we uh, I I like to talk about things like container ization or, you know, an earlier period, the uh, the invention of the flat packed cardboard box that had to be invented. And uh and now all of your deliveries from amazon are made possible by the existence of that technology that the web stuff is important too. But but what would we do without cardboard boxes? So, but all of that stuff is really important in driving economic progress. Well, why don't we invest more uh why don't we invest more uh, in uh again, more prosaic stuff, why aren't, why haven't we built another goddamn real tunnel under the Hudson River, which is for which the need is so totally overwhelmingly obvious. How do you </s><s>[INST] think about, first of all, I don't even know what the word prosaic means, but I inferred it. But how do you think about prosaic, cause it uh the really most basic dumb technology innovation or is it just like the lowest hanging fruit of where benefit can begin? [/INST] Uh, when I say prosaic, I mean stuff that is not sexy and, and fancy and high tech. It's uh building bridges and and tunnels, having uh, inventing the cardboard box or um uh the, I don't know where where do we put uh uh EZ pass in there. Right. That's a uh it is it is actually using some modern technology and all that, but it's not going to have, I don't think you're going to make a movie about uh, about the fact that the guy, whoever it was that invented Ez Pass. But uh, but it's actually a pretty significant productivity booster. </s><s>[INST] To me, it always seemed like it's something that everybody should be able to agree on and just invest. So like in the same way there's the investment in the military and um the D. O. D. Is huge. So everyone kind of, not everyone, but there's a there's a there's an agreement amongst people that somehow that a large defense is important. It always seemed to me like that should be shifted towards. If you want to grow prosperity of the nation, you should be investing in knowledge. Yes. Prosaic stuff. Infrastructure, investing in infrastructure and so on. I mean, sorry to linger on it. But do you have any intuition, Do you have a hope that that changes? Do you have intuition why it's not changing it? I'm [/INST] more than intuition. I have a theory. I'm reasonably certain that I understand why why we don't do it. Uh, and it's it's because um because we have a real values dispute about the welfare state about how much the government should do to help the unfortunate and politicians believe, uh, probably rightly that there's a kind of halo effect that surrounds any kind of government intervention, that even </s><s>[INST] though [/INST] providing people with enhance Social Security benefits is really very different from building a tunnel under the Hudson river. Uh, politicians of both parties seem to believe that the government is seen to be successful at doing one kind of thing. It will make people think more favorably on and doing other kinds of things. And so we have conservatives tend to be opposed to any kind of increase in government spending except military. Uh, no matter how, uh, obviously, a good idea it is because they fear that it's the thin end of the wedge for bigger government in general. Uh, and to some extent, liberals tend to favor spending on these things. Uh, partly because they see it as a way of proving that government can do things well, and therefore it can turn to broader social goals. It's clearly, there's a uh, if you like, the, the what you might have thought would be a technocratic discussion about government investment, both in research and in infrastructure, um, is contaminated by the fact that government is government and people link it to other government actions. </s><s>[INST] Perhaps a silly question, but as a species, we're currently working on venturing out into space one day colonizing mars. So when we started society on mars from scratch what political and economic system should operate under. [/INST] Oh, I'm a big believer in first of all, I don't think we're actually gonna do that. But </s><s>[INST] let's uh let's imagine [/INST] hypothesize that we colonize MArs or something. Um look, representative democracy is </s><s>[INST] ah versus pure democracy. [/INST] Well, yeah, prepare democracy where people vote directly on everything is really problematic because people don't have time to uh to uh to try and master every issue. I mean, we can see what government by referendum looks like. There's a lot of that in in California and it doesn't work so good because it's hard to explain to people the various things they vote for may conflict. So representative democracy uh is uh it's got lots of problems. Um and I kind of the Winston Churchill thing, right? It's the worst system, you know, except for all the others. </s>

<s>[INST] tell you about the nice coffee shops and [/INST] well when coffee shops it was C. D. U. C. D. Places. But yeah it was that kind of thing, nice shops. Oh there's all these students here, Harvard Square is beautiful, you can do all these things and you can walk and something about the outdoors, whichever is the slightest bit interested. The outdoors is for the bugs, not for humans. Um And the that should be a t shirt. I mean that's the way I feel about it. Um The and the black folk told me completely different stories about which part of town you did not want to be caught in after dark and um and I heard of but that was nothing new. So I decided that it was a great place to be as a university and I believed it then, I believe it now. Uh and that whatever it is I wanted to do, I thought I knew what I wanted to do, but what if I was wrong, um someone there, we know how to do it. Of course, then I would pick the one topic nobody was working on at the time, but that's okay. Um it was great and so I thought that I would be fine and not only be there for like four or five years, I told myself it turned out not to be true at all, but I enjoyed my time. I enjoyed my time there, but I did see a lot of um I ran across a lot of things that were driven by what I look like while I was here, I got asked a lot of questions. Uh I ran into a lot of cops. Uh I did a I I saw a lot about the city, but at the time, I mean I've been here a long time. These are the things that I remember. So this is 1990 there is not a single black radio station Now, this is 1990. There are, I don't know if there's any radio stations anymore. I'm sure there are, but you know, I don't listen to the radio anymore and almost no one does. Um at least if you're under a certain age, but the idea is you could be in a major metropolitan area and there wasn't a single black radio station, by which, I mean a radio station that played what we would call black music then um was absurd but somehow captured kind of everything about about the city. I grew up in Atlanta and you know, you've heard me tell you about Atlanta boston had no economically viable or socially cohesive black middle class. Mhm insofar as it existed, it was uniformly distributed throughout large parts of not all parts with large parts of the city. And we had concentrated concentrations of black Bostonians. Uh They tended to be poor. It was very different from where I grew up. I grew up on the poor side of town. Sure. But then in high school, well in ninth grade we didn't have middle school, I went to an eighth grade school where there was a lot of, let's just say we had a riot the year that I was there, there was at least one major fight every week. It was uh it was it was uh it was an amazing, it was amazing experience. But when I went to ninth grade I went to the academy, Math, math and science academy. Makes high was a public school. It was a magnet school. That's why I was able to go there. Uh It was the first school high school I think in the state of Georgia. To sweep the state math and science fairs. It was great. It had 385 students, all but four of whom were black. I went to school with the daughter of the former mayor of Atlanta, Michael Jackson's cousin. I mean, you know, there was, it was an upper middle class, whatever. You know, I just dropped names occasionally, you know, drop the mic, drop some names just to let you know, I used to hang out with Michael Jackson's cousin, 12 cousin, nine times removed. I don't know. The point is that they had, we had a parking problem because the kids have cars. I did not come from a place where you had cars. I got my first car when I came to my t actually. Um, so it was, it was just a, it was just a very, very different experience for me. But I've been to places where whether you were rich or whether you were poor, you know, you could be black and richard black and poor and it was there and there were places and they were segregated my class as well as by race. But um, that existed here at least when I was here didn't feel that way at all. And it felt like a bunch of a really interesting contradiction. It felt like it was the interracial dating capital of the country. Yeah, you really felt that way. Uh, but it also felt like the most racist place. I have respect any time. Uh, You know, you couldn't go up the orange line at that time. I mean, again, that was 30 years ago, I don't know what it's like now, but there were places you couldn't go and you knew it, everybody knew it. And there were places you couldn't live and everybody knew that. uh, and that was just the Greater Boston area in 1992, </s><s>[INST] subtle racism or explicit racism, both in terms of within the institutions. Did you feel was it was their levels in which you were empowered to be first to water in the first black people in a particular discipline and in some of these great institutions that you're part of, you know, Georgia tech MIT. And was there part where it was felt limiting? [/INST] I always felt empowered. Some of that was my own delusion, I think. But it worked out. So I never felt is in fact quite the opposite. Not only did I not feel as if no one was trying to stop me, I had the distinct impression that people wanted me to succeed by people. I meant the people in power. Now my fellow students not, they didn't want me to succeed, but I felt supported or at least that people, we're happy to see me succeed at least as much as anyone else. But, you know, 1990, you're dealing with a different set of problems which you're you're very early, at least in computer science, you're very early in the sort of Jackie Robinson period, you know, there's this thing called the Jackie Robinson syndrome, which is that you you have to, you know, the first one has to be perfect or has to be sure to succeed because of that person fails. No one else comes after for a long time. So, um you know, it was kind of in everyone's best interest, but I think it came from a sincere place, I'm completely sure that people went out of their way to try to make certain that the environment would be good, not just for me, but for the other people who of course were around. And I was hardly the only, I was the only person in the I lab, but I wasn't the only the only person that mighty by a long shot on the other hand, but what at that point we would have been less than 20 years away from the first black PhD to graduate from M. I. T. Right, Shirley Jackson, right, something like that somewhere around then. Um, so we weren't that far away from the 1st 1st and we were still another eight years away from the first black PhD Computer Science. Right? So, um we were in a it was it was a sort of interesting time, but I did not feel as if the institutions of the university where against any of that and furthermore, I felt as if there was enough of a critical mass across the institute from students and probably faculty, that I didn't know them, who wanted to make certain that the right thing happened is very different from the institutions of the rest of the city, which I think were designed in such a way that, uh, they felt no need to be supportive. </s>

<s>[INST] bye bye to them. Which college did you go through? [/INST] So I went to Yale. Uh Princeton would have been way more convenient and it was just beautiful campus and it was close enough to home and I was really excited about Princeton. And I visited, I said, so computer science major. Like, well we have computer engineering. Like, oh, I don't like that word engineering. I like if your science, I really want to do, like you're saying hardware and software, they're like, yeah, like I just want to do software. I couldn't care less </s><s>[INST] about hardware. You grew up in philadelphia. [/INST] I grew up outside Philly. Yeah. Yeah. So the local schools were like penn and Drexel. And uh, Temple, like, everyone in my family went to Temple at least at one point in their lives except for me. So yeah, Philly Philly family, </s><s>[INST] Yale had a computer science department and that's when you, It's kind of interesting. You said 80s and neural networks. That's when you neural networks is the hot new thing or a hot thing period. Uh so what is that in college when you first learn about neural networks? When she learned like [/INST] it was in a psychology class. Not in a wow. </s><s>[INST] Was the psychology of cognitive science or like do you remember like what context [/INST] it was. Yeah. Yeah. Yeah. So so I was a I've always been a bit of a cognitive psychology groupie. So like I I studied computer science but I like I like to hang around where the cognitive scientists are because I don't know brains man, they're like </s><s>[INST] they're wacky cool and they have a bigger picture view of things. They're a little less engineering. I would say they're more they're more interested in the nature of cognition and intelligence and perception and how the division system or that they're asking always bigger questions. Now with the deep learning community there I think more there's a lot of intersections but I do find that the neuroscience folks actually in uh cognitive psychology, cognitive science folks are starting to learn how to program how to use your own artificial neural networks and they are actually approaching problems in like totally new interesting ways. It's fun to watch that grad students from those departments like approached the problem of machine learning. [/INST] They come in with a different perspective. </s><s>[INST] They don't care about like your imaging that data set or whatever they want like to understand the like the basic mechanisms at the at the neuronal level at the functional level of intelligence. It's kind of it's gonna go cool cool to see them work. But yeah, okay, so you always love you were always a group of cognitive psychology. [/INST] Yeah. Yeah. And so so it was in a class by Richard Garrett. He was kind of my my favorite uh psych professor in college and I took like three different classes with him. And yeah, so they were they were talking specifically the class I think was kind of a there was a big paper that was written by steven, Pinker and Prince, I'm blanking on Prince's first name but Prince and Pinker and Prince they wrote kind of a they were at that time kind of like I'm blanking on the name of the current people. Um the cognitive scientists who are complaining a lot about deep networks. Oh, Gary Gary </s><s>[INST] Marcus, sorry Marcus and who else? I mean there's a few but Gary Gary is the most feisty. [/INST] Sure. Gary is very feisty. And with this, with his co author, they you know, they're kind of doing these kind of take downs where they say, okay, well yeah, it does all these amazing, amazing things. But here's a shortcoming, Here's a shortcoming. Here's your shortcoming and so the Pinker Prince paper is kind of like the that generation's version of Marcus and Davis right where there there trained as cognitive scientists but they're looking skeptically at the results in the in the artificial intelligence, neural net kind of world and saying, yeah it can do this and this and this but it can't do that and it can't do that and it can't do that maybe in principle or maybe just in practice at this point. But the fact that matter is your you've narrowed your focus too far to be impressed, you know, you're impressed with the things within that circle, but you need to broaden that circle a little bit, you need to look at a wider set of problems. And so so we have so I was in this seminar in college that was basically a close reading of the pinker prints paper which was like really thick, there's a lot going on in there and um and and and it talked about the reinforcement learning idea a little bit. I'm like oh that sounds really cool because behavior is what is really interesting to me about psychology anyway. So making programs that, I mean programs are things that behave people are things that behave like I want to make learning that learns to behave </s><s>[INST] in which way was reinforcement learning presented. Is this talking about human and animal behavior or we're talking about actual mathematical construct. [/INST] Right? So that's a good question. Right? So this is I think it wasn't actually talked about as behavior in the paper that I was reading, I think that it just talked about learning and to me learning is about learning to behave but really neural nets at that point we're about learning, supervised learning. So learning to produce outputs from inputs. So I kind of tried to invent reinforcement learning. I when I graduated I joined a research group at Belk or which had spun out of Bell Labs recently at that time because of the divestiture of the long distance and local phone service in the 1980s, Uh and I was in a group with Dave Ackley who was the first author of the bolts machine paper. So the very first neural net paper that could handle X or right, so X or sort of killed neural nets the very </s>

<s>[INST] So you quote Finnegans Wake uh mad problems for that James Joyce book. I took a class in James Joyce in college, I think I read parts of fitting in a way it might have been on drugs or some kind or I um I somehow got an A in that class which probably refers to some kind of curve where nobody understood anything. The only thing I understood and really enjoyed is his short stories, The Dead and then ulysses, I kind of uh I think Greta few cliff notes that kind of got to the point and then Finnegans Wake was just a hopeless person. It's a you for people who haven't looked at it, maybe you can elucidate to me better. But I felt like I was reading things, words and the words made sense, like standing next to each other. But when you kind of read for a while you realize you didn't actually understand anything that was said. [/INST] But did you have a feeling though, that's one that's one thing I found interesting about Finnegans Wake, I never fully understood it, but the words caused feelings in me, which I found fascinating and and sometimes I couldn't predict it from the semantic, you know, black and white context of what I was seeing in front of me on the page. But the rhythm of the melody would would make me feel certain ways. And that's that was what I always was intrigued by with Joyce of course that was his, he he existed on a spectrum too and he wrote as you say, you know, more accessible works. You know, I learned a lot about irish history from, you know, portrait of the artist as a young man and there was just so he could be as objective as he wanted to be. But then when he let himself loose, he was in this realm where the words had their own purpose separate from from you know semantic meaning, from their, their dry dictionary definition. </s><s>[INST] You know, there's a funny story that was told, it doesn't matter if it's true or not, but they they said that James Joyce when he was young, when he was in his teen years would go around Sort of Ireland drinking and so on and telling everybody that he's going to be one of if not the greatest writers of the 20th century and he turned out to be that so that that I always think about that little story that somebody told me because I have a lot of people come up to me including myself, I'm a bit of a dreamer. Uh and you know, you get into certain moods where you say I'm going to be the greatest anything ever. You get you get like, people tell you this young, especially young people and it kind of and it makes me feel all kinds of ways, but that story reminds me that like you just might be One of the greatest writers of the 21st century, for example, if somebody were to tell me that and like, don't just don't immediately disregard that because one of the people that say that that's almost like a pre condition. That's a good requirement just to believe in yourself. Maybe it's not a full requirement, but it's an interesting story. I [/INST] think when someone tells you that, then it creates a </s><s>[INST] one [/INST] sees an opportunity and then it would be a tragedy if the opportunity weren't captured, right? And so then that creates a um some some impetus, some motivation to do something, something good. I </s><s>[INST] think the mind, it's like uh you know, I guess that's what, like books like the or books or whatever. I don't even know if the book, the secret plugs into they kind of make a hole industry out of it. But there is something about the mind believing something making it a reality. It is just time and time again with steve jobs, like your belief in yourself, your belief in an idea sort of embracing the me versus the world, embracing the madness of this idea and making it a life pursuit somehow, most reality around for some tiny fraction of the population for everybody else you descend into ah you know, all all the beautiful ways that failure materializes in our lifetime. [/INST] Well, you know that you mentioned love earlier. I mean, that's that's a great example of how belief in something makes it real, right? It's it's not reasonable on the face of it, but because you believe it's reasonable that it actually does become reasonable and then it is real. And so that's a that's a good example. That doesn't happen. I'm also in a bioengineering department. We don't imagine that abridges soundly built and then it is soundly built. That's something that it doesn't come up in too many realms of human existence, but love is one of them and and these the ability to to have a fixed idea and to say it's true and then it is true. </s><s>[INST] A bridge is a kind of manifestation of love. So maybe it does work a little bit, but you know, they can break down like Chernobyl did you can't just say it's safe. The afterwards also prove it's safe. But on Finnegan's wake, I think maybe correct me if I'm wrong, you're using kind of Finnegans wake to give one perspective on what madness is of what's going on in the mind, how much of that is uh that were simply unable to communicate with the person on the other side of of their mind, Like there's almost like a a little person inside the brain. And they have some circuitry that's used to communicate emotion, communicate ideas to the outside world. And there's something about that circuitry that makes it difficult to communicate with the little person on the other side. So [/INST] if you look at what shows up in schizophrenia with many cases, the what we call thought disorders, what we call the, the individual speech symptoms of schizophrenia, Finnegans wake is loaded with them. And it's, it's just full of them. We we talked about clang associations in schizophrenia where the word that is said is echoes in some way the previous word. And as we call that a client association because there's no other reason than the similarity of the sound, like a, like a clang of a, of a garage door being hit. And it has a uh, and sometimes it's not even a word. And we call that a neologism a new word being created. And of course, Finnegans wake is full of that. And then we also in schizophrenia, whether it's what we call loose associations or tangential thought processes, of course, full of that, where things just go off in directions that are not linear, logical. And you can't read Finnegans wake I think without, um, certainly a psychiatrist, you can't read it without thinking about schizophrenia. And then when we look at the families of people with schizophrenia and Joyce was no exception. They're very often are people within the family who are on the spectrum. Some have it, some are able to see it from a distance from a safe distance. There is an association between schizophrenia and what we call, it's a typical personality disorder where people are not quite in this severe state of schizophrenia but have some magical thinking have some unusual thought patterns. Very often. Those are family members of people with schizophrenia. So this points to this again to this idea that that there is a range even along this very severe, very genetic biological illness that human beings dwell on different spots along that spectrum. </s>

<s>[INST] So brainstorming about specific big picture, both big picture and very specific detail of a game that's not working, something's not working, you want to fix it, that kind of stuff because [/INST] you feel like, okay, you've made tangible progress on the actual build of the game or something you played in the beginning the day. Didn't feel great. You've figured out a solution with a group of people. Like it's always with a group and then the next day you're like, yeah, that was that worked out </s><s>[INST] who's on the team is the designers, engineers [/INST] all the above </s><s>[INST] artists voice so [/INST] internal to the studio. It's, you know, a lot of programming. A lot of art you have design, which breaks into some, you know, quest design, writing systems design, who are like doing all the treasures and the loot and the and the skill systems and then level design is making the spaces like those that you'll play through. Production is a big part of it. The producers who organize everything. Um I can't remember, I mentioned art a lot of artists um QA staff as well. They're hugely valuable and saying, hey, we broke your game in these magical ways. What you gonna do about it </s><s>[INST] is the loot design team, still hiring. How do I apply? That seems like the most [/INST] always, </s><s>[INST] I mean, all of this seems like a super fun, [/INST] you know, what's the best, then you have audio. Um and it by far is the greatest job you could possibly have if you're into like technology, it's great if you're in the storytelling creativity and art, it's great and it's really the gaming, you know, the combination of that </s><s>[INST] and like, like I mentioned to you offline, I think video games to me has brought thousands of hours of happiness. And so when you're designing the game, whatever you're doing, you have a part to play in the thing that's going to bring like millions, hundreds of millions of hours of happiness to people. And you know, right? [/INST] It is. And I'm gonna I'm gonna play play you saying that back to our team because you know, people forget your heads down, you're trying to solve these problems and then you do forget how many people it touches like </s><s>[INST] even tiny decisions [/INST] you make </s><s>[INST] tiny little things you create. Yeah. It's weird. I wish there was a way to like um I would notice things in a video game and it's like, huh, okay. Like the if it feels good but you don't get that signal. The creator doesn't get that signal. I wish they did. Um I guess you could get that signal by, you know why is like stuck in this room like digging through the loop. We [/INST] do get we do now get a lot of good data on what the players are doing. Well we we know where they've been and where they've died and how long they play in certain sections and we can sort of tell outside of people just telling us on forums or calling or other things. Um we can tell for some data where people are dropping off or having a, you know, we can tell if there's a key frustration point. </s><s>[INST] Do you ever think about making people feel like human feelings when they play like designing like make them feel fear or excitement, anger, longing, loneliness with stuff. [/INST] Yeah, of course. The Big one I'd like to say is the video games give you his pride outside of other, you know, if you watch movies or things like that like Yeah, but you never think like look what I did and that feeling of like accomplishment and pride in what you did where you overcame, you talked about going back to a game that like those are real feelings of like accomplishment that I felt in games that I've played and when we get to see a player feel that um, it's really, really special. The other one is, there is a, you know, there is an escape or to be someone else that's more powerful in what, in our games that you are in real life that gives you a confidence or a perspective. Um we're doing one next week but we've done a number of make a wish visits kids who could wish for anything and they want to come and I want to see the next game and meet the craters and see how you do it and they come with their family. Um and it is like the greatest thing that we do and it reminds you of like how important it is and the other really awesome thing is that you can see like the family change by the end of the day, like they don't, they don't even realize what it meant to their child or what went into it. And it's just that to me is like been been involved with that foundation for a number of years and it's been really good, you know, reminder of how lucky we are and </s><s>[INST] in general for young people, that sense of accomplishment is hard to find. [/INST] Yeah, where they don't, not everybody has it in the outlets that real life provides. </s><s>[INST] Well that's the thing. I mean the world is cruel to when you're young, nobody takes you seriously. You don't get like that's why you everybody always wants to grow up and get all the script as possible, it's [/INST] the hardest it </s>

<s>[INST] Beautiful. Thanks for talking today, my pleasure thanks for listening to this conversation with Erik Brynjolfsson and thank you to our sponsors and Sarah watches, the maker of Classy while performing watches for Sig Matic, the maker of delicious mushroom coffee express VPN. The VPN. I've used for many years to protect my parents on the internet and cash up the app. I used to send money to friends. Please check out the sponsors and the description to get discount and to support this podcast Future. This thing subscribe on YouTube. Review five stars on Apple podcast following Spotify. Support on Patreon will connect with me on Twitter at Lex Friedman and now let me leave you with some words from Albert Einstein. It has become appallingly obvious that our technology has exceeded our humanity. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] Okay, so, I mean on that topic, let me ask you to play Devil's advocate. Is it possible for you to imagine, Look, years from now and looking at your book ah in which ways might your ideas be wrong? [/INST] Oh, I worry about this all the time. Um, </s><s>[INST] yeah. Still useful. [/INST] Yeah, Yeah. I think there's, you know, um, I can I can best relate it to like, things I'm worried about right now. So we talked about this voting idea, It's happening. There's there's no question it's happening. But it could be far more uh, there's there's enough things I don't know about it that it might be working in the ways differently. I'm thinking about the kind of what's voting, who's voting? You know, where our representations. I talked about like, you have 1000 models of the coffee cup like that. That could turn out to be wrong because it may be Maybe there are 1000 models that are sub models, but not really a single model. The coffee cup. I mean, there's things, these are all sort of on the edges, things that I I present is like, oh, it's so simple and clean. Well, it's not that it's always gonna be more complex. And um, and there's parts of the theory which I don't understand the complexity well, so I think, I think the idea that the brain is a distributed modeling system is not controversial at all. Right. That's not that's well understood by many people. The question then is our each cortical column, an independent modeling system. Um, I could be wrong about that. Um, I don't think so, but I worry about it. </s><s>[INST] My intuition, not even thinking why you could be wrong is the same intuition I have about any sort of physicists like string theory that we as humans desire for a clean explanation. And 100 years from now. Uh, intelligence systems might look back at us and laugh at how we try to get rid of the whole mess by having simple explanation when the reality is, it's it's way messier and in fact, it's impossible to understand. You can only build it. It's like this idea of complex systems and sell your Tamara. You can only launch the thing. You cannot understand it. [/INST] Yeah, I think that, you know, the history of science suggests that's not likely to occur. Um, the history of science suggests that as a theorist and we're theorists. You look for simple explanations, right, fully knowing that whatever simple explanation you're gonna come up with is not going to be completely correct. I mean, it can't be, I mean, it's just it is more complexity, but that's the role of theorists play. They sort of give you a framework in which you now can talk about a problem and figure out, okay, now we can dig in more details. The best frameworks stick around while the details change. You know, again, you know, the classic example is Newton and Einstein, right. You know, um Newton's theories are still used, they're still valuable, they're still practical, they're not like wrong, just they've been </s><s>[INST] refined. Yeah, but that's in physics, it's not obvious by the way. It's not obvious for physics either, that the universe should be such, that amenable to these [/INST] simple, but it's so far it appears to be as far as we can tell. </s><s>[INST] Uh Yeah, I mean, but as far as we can tell, and but it's also an open question whether the brain is amenable to such clean theories, that's the not the brain, but intelligence. [/INST] Well, I don't know, I would take intelligence just say, you know, um well, okay, um the evidence we have suggests that the human brain is a at the one time extremely messy and complex, but there's some parts that are very regular and structured. That's why we started the new york cortex it's extremely regular in its structure and unbelievably so, and then I mentioned earlier, the other thing is it's universal abilities, it is so flexible to learn so many things we don't, we haven't figured out what I can't learn yet, We don't know, but we haven't figured out yet, but he learns things that it never was evolved to learn. So those give us hope. Um, that's why I went into this field because I said, you know, this regular structure, it's doing this amazing number of things. There's gotta be some underlying principles that are common and other scientists have come up with the same conclusions. Um, and so it's promising, it's promising and um, and that's and whether the theories play out exactly this way or not, that is the role that theorists play. And so far it's worked out well, even though, you know, maybe, you know, we don't understand all the laws of physics, but so far it's been pretty damn useful. The ones we have, our theories are pretty useful. </s><s>[INST] You mentioned that we should not necessarily be at least to the degree that we are worried about the existential risks of artificial intelligence, relative to human risks from human nature being existential risk. What aspect of human nature worries you the most in terms of the survival of the human species, [/INST] I mean I'm disappointed in humanity's humans, I mean all of us, I'm one, so I'm disappointed myself too. Um it's kind of a sad state, there's there's two things that disappoint me. One is how it's difficult for us to separate our rational component of ourselves from our evolutionary heritage, which is, you know, not always pretty, you know, um, rape is a, is an evolutionary good strategy for reproduction. Murder can be at times too, you know, making other people miserable at times is a good strategy for reproduction. It's just, and it's just, and so now that we know that and yet we have this sort of, you know, you and I can have this very rational discussion talking about, you know, intelligence and brains in life and some, so it seems like it's so hard. It's just a big transition, forget humans, all humans to, to make the transition from like, let's pay no attention to all that ugly stuff over here. Let's just focus on the incident. What's unique about humanity is our knowledge and our intellect. </s>

<s>[INST] to be heard. And just to feel things in silence together in terms of weddings and marriage. What's the role of that whole? I'm just I need to take some notes here. What's what's [/INST] the </s><s>[INST] role of marriage in in human existence? [/INST] It is first of all to teach you how to care for someone unlike you, which could be anyone you marry. And I think it's to create a home and a family. So </s><s>[INST] there's a commitment to its care for a long time, [/INST] right? Exactly. And also when when couples come to me and they say we don't need to be married because it really won't change how we think about ourselves and our relationship, I said, and that's true. It might not, but it will change how everyone else looks at you. And because it changes how everyone else looks at you, it changes you because it's one thing to say, this is my partner, it's another thing to say, this is my husband. You say this is my husband. That means we've made a real commitment to this. </s><s>[INST] Yeah. What do you do you do you worry that there's a dissolution of that as well in terms of um how, you know, as as religion dissipates like it it loosens its hold on society, loosens its impact on society. Do you worry about that? [/INST] I worry about it. Um I do think that it is possible that were going rather than a dissolution. We're going through a transition that is different kinds of families and different configurations of families. That is. I see some of that, but I also do see uh it's less a dissolution of marriage than it is of the idea of commitment. And I'll give you like a simple example when I was growing up, a player on a sports team was always on that team And you rooted for the team because you knew the players for 20 years now. There are very good reasons starting with curt flood why why people got free agency and they can move around and it's better for the players. I understand all that and I am not, I'm not saying, oh, they should continue, but just like people move jobs and they move sports teams and they change careers, they change partners. And there is uh there's a diminishment of the commitment to commitment that I actually think has serious societal consequences. And that that I am worried about. </s><s>[INST] Yeah, there's a there's a cost to that. I don't know what it is about commitment. That's beautiful like through because like some of the deepest friendships I have is when we've gone through some ship together and so like the hard times going through hard times together, especially when the hard times are between the two of you. That like if I mean that's always a risk, but if it, if you can find a way through that can bond you stronger. That's the fascinating thing about human relations. [/INST] No question. And even if it doesn't keep you forever, you still have a connection that doesn't, that exists that so I can give you one when you said, what is it about commitment? I'll give you one. I think beautiful answer. Someone once asked uh Rabbi Sullivan, who was a great thinker and leader in the Orthodox community in the 20th century? They said, you know, I go from religion to religion, I just take what I think is beautiful in it. And his answer was that you're treating religion like a nomad. He said nomads go from place to place and they eat what they want and they move on. He says farmers stay in one place, the differences farmers make things grow. And I think that that's true. Also when you think about the relationships, you have things have grown out of the relationships that you've invested in that you've farmed basically that can't exist in fly by night relationships, </s><s>[INST] can you talk about? Can we talk about the Torah? [/INST] Yes, What </s><s>[INST] is it? And is it the literal word of God [/INST] Um </s><s>[INST] Easy questions. [/INST] Yeah. Well the Torah is the five books of moses written in Hebrew. Um I like most, I think modern rabbis, non orthodox or non literalist rabbis will tell you that it's a product of human beings um and I believe that they are inspired by God, but it's clear to me that it's a human product and I think that people who study modern biblical criticism, it's really hard to study modern modern criticism gives the wrong impression. I would say modern scholarship on the bible and not appreciate the fact that it's it even has levels of language. I mean it's just like if you read today um somebody writing like Shakespeare, you would say this isn't, it's it's like english has developed, it's different, it's not the english we speak today and if you study the bible and you know Hebrew well enough, you even see that this was written over hundreds and hundreds and hundreds of years. It is a holy book and I like the idea that it is what what you say in Hebrew is to remain hashem. I am and not to remission I that is the Taurus from heaven, but it's not from Sinai, so it has its origin beyond us, but it has things in it that I think and this is one of the, one of the things that was a huge controversy at my congregation when I started to do same sex marriages, there are some people who try to argue that the Torah does not forbid them whether it does or not. It seems to me we understand things that we're not understood in the ancient world about gender and sexuality. And so, </s>

<s>[INST] Again, [/INST] UNIX </s><s>[INST] started being developed 50 years ago. It would be more than 50 years ago. Can you tell the story? Like you describe your new book of how UNIX was created [/INST] ah if I can remember that far back, it was some while ago. Um, so I think the gist of it is that at Bell Labs in 1969 there were a group of people who had just finished working on the Multex project which was itself a follow on to C. T. S. S. So we can go back sort of an infinite regress in time. But the uh CT ss was very very very nice time sharing system. It was very nice to use. I actually used it as that summer I spent in Cambridge in 1966. </s><s>[INST] What was the hardware there? So what's the operating system? What's the hardware there? What's the CTS look like? [/INST] So C. T. S. S looked like uh kind of like a standard time sharing system? Certainly at the time it was the only time sharing. If no, let's </s><s>[INST] go back to the basic, what's it time for my system? [/INST] Okay. In the beginning was the word and the word. </s><s>[INST] And then there was time sharing systems. [/INST] Yeah. If we go back into, let's call it the 1950s and early 1960s, most computing was done on very big computers, physically big, although not terribly powerful by today's standards that were maintained in very large rooms and you use things like punch cards to write programs on and talk to him. So you would take a deck of cards, write your program on it and send it over a counter handed to an operator and some while later back would come something that said, oh you made a mistake and then you'd recycle. And so it's very very slow. So the idea of time sharing was that you take basically that same computer but connect to it with something that looked like an electric typewriter that could be a long distance away. It could be closed. But fundamentally what the operating system did was to give each person who was connected to it and wanting to do something a small slice of time on uh to do a particular job. So I might be editing a file so I would be typing. And every time I hit a keystroke the operating system would wake up and said oh he typed character, let me remember that. Then you go back to doing something else would be going around and around a group of people who were trying to get something done giving each a small slice of time and giving them each the illusion that they pretty much had the whole machine to themselves. And hence time sharing. That is sharing the computing time resource of the computer among a number of people who are doing it </s><s>[INST] without the individual people being aware that there's others in a sense the illusion the feeling is that the machine is your own. [/INST] Pretty much that was the idea. Yes you had if it were well done and if it were fast enough and other people were doing too much you did have the illusion that you had the whole machine to yourself and it was very much better than the punch card model. And so C. T. S. S. The compatible time sharing system was I think arguably the first of these, it was done I guess technically 64 or something like that. It ran on an IBM 70 94 slightly modified to have twice as much memory as the norm. It had two banks of 32 K words instead of one. So </s><s>[INST] 32 K words. Each word [/INST] was 36 beds. So call it you know about 100 and 50 kilobytes times two. So by today's standards that's down in the noise </s><s>[INST] but at the time [/INST] that was a lot of memory and memory was expensive. So C. T. S. S. Was just a wonderful environment to work on. It was done by the people that M. I. T. Led by uh Fernando Corbett over Corby who died just earlier this year. Um and a bunch of other folks and then so I spent the summer of 66 working on that had a great time. Uh meant a lot of really nice people. And indirectly uh knew of people at Bell labs who were also working on a follow on to C. T. S. S. That was called multiplex. The multiplex was meant to be the system that would do everything that C. T. S. S. Did but do it better for a larger population </s><s>[INST] all the usual stuff. Now the actual time sharing the scheduling. Uh how much what's the algorithm that performs the scheduling? What's that look like? How much magic is there? What are the metrics? How does it all work in the beginning? [/INST] So the answer is I don't have a clue. I think the basic idea was nothing more than who all wants to get something done. Suppose that things are very quiet in the middle of the night then I get all the time that I want. I suppose that you and I are contending at high noon for something like that then probably the simplest algorithm is a round robin one that gives you a bit of time, gives me a bit of time and then we could adapt to that. Like what are you trying to do? Are you text editing or are you compiling or something? And we might adjust the schedule according to things like that. </s><s>[INST] So Okay, so Multex was trying to just do some of the clean it up a little bit. Well, [/INST] it was it was meant to be much more than that. So Multex was the multiplex information and computing service and it was meant to be a very large thing that would provide computing utility. Something that where you could actually think of it as just a plug in the wall service sort of like cloud computing today. Same idea. But 50 odd years earlier. And so uh what Multex offered was a richer operating system environment piece of hardware that was better designed for doing the kind of sharing of resources and presumably lots of other things. </s><s>[INST] Do you think people at that time had the dream of what cloud computing is starting to become now. Which is computing is everywhere. That you can just plug in almost you know and you never know how the magic works. You just kind of plug in at your little computation they need to perform and it does. It was that the dream? I [/INST] don't know where that was the dream. I wasn't part of it at that point. Remember I was an intern first summer. But my sense is Given that it was over 50 years ago. Yeah. They had that idea that it was an information utility. That it was something where if you had a computing task to do you could just go and do it now. I'm betting that they didn't have the same view of computing for the masses. Let's call it the idea that you know, your grandmother would be shopping on amazon. I don't think that was part of it. But if your grandmother were a programmer it might be very easy for her to go and use this kind of utility. </s><s>[INST] What was your dream of computers at that time? What did you see as the future of computers? Because you have predicted what computers are today? Oh, [/INST] short answer. Absolutely not. I have no clue. I'm not sure I had a dream. It was a dream job in the sense that I really enjoyed what I was doing. I was surrounded by really, really nice people, Cambridge is very fine city to live in in the summer. Less so in the winter when it snows. But in the summer it was a delightful time. And so I really enjoyed all of that stuff and I learned things and I think the good fortune of being there for summer led me then to get a summer job at Bell Labs the following summer and that was going to be useful for the future. </s><s>[INST] So this Bell Labs is this magical legendary place. So first of all, where is by labs? And can you start talking about that journey towards UNIX at Bell Labs? [/INST] Yeah, So Bell Labs is physically um scattered around at the time, scattered around New Jersey. The primary location was in a town called Murray Hill or a location called Murray Hill is actually That across the boundary between two small towns in New Jersey called New Providence in Berkeley Heights. Think of it as about 15, 20 miles straight west of New york city and therefore about an hour north of here in Princeton. Um And at that time it had make up a number three or 4000 people there, many of whom had phds and mostly doing physical sciences, uh Chemistry, physics, materials kinds of things, but very strong math and it rapidly growing interest in computing as people realized you could do things with computers that you might not have been able to do before. You could replace labs with computers that had worked on models of what was going on. So that was, that was the essence of Bell labs and again, I wasn't a permanent play there. I was, that was another internship I got lucky and internships, </s><s>[INST] I mean if you could just lingering a little bit, what was the, what was in the air there? Because some of the, the number of Nobel prizes, the number of touring awards and just legendary computer scientists that come from their inventions, including developments including UNIX, it's just uh, it's unbelievable. So is was there something special about that place? [/INST] Oh, I think there was very definitely something special. Um I mentioned the number of people is a very large number of people, very highly skilled and working in an environment where there was always something interesting to work on because the goal of Bell labs, which was a small part of AT and T, which provided basically the country's phone service. The goal of a TNT was to provide service for everybody and the goal of Bell labs was to try and make that service, keep getting better. So improving service and that meant doing research on a lot of different things, physical devices like the transistor or fiber optical cables or microwave systems, All of these things, the lives worked on and it was kind of just the beginning of real boom times. And computing as well as when I was there. I went there first in 66. So computing was at that point fairly young and so people were discovering that you could do lots of things with computers. </s><s>[INST] So how's UNIX born? [/INST] So mall ticks in spite of having an enormous number of really good ideas, lots of good people working on it fundamentally didn't live up at least in the short run and I think ultimately really ever to its goal of being this information utility, it was too expensive and certainly what was promised was delivered much too late. Um and so in roughly the beginning of 1969, Bell Labs pulled out of the project. The project at that point had included M. IT. Bell labs and General Electric. General electric made computers. A general electric was the hardware operation. So Bell labs realizing this wasn't going anywhere on a time scale they cared about pulled out of the project. And this left several people with an acquired taste for really, really nice computing environments, but no computing environment. And so they started thinking about what could you do if you're going to design a new operating system that would provide the same kind of comfortable computing as C. T. S. S. Had, but also the facilities of something like Multex brought forward. Um And so they did a lot of paper design stuff and at the same time, Ken, Thompson found what is characterised as a little used PDP seven where he started to do experiments with file systems, just how do you store information on a computer and efficient way? And then this famous story that his wife went away to California for three weeks, taking their one year old son and three weeks and he sat down and wrote an operating system which ultimately became UNIX. So software productivity is good in those days. </s><s>[INST] So PDP what's the PDP seven? So it's a piece of hardware, you [/INST] know, it's a piece of part where it was one of the early machines made by Digital equipment, Corporation deck. Um and it was a mini computer, so called, it Had, I would have to look up the numbers exactly, but it had a very small amount of memory, maybe 16K-16 bit words or something like that, relatively slow. probably not super expensive. Maybe again, making this up, I'd have to look at $100,000 or something like that, </s><s>[INST] which is not super expensive in those days. It [/INST] was expensive, it was enough that you and I probably wouldn't be right by one, but a modest group of people could get together, but in any case in It came out, if I recall in 1964, so by 1969 it was getting a little obsolete and that's why it was little used </s><s>[INST] if you can sort of comment, what do you think it's like to write an operating system like that. So that process that can went through in three weeks because you were I mean you're part of that process, you contributed a lot to uh to UNIX his early development. So what do you think it takes to do that first step, that first kind of from design to reality on the PDP? Well [/INST] let me correct one thing, I had nothing to do with it, so I did not write it. I have never written operating system code </s><s>[INST] um [/INST] and so I don't know now, an operating system is simply code and this first one wasn't very big but it's something that lets you run processes of some let you execute some kind of code that has been written. It lets you store information for periods of time so that it doesn't go away when you turn the power off or reboot or something like that. Um and there's a kind of a core set of tools that are technically not part of an operating system, but you probably need them in this case can wrote an assembler for the PDP seven that worked needed to text editor so that you could actually create text, He had the file system stuff that he had been working on and then the rest of it was just a way to load things, execute able code from the file system into the memory, give it control and then recover control when it was finished or in some other way </s><s>[INST] quit. What was the code written in the primarily the programming language was an assembly PDP [/INST] seven assembler that can created These things were assembly language until probably the call in 1973 or 74, something like that. </s><s>[INST] Uh forgive me if it's a dumb question, but it feels like a daunting task to write any kind of complex system. An assembly absolutely like it feels like impossible to do any kind of what we think of a software engineering. Assembly is to work in a big picture, sort of, [/INST] I think it's hard, it's been a long time since I wrote assembly language. It is absolutely true that assembly language, if you make a mistake, nobody tells you there are no training wheels whatsoever. Um and so stuff doesn't work now. What? And there's nota bookers, Well, there could be de buggers, but that that's the same problem. Right. How do you actually get something that will help you debug it? So part of it is, is um an ability to see the big picture. Now, these systems were not big in the sense that today's pictures. I see the big picture was, in some sense more manageable. I mean, then realistically there's an enormous variation in the capabilities of programmers and ken Thompson who did that first one is kind of the singularity uh in my experience of programmers, uh with no disrespect to you or even to me, he's into now </s><s>[INST] several [/INST] leagues removed. </s><s>[INST] I know there's levels, it's a, it's a fascinating thing that there are unique stars in particular in the programming space and in a particular time, you know, the time matters to the timing of when that person comes along and the wife does have to leave. Like there's this weird timing that happens that and then all of a sudden something beautiful is created. How does it make you feel that there's a system that was created in in three weeks? Or maybe you can even say on a whim, but not really. But of course, quickly that is now you could think of most of the computers in the world run on a UNIX like system. Right? What, how do you enter? Like, uh, if you kind of assume from the alien perspective, if you're just observing Earth that all of sudden these computers took over the world and they started from this little initial seed of UNIX, how does that make you feel? [/INST] It's quite surprising. And you asked earlier about prediction, The answer is no, there's no way you could predict that kind of evolution. And I don't know whether it was inevitable or just a whole sequence of blind luck, I suspect more of the latter. And um, so I look at it and think gee that's kind of neat. I think the real question is what this can think about that because he's the guy arguably from whom it really came tremendous contributions from Dennis Ritchie and then others around in that Bell Labs environment, but you know, if you had to pick a single person that would be can, </s><s>[INST] so you've written in your book UNIX a history and a memoir, are there some memorable human stories, funny or profound from that time to just kind of stand out? [/INST] Oh, there's a lot of them in a sense. And again, it's a question of, can you resurrect them in real time when his memory fails? But I think part of it was the Bell Labs at the time was was a very special kind of place to work because there were a lot of interesting people and the environment was very, very open and free. It was a very cooperative environment, very friendly environment. And so if you had an interesting problem, you go and talk to somebody and they might help you with the solution. And um and it was a kind of a fun environment to in which people did strange things and um often tweaking the bureaucracy in one way or another, </s><s>[INST] rebellious and in certain kinds of ways [/INST] in some ways. Yeah, absolutely. I think most people didn't take too kindly to the bureaucracy and I'm sure the bureaucracy put up with an enormous amount that they didn't really want to, </s><s>[INST] so maybe to linger on it a little bit, you have a sense of what the philosophy that characterizes UNIX is the design, not just the initial, but just carry through the years? Just being there being around? What's the fundamental philosophy behind the system? [/INST] I think one aspect the fundamental philosophy was to provide an environment that made it easy to write or easier productive to write programs. So it was meant as a programmer environment, it wasn't meant specifically as something to do some other kind of job. For example, it was used extensively for word processing, but it wasn't designed as a word processing system, it was used extensively for lab control, but it wasn't designed for that. It was used extensively as a front end for big other systems, big dump systems, but it wasn't designed for that. It was meant to be an environment where it was really easy to write programs. That program which could be highly productive. And part of that was to be a community. And there's some observation from Dennis Ritchie, I think at the end of the book that says that that from his standpoint, the real goal was to create a community where people could work uh as programmers on the system. And I think in that sense, certainly for many, many years it succeeded quite well at that. And part of that is the technical aspects of, because it made it really easy to write programs, people did write interesting programs, those programs tended to be used by other programmers. And so it was kind of a virtuous circle of more and more stuff coming out. That was really good for programmers </s><s>[INST] and you are part of that community of programmers. So what was it like writing programs on that early UNIX? [/INST] It was a blast. It really was. You know, I like to program, I'm not a terribly good program but it was a lot of fun to write code. And in the early days there was an enormous amount of what you would today I suppose called low hanging fruit. People hadn't done things before and this was this new environment and the whole combination of nice tools and very responsive system and tremendous colleagues made it possible to write code. You could have an idea in the morning, you could do a, you know, an experiment with it, you could have something limping along that night or the next day and people would react to it and they would say, oh that's wonderful, but you're really screwed up here. And and the feedback loop was then very, very short and tight. And so a lot of things got developed fairly quickly that um in many cases still exist today and I think that was part of what made it fun because programming itself is fun. It's puzzle solving in a variety of ways. But I think it's even more fun when you do something that somebody else then uses, even if they wine about it not working, the fact that they used, it is part of the reward mechanism </s><s>[INST] and what was the method of an interaction? The communication when that feedback loop. I mean, this is before the internet, [/INST] certainly before the internet, it was mostly physical right there. You know, somebody who come into your office and say something. </s><s>[INST] So these places are all close by like offices are nearby, really lively interaction. Yeah, [/INST] no Bell Labs was fundamentally one giant building and most of the people were involved in this unique stuff. We're in two or three quarters and there was a room, oh, how big was it? Probably Call it 50 ft by 50 ft make up a number of that, which had some access to computers there as well as in offices and people hung out there and it had a coffee machine and so there was, it was mostly very physical. We did use email of course. Um, and but it was fundamentally for a long time, all on one machine, so there was no need for internet. </s><s>[INST] It's fascinating to think about what computing would be today without their labs, it seems so many the people being in the vicinity of each other, sort of getting that quick feedback working together. There's so many brilliant people. I don't know where else that could have existed in the world have been given how that came together. What, how does that make you feel that that's a little element of history? Well, [/INST] I think that's very nice, but in a sense, it's survivor bias and if it hadn't happened at Bell Labs, there were other places that were doing really interesting work as well. Xerox parc is perhaps the most obvious one. Xerox parc contribute enormous amount of good material. Um and many of things we take for granted today in the same way came from xerox parc experience. I don't think they capitalized in the long run as much their parent company was perhaps not as lucky in capitalizing on this. Who knows? But that would that's certainly another place where there was a tremendous amount of influence. There were a lot of good university activities. M. I. T. Was obviously no slouch in this kind of thing and and others as well. Um, </s><s>[INST] so UNIX turned out to be open source because of the various ways that a TNT operated and sort of had to uh it was the focus was on telephones. So [/INST] I think that's a mischaracterization in the sense, it absolutely was not open source, It was very definitely proprietary licensed, but it was licensed freely to universities in source code form for many years. And because of that generations of university students and their faculty people grew up knowing about uh UNIX and there was enough expertise in the community that then became possible for people to kind of go off in their own direction and build something that looked UNIX like. Um the Berkeley version of UNIX started with that licensed code and gradually picked up enough of its own uh, code contributions, notably from people like Bill Joy, that uh eventually it was able to become completely free of any T A T M T code and uh, there was an enormous amount of legal jockeying around this, that in the late, Early to late 80s, early 90's, something like that. Um and then uh Not something that I guess the open source movement might have started when Richard Stallman started to think about this in the late 80s, and by 1991 when Torvalds decided he was going to do a UNIX like operating system, there was enough expertise that uh, in the community that first he had a target, he could see what to do because the kind of the UNIX system called interface and the tools and so on, were there. Um, and so he was able to build an operating system that at this point when you say, you know, in many cases what you're really thinking is Lennox, </s><s>[INST] Lennox. Yeah, but it's funny that from my distant perception, I felt that, you know, it was open source without actually knowing it, but what you're really saying, it was just uh freely licensed, so it's freely licensed, so it felt open sources because universities are not trying to make money so there it felt open source in the sense that you can get access if you want it, right [/INST] and a very, very, very large number of universities had the license and they were able to talk to all the other universities who had the license and so technically not open, technically belonging to A T and T pragmatically pretty open. </s><s>[INST] And so there's a ripple effect that all the faculty and the students then they all grew up and then uh they went throughout the world and uh permeate in that kind of way. So what kind of features do you think made for a good operating system? If you take the lessons of UNIX, you said that, you know, make it easy for programmers like that seems to be an important one, but also UNIX turned out to be exceptionally robust and efficient. Right. So is that an accident when you focus on the programmer or is that a natural outcome? [/INST] I think part of the reason for efficiency was that it began on extremely modest hardware. Very, very, very tiny. And so you couldn't get carried away. You couldn't do a lot of complicated things. Yeah, because you just didn't have the resources, either processor speed or memory. And so that enforced a certain minimum al Itty of mechanisms and maybe a search for generalizations so that you would find one mechanism that served for a lot of different things rather than having lots of different special cases. I think the file system and UNIX is a good example of that file system interface and its fundamental form is extremely straightforward and that means that you can write code very, very effectively for the file system and then one of those ideas? And one of those generalizations is that g that file system interface works for all kinds of other things as well. And so in particular, the idea of reading and writing to devices is the same as reading and writing to a disk that has a file system and then that gets carried further in other parts of the world. Processes become in effect files in a file system. Um and the Plan nine operating system which came along, I guess in the late 80s or something like that, uh took a lot of those ideas from the original UNIX and tried to push the generalization even further. So that in plan nine, a lot of different resources are file systems, they all share that interface. So that would be one example. Uh We're finding the right model of how to do something means that an awful lot of things become simpler and it means therefore that more people can do useful, interesting things with them without him to think as hard about it. </s><s>[INST] So, you said you're not a very good programmer, you're you're the most modest human being. Okay, But you'll continue saying that I understand how this works, but you do radiate a sort of love for programming. So let me ask, uh do you think programming is more an art or science creativity or kind of rigor? [/INST] I think it's some of each, it's some combination, um some of the art is figuring out what it is it do you really want to do? What should that program be? What what would make a good program? And that's some understanding of what the task is, what the people who might use this program want. And I think that's that's art in many respects. The science part is trying to figure out how to do it well. And some of that is a real computer sciences stuff. Like what algorithms should we use at some point? Mostly in the sense of being careful to use algorithms that will actually work properly scale, properly avoiding quadratic algorithms when a linear algorithm should be the right thing, that kind of more formal view of it. Um Same thing for data structures. But also it's I think an engineering field as well and engineer is not quite the same as science because engineering, you're working with constraints, you have to um figure out not only what is a good algorithm for this kind of thing, but what's the most appropriate algorithm given the amount of time we have to compute the amount of time we have to program what's likely to happen in the future with maintenance, who's going to pick this up in the future, all of those kind of things that if you're an engineer you get to worry about. Whereas if you think of yourself as a scientist? Well, you can maybe push them over the horizon in a way and if you're an artist, what's that? </s><s>[INST] So just on your own personal level, what's your process? Like a writing program? Say a small and large sort of tinkering my stuff. Do you just start coding right away and just kind of evolved iterative lee with a loose notion or do you plan on a sheet of paper first and then kind of design in this, what they teach you in the kind of uh software engineering courses in undergrad or something like that. What's your process? Like? [/INST] It's certainly much more the informal, incremental first, I don't write big programs at this point, it's been a long time since I wrote a program that was more than that. I call it a few 100 or more lines, something like that. Many of the programs are right or experiments for either something I'm curious about or often for something that I want to talk about in a class and so those necessarily tend to be relatively small. Um a lot of the kind of code I write these days tends to be for sort of exploratory data analysis where I've got some collection of data and I want to try and figure out what on earth is going on in it and for that, those programs tend to be very small. Sometimes you're not even programming, you're just using existing tools, like counting things or sometimes you're writing och scripts because two or three lines will tell you something about a piece of data and then when it gets bigger, well then I will probably write something in python um because that scales better up to call it a few 100 lines or something like that. And it's been a long time since I wrote programs that were much more than that. </s><s>[INST] Speaking of data exploration in arc first, what is arc? [/INST] So as a scripting language that was done by myself, L A Ho and Peter Weinberger, we did that originally in the late seventies was a language that was meant to make it really easy to do quick and dirty uh tasks like counting things or selecting interesting information from basically all text files, rearranging it in somewhere summarizing, it </s><s>[INST] runs a command on each line or a file. I mean there's uh it's still exceptionally widely used today. Oh, absolutely, yeah, it's so simple and elegant, sort of, the way to explore data turns out you can just write a script that does something seemingly trivial in a single line and that giving you that slice of the data somehow reveals something fundamental about the data and that keeps, that seems to work still. [/INST] Yeah, it's very good for that kind of thing and what that's sort of what it was meant for, I think what we didn't appreciate was that the model is actually quite good for a lot of data processing kinds of tasks and that it's, it's kept going as long as it has because at this point, it's over 40 years old and it's still, I think a useful tool and well this is paternal interest I guess, but I think in terms of programming languages, you get the most bang for the buck by learning. Och, and it doesn't scale the big programs, but it does pretty, pretty darn well on these little things where you just want to see all the something's in something. So yeah, I find I probably write more hawk than anything else at this point. </s><s>[INST] Uh, so what kind of stuff do you love about talk like is there if you can comment on sort of things, uh, that give you joy when you can in a simple program reveals something about the date. Is there something that stands out in particular features? I [/INST] think it's mostly the selection of default behaviors that you sort of hinted at it a moment ago. What art does is to read through a set of files and then within each file, it reads through a, each of the lines and then on each of the lines, it has a set of patterns that it looks for, that's your art program. And if one of the patterns matches, there is a corresponding action that you might perform. And so it's kind of a quadruple the nested loop or something like that. Um, and that's all completely automatic. You don't have to say anything about it, you just write the pattern in the action and then run the day to buy it. And and so that paradigm for programming is a very natural and effective one, and I think we captured that reasonably well and talk and it does other things for free as well. It splits the data into fields so that on each line, their fields separated by white space or something. And so it does that for free, you don't have to say anything about it. Um and it collects information, is goes along like, what line are we on? How many fields are there on this line? So lots of things that just make it, so that a program which in Another language, let's say Python would be 5, 10, 20 lines in August one or 2 lines. </s><s>[INST] So because it's one or two lines, you can do it on the shell, you don't have to open up another whole thing, you can just do it right there in the interaction with the operating directly. Is there other shell commands that you love over the years? Like you really enjoy using major [/INST] grab? </s><s>[INST] The only. Right, Yeah, [/INST] graph does everything. Uh so </s><s>[INST] grab is a kind of uh what is it a simpler version of arc? I would say [/INST] in some in some sense. Yeah. Right, because what is grap, so grab is basically searches the input for particular patterns, regular expressions, technically of a certain class um and it has that same paradigm that och does its pattern action thing. It reads through all the files and then all the lines in each file, but it has a single pattern which is the regular expression you're looking for in a single action printed if it matches. So it's in that sense, it's a much simpler version and you could write crap in och as a one liner. Um and I use grab probably more than anything else at this point. Uh just because it's so convenient and natural, </s><s>[INST] Why do you think it's such a powerful tool grab knock? Why do you think operating systems like Windows for example, don't have it sort of, you can of course I use the which is amazing. Now there's a Windows for Lennox like, which you can basically use all the fun stuff like all can grab and inside of Windows, but Windows naturally sort of as part of the graphical interface. The simplicity of grap, sort of searching through a bunch of files and just popping up naturally, why don't you think that what you think that's unique to the UNIX and Linux environment? I [/INST] don't know, it's not strictly unique, but it's certainly focused there and I think some of it's the weight of history um that Windows came from MSD SMS Das was a pretty pathetic operating system, although common on, you know, unbounded lee large number of machines, But somewhere in roughly the 90s windows became a graphical system and I think Microsoft spent a a lot of their energy on making that graphical interface what it is. Um And that's a different model of computing and it's a model of computing that where you point and click and sort of experiment with menus, it's a model of computing works right rather well for people who are not programmers just want to get something done. Whereas teaching something like the command line to non programmers turns out to sometimes be an uphill struggle. And so I think Microsoft probably was right and what they did now you mentioned whistle or whatever it's called when it's like </s><s>[INST] this. I wonder was pronounced WsL is never actually pronounced the whistle. I like it. I have no [/INST] idea. But there have been things like that for longest seguin, for example, which is a wonderful collection of. Take all your favorite tools from UNIX and Linux and just make them work perfectly on Windows. And so that's a something that's been going on for at least 20 years if not longer. And I use that on my one remaining Windows machine uh routinely because it's for if you're doing something that is batch computing command suitable for command line, uh That's the right way to do it because the Windows equivalents are if nothing else not familiar to me, </s><s>[INST] but I should, I would definitely recommend to people to if they don't use sigma to try whistle. I I've been so excited that I could use bash you bash, write scripts quickly In windows, it's changed my my life. Okay. What's your perfect programming set up? What computer, what operating system with keyboard? What editor? [/INST] Yeah. Perfect. Is too strong a word. It's way too strong a word. Um What I used by default, I have a uh at this point a 13 inch Macbook Air which I use because it's kind of a reasonable balance of the various things I need. I can carry it around. It's got enough computing, Horse bar screens, big enough keyboards. Okay. Um and so I basically do most of my computing on that. Um I have a big Mac in my office that I use from time to time as well, especially when I need a big screen, but otherwise it tends not to be used as much. Um uh </s><s>[INST] editor [/INST] I use mostly SaM, which is an editor that rob pike wrote long ago at Bell Labs. And </s><s>[INST] that sarge interrupted. Does that proceed vis I does that post [/INST] it. Post dates both V. I. And E max. Um It is derived from rob's experience with E. D. E. N. V. I. </s><s>[INST] Um [/INST] that's the original UNIX editor. Oh wow. Dated probably before you were born. So </s><s>[INST] what's uh actually, what's the history of editors? Can you, can you briefly because this is your fat. I used imax, I'm sorry to say, I'm sorry to come out with that. But uh what's what's the kind of interplay there? So in [/INST] ancient, ancient times, like, call it the first time sharing systems. Going back to what we're talking about, there were editors. There was an editor on C T S. S that I don't even remember what it was called. It might have been edit um where you can type text, program text and it would do something or document text </s><s>[INST] and [/INST] save it. You could edit it usual thing that you would get an editor. Um and ken, Thompson wrote an editor called QED, which was very, very powerful. But these were all totally a command base. They were not most or cursor based because it was before mice and even before cursors because they were running on terminals that printed on paper. Okay. No, no CRT type displays, let alone L A D S. Um And so then when UNIX came along, ken took Q E D and stripped way way, way, way down. And that became an editor that he called A D. It was very simple, but it was a line oriented editor. And so you you could load a file and then you could talk about the lines, one through the last line and you could um, you know, print ranges of lines, you could add text, you could delete text, you could change text or you could do a substitute command that would change things within a line or within groups of lines. So you can work on the parts of a </s><s>[INST] file essentially. You know [/INST] you can work on any part of it the whole thing or whatever but it was entirely command line based and it was entirely on paper. Okay. Paper. And that meant that you changed. Yeah. Right. Real paper. And so if you change the line you have to print that line using up another line of paper to see what change caused. Okay so when when CRT displays came along then you can start to use cursor control and you can sort of move where you were on the screen in uh </s><s>[INST] without reprinting every printing. [/INST] And one of there were a number of editors there. The one that I was most familiar with and still use is V. I. Which was done by Bill Joy. And so uh that dates from probably the late 70s as a guess. And uh it took full advantage of the cursor controls. I suspected imax was roughly at the same time but I don't know I've never internalized imax. So so I use at this point I stopped using E. D. All I can um I use V. I sometimes and I use SAM when I can. </s><s>[INST] And SAM is available on most systems. It [/INST] Was it is available. You have to download it yourself from typically the Plan nine operating system distribution. It's been maintained by people there and so </s><s>[INST] I'll get home tonight. I'll try it. It's cool. It sounds it's uh sounds sounds fascinating although my love is with Lisbon imax have went into that hippie world of, I [/INST] think it's a lot of things. What religion were you brought up with? That's </s><s>[INST] true, that's right. Most of the actual programming I do is C. C plus plus and python. But my weird sort of my religious upbringing is in list. So can you take on the impossible task and give a brief history of programming languages from your perspective? [/INST] So I guess you could say programming languages started probably in what the late forties or something like that. People used to program computers by basically putting in zeros and ones using something like switches on a console. Uh and then or maybe holes in paper tapes, something like that. So extremely tedious, awful whatever. And so I think the first programming languages were uh relatively crude assembly languages. Where people would basically write a program that would convert Pneumonic, like add a. D. D. Into whatever the bit pattern was that corresponded to an add instruction. And they would do the clerical work of figuring out where things were. So you could put a name on a location in the program and the assembler would figure out where that corresponded to when the thing was all put together and dropped into memory. Um and early on. And this would be the late forties uh very early fifties there were assemblers written for the various machines that people used you may have seen in the paper just a couple of days ago, Tony Becker died. He did this thing in Manchester called the called Auto Code, a language which I knew only by name. Um, but it sounds like it was a flavor of assembly language sort of a little higher in some ways. Um, and it replaced the languages alan turing wrote, which you put in zeros and ones, but you put it in backwards order because that was a hardware worked very </s><s>[INST] That's right, yeah, that's right backwards. [/INST] So assembly languages then let's call that the early 19 fifties and so every different flavor of computer has its own assembly language. So the egg sac had hits in the Manchester had it and the IBM whatever, 70 90 or 704 or whatever had hits and so on. So everybody had their own assembly </s><s>[INST] language. Assembly languages have a few commands, addition, subtraction, then branching of some kind if then type of situation right. [/INST] They have exactly in their simplest form, at least one instruction per or one assembly language instruction per instruction in the machines repertoire. And so you have to know the machine intimately to be able to write programs in it. And if you write an assembly language program for one kind of machine and then you say, geez, it's nice and like a different machine Start over. Okay, so very bad. And so what happened in the late 50s was people realized you could play this game again and you could move up a level in writing or creating languages that were closer to the way that real people might think about how to write code. Um, and where, I guess arguably three or four at that time period, there was FORTRAN which came from IBM, which was formula translation. Meant to make it easy to do scientific and engineering computation. </s><s>[INST] That formula translation, that's what I stood for. It was COBOL, [/INST] which is the common business oriented language that Grace Hopper and others worked on, which was aimed at business kinds of tasks. Uh, there was alcohol which was mostly meant to describe algorithmic computations. Um, I guess you could argue basic was in there somewhere. I think it's just a little later. Um, and so all of those moved to level up and so they were closer to what you and I might think of as we were trying to write a program and they were focused on different domains for trend. For formula translation, engineering computations, let's say cobalt for business, that kind of thing. </s><s>[INST] Still used today for trend? Probably. Oh [/INST] yeah, cobalt to. But the deal was that once you moved up that level, then you let's call it for trend. You had a language that was not tied to a particular kind of hardware because different compiler would compile for different kind of hardware and that meant two things. It meant you only had to write the program once, which is very important. And it meant that you could in fact, if you were random engineer physicist, whatever you could write that program yourself, you didn't have to hire a programmer to do it for. You might not be as good as you get the real programmer, but it was pretty good and so democratized and made much more broadly available. They build it right code. </s><s>[INST] So it puts the power of programming into the hands of people like you, [/INST] anybody who wants, who is willing to invest some time in learning a programming language and is not then tied to a particular kind of computer. And then in the seventies you get system programming languages of which C is the survivor. And </s><s>[INST] what is system programming languages, programs [/INST] that programming languages that would take on the kinds of things that would necessary to write so called system programs, things like text editors or assemblers or compilers or operating systems themselves, those kinds of things. And </s><s>[INST] fortunately feature rich, they have to be able to do a lot of stuff, a lot of memory management, access processes and all that kind of stuff. They have a process, it's a [/INST] different flavor of what they're doing, they're much more in touch with the actual machine in a but in a positive way that is, you can talk about memory in a more controlled way. Um you can talk about the different data types that the machine supports and underway there, um And more ways to structure and organize data and so the system programming languages, there was a lot of effort in that in the call it the late 60s, early 70's C is I think the only real survivor of that. Um and then what happens after that? Uh, you get things like object oriented programming languages, because as you write programs in a language like C at some point scale gets too and it's too hard to keep track of the pieces and there's no guardrails or training wheels or something like that to prevent you from doing bad things. So, C Plus Plus comes out of that tradition </s><s>[INST] and then it took off from there. I mean, there's also a parallel, slightly parallel track with a little bit of functional stuff with list and so on. But I guess from that point it's just an explosion of languages. Is the java story. There's the javascript, there's all the stuff that the cool kids these days are doing with rust and all that. So, what's to use? You wrote a book, C programming language? What And see is probably one of the most important languages in the history of programming languages if you kind of look at impact, What do you think is the most elegant or powerful part of C? Why did it survive? Why did it have such a long lasting impact? [/INST] I think it found a sweet spot that in of expressiveness. So you can really write things in a pretty natural way and efficiency, which was particularly important when computers were not nearly as powerful as they are today. Put yourself back 50 years, uh almost in terms of what computers could do and that's, you know, roughly four or five generations decades of more Islam. Right. Um so expressiveness and efficiency and I don't know, perhaps the environment that it came with as well, which was UNIX. So it meant if you wrote a program, it could be used on all those computers that ran UNIX. And that was all of those computers because they were all written in C and that way was UNIX. The operating system itself was portable as were all the tools. So all we're together again in one of these things where things fed on each other in a positive cycle. </s><s>[INST] What did it take to write, sort of a definitive book? Probably the definitive book on all of programs, like it's more definitive to a particular language than any other book on any other language and did two really powerful things, which is popularized the language, at least from my perspective, maybe you can correct me And second is created a standard of how, you know, the how this language is supposed to be used and apply. So what did it take? Did you have those kinds of ambitions in mind when we're working on that? [/INST] It's some kind of joke? No, of course not. Uh </s><s>[INST] an accident is an accident of uh of timing skill and just look [/INST] A lot of it is clearly timing was good. Now Dennis and I wrote the book in 1977 on the stretchy, right? Um And at that point UNIX was starting to spread. I don't know how many there were but it would be dozens to hundreds of unique systems. Um And see was also available on other kinds of computers that had nothing to do with UNIX. And so the language had some potential. Um And there were no other books on C. And Bell Labs was really the only source for it. And Dennis of course was authoritative because it was his language. And he had written the reference manual which is a marvelous example of how to write a reference manual. Really really very very well done. So I twisted his arm until he agreed to write a book and then we wrote a book. And the virtue or advantage at least I guess of going first is that then other people have to follow you if they're going to do anything. Um And I think it worked well because Dennis was superb writer. I mean he really, really did. And the reference manual in the book is his period, I had nothing to do with that at all. Um So just crystal clear prose. Very, very well expressed. Um And then he and I I wrote most of the expository material and then he and I sort of did the usual ping ponging back and forth refining it. But I spent a lot of time trying to find examples that would sort of hang together and they would tell people what they might need to know. What about the right time that they should be thinking about needing it. Um And I'm not sure it completely succeeded but it mostly worked out fairly well. </s><s>[INST] What do you think is the power of example? I mean you're you are the creator at least one of the first people to do the Hello World program. Just like the example if aliens discover our civilization hundreds of years from now, it'll probably Hello World programs just like a half broken robot communicating with them with the Hello World. So what and that's a representative example. So what what do you find powerful about examples? [/INST] I think a good example will tell you how to do something and it will be representative of, you might not want to do exactly that but you will want to do something that's at least in that same general vein. Um And so a lot of the examples in the sea book were picked for these very very simple straightforward text processing problems that were typical of UNIX. I want to read input and write it out again. There's a copy command, I want to read input and do something to it and write it out again there's a grab. And so that kind of find things that are representative of what people want to do and spell those out so that they can then take those and see the core parts and modify them to their taste. And I think that a lot of programming books that I don't look at programming books a tremendous amount these days, but when I do a lot of don't do that, they don't give you examples that are both realistic and something you might want to do. Some of them are pure syntax, here's how you add three numbers. Well, come on, I could figure that out. Tell me how I would get those three numbers into the computer and how he would do something useful with them and then how I put them back out again, deeply formatted </s><s>[INST] and especially if you follow the example, there's something magical of doing something that feels useful. Yeah. [/INST] Right. And I think it's the attempt and it's absolutely not perfect. Uh but the attempt in all cases was to get something that was going to be either directly useful or would be very representative of useful things that a programmer might want to do. But within that vein of fundamentally text processing, reading text, doing something, writing text. Yeah. </s><s>[INST] So you've also written a book on goal language. Uh I have to admit. So I worked at google for a while, I've never used go not you miss something well, I know I missed something for sure. I mean so go and rust, the two languages that I hear very uh spoken very highly of and I wish I would like to try. Well there's a lot of them, there's Julia, there's there's all these incredible modern languages, but if you can comment before or maybe comment on, what do you find, where does go sit in in this broad spectrum of languages. And also how do you yourself feel about this wide range of powerful, interesting language is that you may never even get to try to explore because of time. [/INST] So I think so go um go first comes from that same bell labs tradition in part, not exclusively, but two of the three creators, ken, Thompson and rob pike, </s><s>[INST] literally the people, yeah, people. Um [/INST] and then with this very, very useful uh influence from the european school uh in particular the clothes sphere of influence through robert Chrismer. Um it was, I guess the second generation down student uh e th um and so that's an interesting combination of things. And so some ways go captures the good parts of c, It looks sort of like, see it sometimes characterized to see for the 21st century um on the surface, it looks very, very much like c but at the same time it has some interesting data structuring capabilities. And then I think the part that I would Say is particularly useful and again, I'm not ago expert. Um in spite of co authoring the book, about 90% of the work was done by Alan Donovan and my co author who is ago expert. Um But go provides a very nice model of concurrency. It's basically the cooperating uh communicating sequential processes that Tony Hoare uh set forth, Jeez, I don't know, 40 plus years ago. Um and go routines are to my mind a very natural way to talk about um parallel computation. And in the few experiments I've done with them, they're easy to write and typically it's going to work and very efficient as well. So I think that's one place where go stands out that that model of parallel computation is very, very easy and nice </s><s>[INST] just to comment on that, do you think? See foresaw or the early unique states, foresaw threads and massively parallel computation? [/INST] I would guess not really, I mean maybe it was seen but not at the level where it was something you had to do anything about. Um for a long time processors got faster and then processors stopped getting faster because of things like power consumption and heat generation. And so what happened instead was that instead of process is getting faster, there started to be more of them. And that's where that parallel thread stuff comes in. </s><s>[INST] So if you can comment on all the other languages, is it break your heart that you'll never get to explore them. How do you feel about the full variety? It's [/INST] not break my heart, but I would love to be able to try more of these languages. The closest I've come is in class that I often teach in the spring here. Uh, it's a programming class and I often Give, I have one sort of small example that I will write in as many languages as I possibly can. I've got it in 28 languages at this point. And Um, and that's so I do a minimal experiment with a language just to say, Okay, I have this trivial task which I understand the task and it should, it takes 15 lines in och and not much more in a variety of other languages. So how big is it? How fast does it run? And what pain did I go through to learn how to do it. Um and that's a, it's like an IQ data, right? It's very, very, very narrowly </s><s>[INST] data like that. So yeah. But it's still, it's a little sample because you get, I think the hardest step of the programming language is probably the first step. Right? So there you're taking the first step. [/INST] Yeah. And so my experience um with Some languages very positive, like Lua scripting language I've never used. Um and I took my little program, the program is a trivial for matter. It just takes in lines of text of varying lengths and it puts them out in lines that have no more than 60 characters on each line. So I think it was just kind of the flow of process in a browser or something. Um So it's a very short program and um in lua I downloaded lou and in an hour I had it working, never having written you in my life just going with online documentation. I did the same thing in scalar which you can think of as a flavor of java equally trivial. Um I did it in Haskell. It took me several </s><s>[INST] weeks [/INST] But it did run like a turtle. Um and and I did it in FORTRAN 90 and it's painful but it worked and I tried it in rust and it took me several days to get it working because the model of memory management, it was just a little unfamiliar to me and the problem I had with rust and it's back to what we were just talking about. I couldn't find good consistent documentation on rust. Now this was several years ago and I'm sure things have stabilized but at the time everything in the restaurant seemed to be changing rapidly and so you would find what looked like a working example and it wouldn't work with the version of the language that I had. So it took longer than it should have. Um rest is a language I would like to get back to but probably won't. I think one of the issues you have to have something you want to do and if you don't have something that is the right combination if I want to do it and yet I have enough disposable time, whatever to make it worth learning a new language at the same time, it's never gonna happen. </s><s>[INST] So what do you think about another language of javascript? That's this? Well, let me just sort of comment on. And I said what I was brought up sort of javascript was seen as uh the probably like the ugliest language possible and yet it's quite arguably, quite possibly taking over not just the front and the back end of the internet, but possibly in the future taking over everything because they have now learned to make it very efficient. Yeah. So what do you think about this? Yeah, Well I [/INST] think you captured in a lot of ways when it first came out, javascript was deemed to be fairly irregular and an ugly language and certainly in the academy, if you said you were working on javascript, people would ridicule you, it was just not fit for academics to work on. Um, I think a lot of that has evolved, The language itself has evolved and uh, certainly the technology of compiling it is fantastically better than it was. Um, and so in that sense, it's a absolutely a viable solution on back ends as well as the front ends um used, well, I think it's a pretty good language, I've written a modest amount of it and I've played with javascript translators and things like that. I'm not a real expert and it's hard to keep up even there with the new things that come along with it. Um, so I don't know whether it will ever take over the world, I think not, but uh, it's certainly an important language and whereas knowing more about </s><s>[INST] there's uh, there's maybe to get your comment on something which javascript and actually, most languages sort of python, such a big part of the experience of programming with those languages includes libraries, sort of using building on top of the code that other people have built. I think that's probably different from the experience that we just talked about from UNIX and C days when you're building stuff from scratch. What do you think about this world of essentially leveraging, building up libraries on top of each other and leveraging them? [/INST] Yeah, that's a very perceptive kind of question. One of the reasons programming was fun in the old days was that you were really building it all yourself. The number of libraries you had to deal with was quite small. Maybe it was printed after the standard library or something like that. Um, and that is not the case today. And if you want to do something and you mentioned python and javascript and those of the to find examples, you have to typically download a boat load of other stuff and you have no idea what you're getting absolutely nothing. I've been doing some, playing with machine learning over the last couple of days and gee something doesn't work? Well you pip install this okay. And down comes another gazillion megabytes of something and you have no idea what it was. And if you're lucky it works. And if it doesn't work, you have no recourse. There's absolutely no way you could figure out which of these 1000 different packages. And I think it's worse in the M. P. N. N. P. M. Environment for javascript. I think there is less discipline, less control there. </s><s>[INST] And there's aspects of uh not just not understanding how it works, but there's security issues. They're business issues. So you don't want to run a nuclear power plant using Javascript essentially. Probably [/INST] not. </s><s>[INST] So speaking to the variety of languages, do you think that variety is good or do you hope think that over time we should converge towards 123 programming languages that you mentioned to the bell out days when people could sort of the community of it. And the more languages you have, the more you separate the communities, There's the ruby community, there's the python community, there's C++ community. Do you hope that they will unite one day to just one or two languages? [/INST] I certainly don't hope it. I'm not sure that that's right because I honestly don't think there is one language that will suffice for all the programming needs of the world, Are there too many at this point? Well, arguably. Um but I think if you look at that, that's sort of the distribution of how they are used, there's something called a dozen languages that probably account for 95% of all programming at this point. And that doesn't seem unreasonable. Uh, and then there's another, Well, languages that are still in use that nobody uses and or at least don't use in any quantity. But I think new languages are a good idea in many respects, because they're often a chance to explore an idea of how language might help. I think that's one of the positive things about functional languages, for example, they're particularly good place where people have explored ideas that at the time didn't seem feasible, but ultimately have wound up as part of mainstream languages as well. Let me just go back as early as Rikers in lisp and then follow forward functions as first class citizens and pattern based languages and gee, I don't know, closures and just on and on and on lambdas, interesting ideas that showed up first, and let's call it broadly the functional programming community and then find their way into mainstream languages. </s><s>[INST] It's a playground for rebels. [/INST] Yeah, exactly. And, and so I think the languages in the playground themselves are probably not going to be the mainstream, at least for some while, but the ideas that come from, there are invaluable. </s><s>[INST] So let's go to something that when I found out recently, so I've known you've done a million things, but one of the things I wasn't aware of that you had a role in ample. And before you interrupt me by minimizing your role in it, [/INST] which was for minimizing functions. </s><s>[INST] Yeah. Minimizing functions. Right. Exactly. Um uh Can I just say that the elegance and abstraction power of ample is incredible when I first came to it about 10 years ago, or so. can you describe what is the ample language? [/INST] Sure. So ample is a language for mathematical programming? Uh technical term. Think of it as linear programming that is setting up systems of linear equations that are offer some sort of system of constraints so that you have a bunch of things that have to be less than this, greater than that, whatever. And you're trying to find a set of values for some decision variables that will maximize or minimize some objective function. So it's it's a way of solving a particular kind of optimization problem. A very formal sort of optimization problem. But one that's exceptionally useful </s><s>[INST] and specifies. So there's objective function and constraints and variables that becomes separate from the data that operates on. Right? So the that kind of separation allows you to uh you know, put on different hats when put the hat of an optimization person and then put another hat of a data person and dance back and forth. And and also separate the actual uh solvers the optimization systems that do the solving that you can have other people come to the table and then build their solvers whether it's linear or nonlinear uh convex, non convex, that kind of stuff. So of what is the do you as maybe you can comment how you got into that world and what is the beautiful or interesting idea to you from the world of optimization? Sure. [/INST] So I preface it by saying I'm absolutely not an expert on this. And most of the important work in ample comes from why two partners in crime on that bob forer who was a professor of and in the industrial Engineering and Management science department at Northwestern and my colleague at Bell Labs, Dave Gay, who was numerical analyst uh and optimization person. So the deal is linear programming preface this by saying, let's stay with linear program, linear program is the simplest example of this. So linear program is it's taught in school, is that you have a big matrix which is always called A. And you say A X. Is less than or equal to B. So B as a set of constraints. Access the decision variables and a as to how the decision very polls are combined to set up the various constraints. So A as a matrix and X. And be your vectors. Um and then there's an objective function which is just the sum of a bunch of excess and some coefficients on them. And yet that's the thing you want to optimize. Um the problem is that in the real world that that matrix A is a very, very, very intricate, very large and very sparse matrix where the various components of the model are distributed among the coefficients in a way that is totally un obvious to anybody. Um And so what you need is some way to express the original model, which you and I would write, you know, we'd write mathematics on the board, the some of this is greater than the sum of that kind of thing. So you need a language to write those kinds of constraints. And bob for for a long time had been interested in modeling languages, languages that made it possible to do this, there was a modeling language around called Gams, the general algebraic modeling system, but it looks very much like FORTRAN was kind of clunky. Um and so bob spent a sabbatical year at Bell Labs in 1984 and uh he and wasn't in the office across from me and uh it's always geography and uh he and Dave Gay and I started talking about this kind of thing. And um he wanted to design a language that would make it so that you could take these algebraic specifications, you know, summation, signs over sets and that you would write on the board and convert them into basically this a matrix and then pass that off to a solver, which is an entirely separate thing. And so we talked about the design of the language. I don't remember any of the details of this now, but it's kind of an obvious thing, You're just writing out mathematical expressions in a FORTRAN like, or sorry, an algebraic but textual like language. Um and um I wrote the first version of this ample program, uh my first C plus plus program and uh </s><s>[INST] it's written in C plus plus. Yeah. [/INST] And so I did that fairly quickly. We wrote it was, you know, 3000 lines or something, so it wasn't very big, but it sort of showed the feasibility of it that you could actually do something that was easy for people to specify models and converted into something that the solver could work with at the same time as you say, that model and the data are separate things. So one model would then work with all kinds of different data in the same way that lots of programs do the same thing, but with different </s><s>[INST] data. So one of the really nice things is the specification of the models. Human um just kind of like as you say, it's human readable, like I literally remember and stuff, I worked, I would send it to colleagues that I'm pretty sure, never programmed in their life, just just to uh understand what optimization problem is, I think how hard is it to convert that you said you there's a first prototype in people as well as to convert that into something that could actually be used by the solver. [/INST] It's not too bad because most of the solvers have some mechanism that lets them import a model in a form. It might be as simple as the matrix itself in just some representation or if you're doing things that are not linear programming, then there may be some mechanism that you provide things like functions to be called or other constraints on the model. So uh, so all ample does is to generate that kind of thing and then solver deals with all the hard work and then when the solver comes back with numbers, ample converts those back into your original form. So you know how much of each thing you should be buying or making or shipping or whatever. Um, so we did that in 84 and I haven't had a lot to do with it since except that we wrote a couple of versions of the book on which is one of the </s><s>[INST] greatest books ever written. I love that book. I don't know why [/INST] it's an excellent book bob for wrote most of it. And so it's really, really well done. He must spend dynamite teacher </s><s>[INST] and typeset in late check. [/INST] No, no, no. Are you </s><s>[INST] kidding? I remember liking the typography so I don't know [/INST] we did it with the roof. </s><s>[INST] I don't even know what that is. Yeah, [/INST] exactly. You're too </s><s>[INST] young to think [/INST] of tear off as as a A predecessor to the tech family of things. It's a format that was done at Bell Labs in this same period of the very early 70's That predates tech and things like that play 5-10 years. </s><s>[INST] But it was nevertheless just, I'm going by memories. It was, I remember it being beautiful. Uh It was nice outside of UNIX C are going all the things we talked about all the amazing work you've done. You've also done working graph theory. Ah let me ask this this crazy out there question. Uh if you have to make a bet and I had to force you to make a bet, Do you think P equals np? [/INST] Uh The answer is no. Although I'm told that somebody asked Jeff Dean if that was the under what conditions? Beauty Glenn P And he said either P A zero or in his one or vice versa. I've forgotten. This is what Jeff was a lot smarter than </s><s>[INST] I. Uh So, but your intuition is uh I [/INST] have no, I have no intuition. But I've got a lot of colleagues who got intuition and they're betting is no, </s><s>[INST] that's the popular, that's the popular bet. Okay. So what is computational complexity theory? And do you think these kinds of complexity classes especially as you've taught in this modern world are still useful way to understand the hardness of problems. [/INST] I don't do that stuff the last time I touched anything to do with many, many years ago was before it was invented because I it's literally true. I did my PhD thesis on before big </s><s>[INST] o notation. Absolutely. [/INST] Before I did this in 1968. Um and I worked on graph partitioning, which is this question, you got a graph that is a nodes and edges kind of graph and the edges have weights and you just want to divide the nodes into two piles of equal size so that the number of edges that goes from one side to the other is as small as possible. And </s><s>[INST] we developed so that problem is hard. [/INST] Well, as it turns out, I worked with shen lin at bell labs on this and we were never able to come up with anything that was guaranteed to give the right answer. We came up with heuristics that worked pretty darn well. And I peeled off some special cases for my thesis, but it was just hard and that was just about the time that steve cook was showing that there were classes of problems that appeared to be really hard of which graph partitioning was one. But this my expertise such as it was totally predates that development. </s><s>[INST] So the purest which now uh your cares the two of yours names for the traveling salesman problem and for the graph partitioning that was like how did you, you weren't even thinking in terms of classes, you were just trying to was no such idea a heuristic That kind of does the job pretty well. You [/INST] were trying to find a something that did the job and there was no nothing that you would call it a closed form or algorithmic thing. That would give you a guaranteed right answer. I mean, compare graph partitioning to max flow min cut or something like that. That's the same problem. Except there's no constraint on the number of nodes on one side or the other of the cut. And that means it's an easy problem, at least as I understand it. Whereas the constraint that says the two have to be constrained in size makes it a hard problem. </s><s>[INST] Yes. So the robert Frost has that poem where you choose to path. So why did you uh is there another alternate universe in which you pursued the don canoe? The path of you know, algorithm design, sort [/INST] of not smart enough, </s><s>[INST] smart enough. Uh uh You're infinitely uh modest, but so you you proceed your kind of love of programming. I mean when you look back to those, I mean, just looking into that world, does that just seem like a distant world of theoretical computer science, then? Is it fundamentally different from the world of programming? [/INST] I don't know. I mean, certainly in all seriousness, I just didn't have the talent for it. I when I got here as a grad student at Princeton and I started to think about research at the end of my first year or something like that. I worked briefly with john Hop Croft, who is absolutely you know, you mentioned during award winner et cetera, a great guy. Um and it became crystal clear. I was not cut out for this stuff period. Okay. And so I moved into things where I was more cut out for it. And that tended to be things like writing programs and then ultimately writing books. </s><s>[INST] Yeah, you said that in Toronto as an undergrad, you did a senior thesis or literature survey on artificial intelligence. This was 1964, correct. What was the ai landscape ideas, dreams at that time? [/INST] I think that was one of the, what you've heard of ai winters. This is whatever the opposite was a I summer or something. It was one of those things where people thought that boy, we could do anything with computers, that all these hard problems we could, computers will solve them, they will do machine translation, they will play games like chess, that they will do miss uh you know, prove theorems in geometry. There are all kinds of examples like that where people thought, boy, we could really do those sorts of things. Um and you know, I I read the kool aid in some sense, I still have a wonderful collection of papers called computers and thought that was published in about that era and people were very optimistic. And then of course it turned out that What people thought was just a few years down the pike uh was more than a few years down the pike and some parts of that are more or less now sort of under control. Uh we finally do play games like going chess and so on better than than people do. But there are others. Machine translation is a lot better than it used to be, but that's you know, 50 close to 60 years of progress and a lot of evolution and hardware and a tremendous amount more data upon which you can build uh systems that actually can learn from some of </s><s>[INST] that and and the uh the infrastructure to support developers working together like an open source movement. The internet period is also an empowering. But what what lesson do you draw from that the opposite of winter? That optimism? [/INST] Well, I guess the lesson is that in the short run it's pretty easy to be too pessimistic or maybe too optimistic and in the long run you probably shouldn't be too pessimistic. I'm not saying that very well, it reminds me of this uh remark from Arthur Clarke, science fiction author says, you know, when some distinguished but elderly person says that something is in is possible. He's probably right if he says it's impossible, he's almost surely wrong. But you don't know what the time scale </s><s>[INST] is that the time scale is good? All right. So what are your thoughts on this new summer of Ai now in the work with machine learning and neural networks you've kind of mentioned that you started to try to explore and look into this world that seems fundamentally different from the world of heuristics and algorithms. Like search that. It's now purely sort of trying to take huge amounts of data and learn learned from that day to write programs from the data. Yeah. [/INST] Look I think it's very interesting. I am incredibly far from an expert most of what I know I've learned from my students and they're probably disappointed in how little I've learned from them. But um I think it has tremendous potential for certain kinds of things. I mean games is one where it obviously has had an effect on some of the others as well. I think there's um and this is speaking from definitely not expertise. I think there are serious problems in certain kinds of machine learning at least because what they're learning from is the data that we give them and if the data we give them has something wrong with it then what they learned from it is probably wrong to the obvious thing is some kind of bias in the data. The data has stuff in it. Like I don't know, women aren't as good at men as men. It's something okay. That's just flat wrong. But if it's in the data because of historical treatment then that machine learning stuff will propagate that. And that is a serious worry. </s><s>[INST] The positive part of that is what machine learning does is reveal the bias in the data and puts a mirror to our own society and in so doing helps us remove the by you know, helps us work on ourselves but some mirror to ourselves. Yeah, [/INST] that's an optimistic point of view. And if it works that way, that would be absolutely great. And what I don't know is whether it does work that way or whether the the you know, the ai mechanisms or machine learning mechanisms reinforced and amplify things that have been wrong in the past and I don't know. But I think that's a serious thing that we have to be concerned about. </s><s>[INST] Let me ask you another question. Okay. I know nobody knows. But What do you think it takes to build a system of human level intelligence that's been the dream from the 60s. We talk about games about language about about image recognition, but really the dream is to create human level or super human level intelligence. What do you think it takes to do that? And are we close? [/INST] I haven't a clue. And I don't know roughly speaking. I mean this was trying to </s><s>[INST] trick you into a hypothesis. [/INST] Yeah. I mean turing talked about this in his paper on machine intelligence back and she's on an early 50s or something like that. And he had the idea of the turing test and I don't know what the turing test is a </s><s>[INST] good, I don't know, [/INST] it's an interesting test, at least in some vague sense, objective, whether you can read anything into the conclusions is a different story. </s><s>[INST] Do you have worries uh concerns excitement about the future of artificial intelligence? So there's a lot of people who are worried and you can speak broadly than just artificial intelligence is basically computing taking over the world in various forms. Are you excited by this future, this possibility of computing being everywhere or are you worried? [/INST] It's some combination of those. I think almost all technologies over the long run are for good but there's plenty of examples where they haven't been good either over a long run for some people or over a short run. Um And computing is one of those and ai within it is going to be one of those as well. But computing broadly, I mean for just a today example is privacy that um the use of things like social media and so on means that and and the commercial surveillance means that there's an enormous amount more known about us by people, other businesses, government whatever than perhaps one ought to feel comfortable with. So that's an example. Um </s><s>[INST] That's an example of possible negative negative effect of computing being everywhere. It's a it's an interesting one because uh it could also be a positive leverage correctly. [/INST] Uh There's a big gift there. </s><s>[INST] So I I you know I have a deep interest in human psychology and humans there seem to be very paranoid about this data thing? Ah At uh but that varies depending on age group, it seems like the younger folks, so it's exciting to me to see what society looks like 50 years from now that the concerns about privacy might be flipped on their head based purely on human psychology versus actual concerns or not. Uh huh. What do you think about moore's Law? You said a lot of stuff we've talked, you talked about with programming languages in their design and their ideas are coming from the constraints of the systems they operate. And do you think moore's Law? The ah the exponential improvement of systems will continue indefinitely. There's a mix of opinions on that currently, or do you think uh do you think there will be uh Do you think there will be a plateau? [/INST] Well, the frivolous answer, There's no exponential can go on forever. Uh Run out of something? Um Just </s><s>[INST] as we said, time scale matters. So if it goes on long enough, that might be all we needed. Yeah. Right. What matter does [/INST] uh So I don't know, we've seen places where moore's law has changed for example, mentioned earlier, process processors don't get faster anymore, but you use that same growth of, you know, ability to put more things in a given area to go to horizontally instead of vertically classes. Where so you can get more and more processors or memory or whatever on the same chip. Is that going to run into a limitation presumably because, you know, at some point you get down to the individual atoms and so you've got to find some way around that. Will we find some way around that? I don't know. I just said that if I say it won't, I'll be wrong. Perhaps we </s><s>[INST] will. So I just talked to jim Keller and he says he actually describes, he argues that the moore's law will continue for a long, long time. Because you mentioned the atom we actually have I think a thousandfold increase still uh decrease in threatened transistor size. Still possible before we get to the quantum level. So it's there's still a lot of possibilities. So you think he'll continue indefinitely, which is an interesting optimistic, uh optimistic viewpoint. But how do you think the programming languages will change with this increase? Whether we hit a wall or not? What do you think? Do you think there'll be a fundamental change in the way programming languages are designed? [/INST] I don't know about that. I think what will happen is continuation of what we see in some areas at least, which is that um more programming will be done by programs than by people and that more will be done by sort of declarative rather than procedural mechanisms where I say I want this to happen, you figure out how um, and that is in many cases at this point, domain of specialist languages for narrow domains, but you can imagine that broadening out and so I don't have to say so much in so much detail, some collection of software, let's call it languages or programs or something. We'll figure out how to do what I want to do. </s><s>[INST] Some increased levels of abstraction. Yeah. And one day getting to the human level, we're gonna just use that as possible. So, Utah still teach of course computers in our world here at Princeton that introduces computing and programming to non majors. What uh, just from that experience, what advice do you have for people who don't know anything about programming? But I'm kind of curious about this world or programming seems to become more and more of a fundamental skill that people need to be at least aware of. [/INST] Yeah, well, I can recommend a good book. </s><s>[INST] What's that [/INST] book I wrote for the course. I think this is one of these questions of should everybody know how to program. And I think the answer is probably not, but I think everybody should at least understand sort of what it is. So that if you say to somebody I'm a program or they have a notion of what that might be or if you say this is a program or this was decided by a computer running a program that they have some vague intuitive understanding, an accurate understanding of what that might imply. Um so part of what I'm doing in this course, which is very definitely for non technical people. I mean, typical person and it is a history or english major. Um try and explain how computers work, how they do their thing, what programming is, how you write a program um and how computers talk to each other and what do they do when they're talking to each other? And then I would say nobody very rarely. And it does anybody in that course go on to become a real serious programmer. But at least they've got a somewhat better idea of what all this stuff is about, Not just the programming, but the technology behind computers and </s><s>[INST] communications. Do they write a do they try and write a program themselves? Yeah, [/INST] very small amount. I introduced them to how machines work at a level below high level languages. So we have a kind of a toy machine that has a very small repertoire, a dozen instructions. And they write trivial assembly language programs for that, </s><s>[INST] wow. So just if you were to give a flavor to people of the programming world, of the competing world, what are the examples they should go with a little bit of assembly to to get a sense at the lowest level of what the program is really doing? [/INST] Yeah, I mean in some sense there's no such thing as the lowest level because you can keep going down. But that's the place where I drew the line. So the idea that computers have a fairly small repertoire of very simple instructions that they can do, like add and subtract and branch and so on, as you mentioned earlier, um and that you can write code at that level and it will get things done and then you have the levels of abstraction that we get with higher level languages like FORTRAN or C or whatever, and that makes it easier to write the code and less dependent on particular architectures. Um and then we talk about a lot of the different kinds of programs that they use all the time, that they don't probably realize our programs like they're running uh Mac Os on their computers or maybe windows um and they're downloading apps on their phones and all of those things are programs that are just what we just talked about except at a grand scale. </s><s>[INST] It's easy to forget that their actual programs that people program, there's engineers, they wrote, wrote those things right. [/INST] And so In a way, I'm expecting them to make an enormous conceptual leap from there. five or 10 line toI assembly language thing that adds two or three numbers to, you know, something that is a browser on their phone or whatever, but but it's really the same thing. </s><s>[INST] So if you look at the bright and broad strokes in history, what do you think the world like? How do you think the world changed because of computers. Uh It's hard to sometimes see the big picture when you're in it, but I guess I'm asking if there's something you've noticed over the years that uh like you mentioned, the students are more distracted looking at there. Now there's a device to look at. Right? Well, [/INST] I think computing has changed a tremendous amount and obviously, but I think one aspect of that is the way that people interact with each other, Both locally and far away. And when I was, you know, the age of those kids making a phone call to somewhere was a big deal because it costs serious money and this was in the 60s, right? Um and today people don't make phone calls, they send texts or something like that. So it's there's a up and down and what people do, people think nothing of having, okay correspondence, regular meetings, video whatever with friends or family or whatever in any other part of the world. And they don't think about that at all, they and so that's just the communication aspect of it. And </s><s>[INST] do you think that brings us closer together or does it make us uh does it take us away from the closeness of human to human contact? I [/INST] think it depends a lot on all kinds of things. So I trade mail with my brother and sister in Canada much more often than I used to talk to them on the phone. So probably every two or three days I get something or send something to them. Whereas 20 years ago I probably wouldn't have talked to them on the phone. Nearly as much so in that sense that's brought my brother and sister and I closer together. That's a good thing. Um I watched the kids on campus and they're mostly walking around with their heads down fooling with their phones to the point where I have to duck them. I don't know that that has brought them closer together in some ways. There's sociological research that says people are in fact not as close together as they used to be. I don't know where that's really true but I can see potential downsides and kids where you think come on, wake up and smell the coffee or whatever. </s><s>[INST] That's right. But if you look at again nobody can predict the future. But are you excited? Kind of touch this a little bit with ai but Are you excited by the future in the next 10, 20 years? The computing, we'll bring you you were there when there was no computers really? And now computers are everywhere all over the world and africa and asia and just every every person, almost every person in the world has a device. So are you hopeful optimistic about that future? [/INST] Uh It's mixed if the truth be told. I mean I think there are some things about that that are good. I think there's the potential for people to improve their lives all over the place and that's obviously good. Um and at the same time, it at least in the short time short run you can see lots and lots of bad as people become more try ballistic or parochial in their interests and it's an enormous amount more us and them and people are using computers and all kinds of ways to mislead or misrepresent or flat out lie about what's going on and that is affecting politics locally. And I think everywhere in the world, </s><s>[INST] yeah, the long term effect on political systems and so on. Who knows, who knows indeed the people now have a voice which is a powerful thing. People who are oppressed have a voice but also everybody has a voice and the chaos that emerges from that is fascinating to watch. It's [/INST] kind of scary </s><s>[INST] if you can go back and relive a moment in your life one that made you truly happy outside of family or was profoundly transformative. Is there a moment or moments that jump out at you from memory? [/INST] I don't think specific moments, I think there were lots and lots and lots of good times at Bell Labs where you would build something and it worked just, they </s><s>[INST] worked at the moment. We [/INST] used it. Yeah. And and somebody used it and they said gee that's neat, those kinds of things happened quite often in that sort of golden era and the the seventies when UNIX was young and there was all this low hanging fruit and interesting things to work on a group of people who kind of, we were all together in this and if you did something they would try it out for you and I think that was in some sense a really really good time. </s><s>[INST] And AUC was was also an example of that that when you built it and people used it. [/INST] Yeah, absolutely. </s><s>[INST] And now millions of people use you and all [/INST] your stupid mistakes are right there for them to look </s><s>[INST] at. So it's mixed. Yeah, it's terrifying, vulnerable but it's beautiful because it does have a positive impact on so so many people so I think there's no better way to end it brian thank you so much for talking to. It was an honor. [/INST] Okay, my pleasure. Good fun. </s><s>[INST] Thank you for listening to this conversation with Brian Cardigan and thank you to our sponsors eight sleep mattress and Raycon earbuds. Please consider supporting this podcast by going to a sleep dot com slash lex and to buy recon dot com slash lex click the links by the stuff. These both are amazing products. It really is the best way to support this podcast and the journey I'm on. It's how they know I sent you and increases the chance that they'll actually support this podcast in the future If you enjoy this thing. Subscribe my youtube, review it with five stars on apple podcast, support on Patreon or connect with me on twitter. Elex Friedman spelled somehow miraculously without the letter E. Just F R I D M A N because when we immigrated to this country, we were not so good at spelling. And now let me leave you with some words from brian cardigan. Don't comment. Bad code. Rewrite it. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] I think we're going to get into language, we're going to get into communication, intelligence and neural networks and so on. But let me return to the high level the futuristic for for a brief moment. So you've written in your book, your new book, It would be arrogant to suppose that we could forecast where Ai will be where the impact will have in 1000 years or even 500 years. So let me ask you to be arrogant uh what do AI systems with or without physical bodies look like 100 years from now, If you would just uh you can't predict. But if you were to classify eyes and imagine, do can [/INST] I first justify the arrogance before you try to push me beyond it? I mean, there are examples like, you know, people figured out how electricity works. They had no idea that that was going to lead to cell phones, right? I mean, things can move awfully fast once new technologies are perfected even when they made transistors, they weren't really thinking that cell phones would lead to social networking. </s><s>[INST] There are nevertheless predictions of the future which are statistically unlikely to come to be but nevertheless [/INST] is asking me to be wrong. </s><s>[INST] Asking [/INST] you to be way, would I like to be wrong, </s><s>[INST] pick the least unlikely to be wrong thing, even though it's most very likely to be wrong. [/INST] I mean here's some things that we can safely predict. I suppose we can predict that ai will be faster than it is now. It will be cheaper than it is now. It will be better in the sense of being more general and applicable in more places. It will be pervasive. Yeah, I mean, these are easy predictions that I'm sort of modeling them in my head on um Jeff Bezos is famous predictions. He says I can't predict the future. Not in every way I'm paraphrasing but I can predict that people will never want to pay more money for their stuff. They're never going to want to take longer to get there. And you know, so like you can't predict everything, but you can predict something sure. Of course it's going to be faster and better and well, we can't really predict is the full scope of of wary I will be in a certain period. I mean, I think it's safe to say that although I'm very skeptical about current day, I it's possible to do much better. There's no in principle argument that says ai is an insolvable problem that there's magic inside our brains that will never be captured. I mean, I've heard people make those kind of arguments, I don't think they're very good. So ai is going to come and Probably 500 years of planning to get there, and then once it's here it really will change everything. </s><s>[INST] So when you say, yeah, it's going to come. Are you talking about human level intelligence? So maybe I [/INST] like the term general intelligence, so I don't think that the ultimate ai if there is such a thing is going to look just like humans, I think it's going to do some things that humans do better than current machines, like reason flexibly, um and understand language and so forth, but that doesn't mean they have to be identical to humans. So for example, humans have terrible memory and they suffer from what some people call motivated reasoning, so they like arguments that seem to support them and they dismiss arguments that they don't like, there's no reason that a machine should ever do that. </s><s>[INST] So you see that those limitations of memory as a bug, not a feature. [/INST] Absolutely, I'll say two things about that one is I was on a panel with Danny Kahneman, the Nobel Prize winner last night and we were talking about this stuff and I think, you know, we converged on is that the humans are a low bar to exceed. They may be outside of our skill right now, but as you know, ai programmers, but eventually I will exceed it. So we're not talking about human level ai we're talking about general intelligence that can do all kinds of different things and do it without some of the flaws that human beings have. The other thing I'll say is I wrote a whole book actually about the flaws of humans. And it's actually a nice bookend to the counterpoint to the current book. So I wrote a book called Cluj which was about the limits of the human mind. Current book is kind of about those few things that humans do a lot better than machines. </s><s>[INST] Do you think it's possible that the flaws of the human mind the limits of memory, our mortality, our bias is a strength not a weakness. That that is the thing that enables from which motivation springs and meaning springs are [/INST] part of a lot of arguments like this. I've never found them that convincing. I think that there's a lot of making lemonade out of Lemons. So we for example do a lot of free association where one idea just leads to the next and they're not really that well connected and we enjoy that and we make poetry out of it and we make kind of movies with free associations and it's fun or whatever. I don't I don't think that's really a virtue of the system. I think that the limitations and human reasoning actually get us in a lot of trouble. Like for example politically we can't see eye to eye because we have the motivational reason I was talking about and something related called confirmation bias. So we have all these problems that actually make for a rougher society because we can't get along because we can't interpret the data in shared ways and then we do some nice stuff with that. So my free associations are different from yours and you're kind of amused by them and that's great. And hence poetry. So there are lots of ways in which we take a lousy situation and make it good. Another example would be our memories are terrible. So we play games like concentration where you flip over two cards, try to find a pair. You imagine a computer playing that computers like this is the dullest game in the world. I know where all the cards are. I see it once. I know where it is. What are you even talking about when we make a fun game out of having this terrible memory. </s>

<s>[INST] Right? So, I mean, there's certain properties to an ideology that make it harmful. One of them is that, like dogmatism of just certainty, dogged certainty in that you're right, you have the truth [/INST] and nobody else. But what is creating the certainty? It's very interesting to look at the type of model that is being produced is it basically tries to draw a strong prior and you tell people, oh, this idea that you consider to be very true, the evidence that this is actually just much weaker than you thought. And look here, that some studies No, this is not how it works. It's usually normative, which means some thoughts are unthinkable because they would change your identity into something that is no longer acceptable. And this cuts you off from considering an alternative, and many defective religions used this trick to lock people into a certain mode of thought. This removes agency over your own thoughts. It's very ugly to me, it's basically not just a process of domestication, but it's actually an intellectual castration that happens. It's an inability to think creatively and to bring forth new thoughts. </s><s>[INST] Can I ask you about substances, chemical substances that affect the video game the dream world? So psychedelics that increasing have been getting a lot of research done on them. So in general psychedelics, uh psilocybin, M. D. M. A. But also really interesting one, the big one which is D. M. T. What and where are the places that these substances take? The mind that is operating in the dream world? Do you have an interesting sense how this throws a wrinkle into the prediction model? Is it just some weird little quirk, or is there is there some fundamental expansion of the mind going on? [/INST] I suspect that a way to look at psychedelics is that they induce particular types of lucid dreaming states. So it's a state in which certain connections are being serviced in your mind, there are no longer active. Your mind basically gets free to move in a certain direction because some inhibition, some particular innovation doesn't work anymore. And as a result you might stop having yourself or you might stop perceiving the world as three dimensional and you can explore that state. And I suppose that for every state that can be induced with psychedelics, there are people that are naturally in that state. So sometimes psychedelics that shift you through a range of possible mental states and they can also shift you out of the range of permissible mental state. That is where you can make predictive models of reality. And what I observe in people that use psychedelics a lot is that they tend to be overfitting. Overfitting means that you are um using more bits for modeling the dynamics of a function than you should. And so you can fit your curve to extremely detailed things in the past. But this model is no longer predictive for the future. </s><s>[INST] What is it about psychedelics that forces that I thought it would be the opposite. I thought I thought uh it's a good mechanism for uh for generalization, for regularization. So it feels like psychedelics expansion of the mind. Like taking you outside of like forcing your model to be uh non predictive is a good thing, meaning like uh it's almost like, okay, what I would say it's psychedelic psych into is traveling to a totally different environment. Like going if you've never been to like India or something like that from the United States, very different set of people, different culture, different food, different roads and values and all those kinds of [/INST] things. So, psychedelics can for instance, um teleport people into a universe. That is hyperbolic, Which means that if you imagine a room that you're in, you can turn around in the 60° and you didn't go full circle, you need to go seven and 20° to go full circle. Exactly. So the things that people learn in that state cannot be easily transferred in this universe that we are in, it could be that if they are able to abstract and understand what happened to them, that they understand that some part of their spatial cognition has been de synchronized and has found a different synchronization and this different synchronization happens to be a hyperbolic one. Right? So if you learn something interesting about your brain, it's difficult to understand what exactly happened, but we get a pretty good idea once we understand how the brain is representing geometry, </s><s>[INST] but doesn't give you a fresh perspective on the physical reality. Mhm. Who's making that sound is inside my head? Or is it external? [/INST] There is no sound outside of your mind. But uh it's making sense phenomenon. Physics. </s><s>[INST] Uh In the physical reality there's their sound waves traveling through air. Okay, [/INST] that's our model of what happened </s><s>[INST] tomorrow. What happened? Right. Ah That doesn't uh don't psychedelics give you a fresh perspectives on this physical reality. Like uh not this physical reality but this this more, what do you call the dream world that's mapped [/INST] directly to dreaming at night I think is documentation. </s><s>[INST] Exactly. [/INST] So that that's very similar to exchange parameters about the things that you have learned. And uh for instance when you are young, you have seen things from certain perspectives, but not from others. So your brain is generating new perspectives of objects that you already know, which means you can learn to recognize them later from different perspectives. And I suspect that's the reason why many of us remember to have flying dreams as Children, because it's just different perspectives of the world that you already know and that it starts to generate these uh different perspective changes. And then it fluidly turns this into a flying green to make sense of what's happening, right? So you fill in the gaps and certainly you see yourself flying and similar things can happen with semantic relationships. So it's not just special relationships but it can also be the relationships between ideas that are being changed. And it seems that the mechanisms that make that happen during dreaming um are interacting with the same receptors that are being simulated by psychedelics. So uh I suspect that there is a thing that I haven't read really about the way in which dreams are induced in the brain. It's not just that the activity of the brain gets tuned down because you are somehow your eyes are closed and you no longer get enough data from your eyes. But there is a particular type of neurotransmitter that is saturating your brain during these phases, during your emphases and you produce uh controlled hallucinations and psychedelics are linking into these mechanisms I suspect. So </s>

<s>[INST] that's brilliant. So such a subtle but such a fundamental difference. Yeah. Yeah. Goodness me sleep with the problem. Yes, exactly. [/INST] I love french. Um uh so, and we can double click on any one of these and going into detail. But the fourth, I became really enchanted by about eight years ago in our research, which was this idea of forgetting. And I started to think that forgetting maybe the price that we pay for remembering. Mhm. And in that sense, there is an enormous benefit to letting go. And you may be thinking, that sounds ridiculous. I don't want to forget. In fact, my biggest problem is I keep forgetting things, but the brain has a fight. Will we believe has a finite storage capacity? We can't prove it yet. But my suspicion is that that's probably true. It doesn't have an infinite storage capacity, it has constraints if that's the case, we can't simply go through life being constantly informational aggregators Unless, you know, we are programmed to say we've got a hard drive space of about 85-90 years and we're good and we can do that. Maybe that's true. I don't think that's true. I think forgetting is an incredibly good and useful thing. So for example, you know it's not beneficial from an evolutionary perspective for me to remember where I parked my car three years ago. So it's important that I can remember today's parking spot but I don't want to have the junk kind of D. N. A. From a memory perspective of you know, where my I parked my car, you know two years ago. Um Now I actually have a in some ways a problem with forgetting. I'm again I'm not trying to be laudatory but you know I tend not to forget to many things and I don't think that that's a good thing. And um the there was a wonderful neurologist Lauria who wrote a book called The Mind of the norman assist and it was a brilliant book both because it was written exquisitely but he was studying these sort of memory savants who basically could remember everything that he gave them and he tried to find a *** in their armor and the first half of the book is essentially about him seeing how far he can push them before they fail. And he never found that place, he could never find a place where they stopped remembering. And then in his brilliance, he turned the question on its head, he said not what is the benefit of constantly remembering, but instead what is the detriment to never forgetting. And when you start to realize his descriptions of those individuals, it's probably a life that you would not want. </s><s>[INST] It's just fascinating both from a human perspective, but also a i perspective, there's a there's a big challenge in the machine learning community of how to build systems that are able to remember for prolonged periods of time, lifelong continuous learning. So where you build up information over time. So memory is one of the biggest open problems in uh in A I and machine learning, but at the same time the right way to formulate memory is actually forgetting because you have to be exceptionally selective at which kind of stuff you remember. And that's where the step of a simulation integration that you're referring to is really important. I mean we forget most of the things and the question is exactly the cost of forgetting at the very edge of stuff that could be important or could not be, How do we remember or not those things like for example, have you know, doing a podcast, I've become cognizant of one feature of my forgetting that's been problematic, which is I forget names and titles of books and so on. So when I read, I remember ideas, I remember quotes remember statements and like that's the space in which I'm thinking. But when you communicate to others, you have to say this person in this book said that it's the same thing with like Andrew Huberman is masterful at this, This is important academia remembering the authors of the paper and the title of the paper as part of remembering the idea and I've been feeling the cost of not not being able to naturally remember those things and so that that's something I need to sort of work on but that's an example with faces. Yes, very good at faces [/INST] but not good with names. So I'm exactly like you and there is you know an understanding of that in the brain to we understand that there is partitioning of those in terms of the territory of the brain that takes care of faces and facts and places and they can be separate. So I will never forget face but you know, and as I said, I usually forget very little but for some reason, names are struggle, I think in some ways because I'm probably just a slightly anxious person. So when you first meet someone which is usually the time when a name is introduced, you know you were saying you were sort of anxious maybe about sort of sitting down with me, but I find that a little bit, you know activating and so it's not as though there's anything wrong with my memory, it's just the emotional state. I mean when I first meeting someone. Yeah. You know it's a little bit for something but I will never forget that face. But </s>

<s>[INST] I'm a fan of so many people. I love celebrating people, and wearing merch or owning merch is a way to one, celebrate those people. But two, if it's a thing you can wear, you can meet other people, then it starts a conversation, you're like, holy crap, we both like the same thing. Or, holy crap, what is that? I don't know anything about it, tell me about it. And you get to share the things you love with others. It's so awesome. And obviously Shopify is a place you can do that really effectively and make that part of your income a part of your life, a part of your lifestyle, part of your career. So it's awesome that they've created this kind of platform. Sign up for $1 per month trial period at Shopify.com slash Lex. That's all lowercase. Go to Shopify.com slash Lex to take your business [/INST] to the next level today than yours do. </s><s>[INST] But early signs. This episode is also brought to you by ExpressVPN, a VPN provider I've used for many years. Way, way, way before there was a podcast, way before there were a sponsor, there were a source of happiness, a source of security, a source of a kind of blanket protection and help ensure that to whatever degree, at the very basic level, my privacy is protected on the internet. Everybody should be using a VPN. It's in the digital space, it's the first layer of protection. You should definitely be using it. And the one I recommend, the one I've always used is ExpressVPN. There's a lot of other features, like you can watch different shows that are geo-constrained on Netflix and other services. But honestly, to me, the security and the privacy that comes with the VPN is the big plus. And the other point is with ExpressVPN, it's fast and it works everywhere, works on Linux. I don't know why that's such a awesome thing for me. Maybe now it's obvious, but in the early days when they worked on Linux, like early, early days, I don't know how many years ago, but I was like, holy crap, this is so awesome. This is so awesome that they care about this operating system that I love so much. So yeah, it works anywhere. Android, your iPhone, all of that. By the way, I should mention that the peer pressure I have to switch to an iPhone. I do have an iPhone, I just don't use it. The peer pressure is immense, but I remain with the Android. It is the phone of the people. I still like the customization. I like developing for the Android as well. And I have several Android phones and I love them very much. The customization, the freedom, actually the principles that's behind it. But of course, I also love beautiful design. Johnny Ives is one of, if not the greatest tech designers of all time and he's a hero of mine. He's somebody that inspires me as a human being, as a designer. So all of that combined, I don't know. I should be using both actively and giving both love and developing for both. Let's see what the future holds. The point is ExpressVPN works across all of those, no matter what choice I make. Go to expressvpn.com slash LexPod for an extra three months free. This is the Lex Friedman podcast. To support it, please check out our sponsors in the description. And now, dear friends, here's David Backman. Are there interesting differences to you between terms like liberal, Democrat, left-wing, leftist, progressive, socialist, communist, Marxist, far left, center-left, all of these labels. [/INST] Is there interesting distinctions between them? Yeah, there's two sets of distinctions. One is if you just want to say, let's define each of these as political terms. They're all different terms. You can be a progressive ideologically but not be a member of the Democratic party. Many say the Democratic party isn't even really very progressive. So these are certainly terms that we could define in order to have a conversation about the next thing kind of as a precursor to a conversation. Sometimes the terms are used in order to tag someone with a certain ideology that's not really linked to policy or any particular political question, but they can be used positively or negatively to just kind of say, here's the image of this individual that I have in my mind. So like Marxist is right now very popularly being used by some on the right to attack Democrats. There's very few actual Marxists, certainly not in positions of power in the United States but even among the general population. So I think it's important to distinguish are we defining these terms because we want to compare and contrast the ideas that a particular group might bring to the discussion or are we using them as insults or to stifle conversation? There are terms that can be used to start a conversation </s><s>[INST] or to stop it. And the use of those terms is evolving rapidly month by month. So the term leftist, I think is a relatively popular term now to use in the negative context to describe [/INST] what an outraged left wing commentator. I think what you're kind of grasping onto is that there's probably some set of ideas that would apply to most of those who consider themselves to be on the left. The discussion of how that term is mostly being used is not about policy ideas. You're accurately kind of identifying that. And it does seem like progressive is no longer being used as a smear and leftist is being used as a smear </s><s>[INST] more at this point. Okay, but sometimes some of these terms are useful. Like can we try to pick the terms that are useful [/INST] like liberal and progressive and Democrat </s><s>[INST] and liberal and progressive. Is there an interesting definable distinction [/INST] between liberal and progressive to you? That's maybe one of the most interesting ones. 10 years ago, liberal often meant what now we mean by progressive. More recently, the progressive socialist leaning part of the political spectrum has started to use liberal to mean Joe Biden, to mean someone who is not really left enough. So liberal is very interesting because I remember talking with my audience years ago, maybe eight years ago or something like that, where I identified, I'm going to now use the term progressive more commonly to describe my own beliefs because liberal has now been made a smear, it's being shifted into something else and it also means more of like a center left politics. So it's changed in some sense by necessity, by force </s>

<s>[INST] and the common plasma. That's all the same. Before you add [/INST] exactly the rest of the plasma is the same. So it's it's called an epi zero mo reporter assay. Episode means not inside the genome, it's outside the chromosomes. It's an episode. Last thing that allows you to have a variable region where you basically test 10,000 different enhancers and you have a common region which basically has the same reporter gene. You know some condition. Very cool molecular biology. You can basically take the 450,000 elements that you've generated. And you have a piece of the puzzle here, piece of the puzzle here which is identical. So they're compatible with that plasma. You can chop them up in the middle to separate a barcode reporter from the enhancer and in the middle put the same gene again using the same pieces of the bottle. You now can have a barcode readout of what is the impact of 10,000 different versions of an enhancer on gene expression. So we're not doing one experiment, we're doing 10,000 experiments. And those 10,000 can be 5000 of different lo sai and each of them in two versions risk or non risk. I can now test tens of </s><s>[INST] These little hypotheses exactly. And then you can do 10,000 and you can [/INST] Test 10,000 hypothesis at once. </s><s>[INST] How hard is it to generate those 10,000 [/INST] trivial, </s><s>[INST] trivial. But it's biology. [/INST] No no generating the 10,000 is trivial. Because you basically add, it's made by technology. You basically have these arrays that add one include that at the time at every spot. </s><s>[INST] So it's printing and so you're able to you're able to control Uh super costly. Is it $10,000? So this isn't millions of [/INST] dollars for 10,000 experiments. Sounds like the right, You know, I </s><s>[INST] mean, that's super. That's exciting because you don't have to do one thing at a time. [/INST] You can now use that technology, these massively parallel reporter assays to test 10,000 locations at a time. We've made multiple modifications. That technology one was sharper NPR A which stands for you know, basically getting higher resolution view by tiling these uh these elements. So you can see where along the region of control are they acting? And we made another modification called hydra for high, you know, definition regulatory uh notation or something that which basically allows you to test seven million of these at the time by cutting them directly from the DNA. So instead of synthesizing which basically has the limit of 450,000 that you can synthesize at the time, basically said, hey if we want to test all accessible regions of the genome. Let's just do an experiment that cuts accessible regions. Let's take those accessible regions. Put them all with the same end joints of the puzzles and then now use those to create a match, much, much much larger uh array of things that you can test and then tiling all of these regions. You can then pinpoint what are the driver nucleotides, What are the elements, how are the acting across seven million experiment at the time. So basically this is all the same family of technology where you're basically using these parallel readouts of the barcodes and then you know to to do this. We used a technology called Star Sick for self transcribing reporter assays. Technology developed by Alex Stark, my former post dog who is now a P. I. Over in Vienna. So we basically coupled the Star Sick the self transcribing uh reporters where the enhancer can be part of the gene itself. So instead of having a separate barcode, that enhancer basically act to turn on the gene and is transcribed as part of the genes, you have to have the two separate exactly. So you can just read them. So there's a constant </s><s>[INST] improvements in this whole process by the way, generating all these options are is it basically brute force? How much human intuition [/INST] Is? Oh gosh of course it's human intuition and human creativity and incorporating all of the input datasets. Because again the the genome is enormous 3.2 billion. You don't want to test that. Instead you basically use all of these tools that we've talked about already. You generate your top favorite 10,000 hypothesis And then you go and ask all 10,000 and then from what comes out you can then go go the next step. So that's technology. Number two technology want. Number one is robotics automation where you have thousands of wells and you constantly test them. The second technology is instead of having wells, you have these massively parallel readouts in sort of these pooled asses. The 3rd technology is cobbling CRISPR perturbations with these single cell RNA readouts. So let me make another parentheses here to describe. Now single cell RNA sequencing. Okay, so what a single cell RNA sequencing me. So are any sequencing is what has been traditionally used? Well traditionally in the last 20 years uh ever since the advent of next generation sequencing. So basically before RNA expression profiling was based on this micro race. The next technology after that was based on sequencing. So you chop up your RNA and you just sequence small molecules just like you would sequence the genome, basically reverse transcribed the small RNA into DNA. And you sequenced that DNA in order to get the number of sequencing reads corresponding to the expression level of every gene in the genome. You now have are any sequencing. How do you go to single cell RNA sequencing? That technology also went through stages of evolution. The first was micro fluid. It's you basically had these even even chambers you basically have these ways of isolating individual cells, putting them into a well for every one of these cells. So you have 384 well plate and you know, do 304 parallel reactions to measure the expression of 304 cells. That sounds amazing and it was amazing. But we want to do A million sales. How do you go from these wells? two million sales? You can't So what the next technology was after that is instead of using a well for every reaction, you now use a lipid droplet for every reaction. So you use micro droplets as reaction chambers to basically amplify RNA. So here's the idea you basically have micro fluid. It's where you basically have every single cell coming down one tube in your microphone edicts and you have little bubbles getting created in the other way with specific primers that mark every cell with its own barcode. You basically couple the two and you end up with little bubbles that have a cell and tons of markers for that cell. You now mark up all of the RNA for that one cell with the same exact barcode and you then lies all of the droplets and you sequence the heck out of that and you have for every RNA molecule, a unique identifier that tells you what cell was it on? That is such good engineering. </s>

<s>[INST] So fundamentally. What is the problem with the state of [/INST] existence? </s><s>[INST] Okay. Well but but uh to play angels advocate again, you know government is the people. [/INST] Come on. You don't you don't do you really think this </s><s>[INST] as best? I think it's possible to have represent representation. [/INST] Can you imagine if like you have an attorney, you're like oh you can't have the attorney you want. You're gonna have this guy who you absolutely hate, who you share no values with. Why? </s><s>[INST] Because he drives. I mean leaders, political leaders and political representation drive the discourse like we you know, uh the majority of people voted for him or whatever. However however you define that and now we get to have a discussion while was this the right choice and then we'll get to make that choice again in four years and so on. [/INST] First of all the fact that I have to be under the thumb of somebody four years makes no sense. There's no other relationship that's like this including marriage. You can leave any other relationship at any time. Number one. Number two. Is </s><s>[INST] it always impeach But they did that part of it. I'm just saying that the mechanisms are flawed in many ways. [/INST] Yeah. Right. And and so that's number one, Number two is it doesn't make sense that if I don't want someone to represent me that because that person is popular that they are now in a position to. So having uh representation and having citizenship based on geography is a pre landline technology in a post cell phone world. There's no reason why I have to just because we're physically between two oceans, we all have to be represented by the same people. Whereas I can very easily have my security be under someone and switch it as easily as cell phone providers. </s><s>[INST] So okay, but it doesn't have to be geographical. It can be ideas. Sure. This country represents a certain set of ideas. It started out geographically, it's still geographic. [/INST] Start off as ideas as </s><s>[INST] well. But like there's a it was intricately. I mean that's the way humans are is there's I mean uh there was no internet. So it was your geographically the same location and you signed a bunch of documents and then you kind of debated and you wrote a bunch of stuff and then you agreed on it. [/INST] Okay. You understand that no one signed these documents and no one agreed to it as Lysander spooner pointed out over 100 50 years ago. The constitution or the social contract if anything is only binding to the signatories. Um and even then they're all long dead. Uh so it's it's this fallacy that somehow because I'm in a physical place, I have agreed, even though I'm screaming to your face that I don't agree to be subordinate to, uh, some imaginary invisible monster that was created 100 50 years ago. Uh, and this idea of like if you don't like it, you have to move. That's not what freedom means. Freedom means, I do what I want, not what you want. So if you don't like it, you move. </s><s>[INST] Okay. Just to put some I don't like words in [/INST] terms 11101110. Yeah, exactly what your language like </s><s>[INST] it is translating it all in real time. But would you call the kind of ideas that you're advocating for? And we're talking about anarchy? Yes, anarchism. Yes. Okay, so let's get into it. Can you, can you try to paint the utopia that an anarchist worldview dreams about [/INST] the only people who describe anarchism as Utopia or its critics? If I told you right now, And I wish I could say this factually that I have a cure for cancer that would not make us utopia. Uh, that would still probably be expensive. We would still have many other diseases. However, we would be fundamentally healthier, happier and better off all, all of us </s><s>[INST] then democracy. So that democracy, I jumped back from the cancer [/INST] then democracy or or government? So it's only curing one major major life threatening problem, but in no sense? Is it a utopia? </s><s>[INST] So what can we try to uh, answer this question? Same question many times. Which is what exactly is the problem with democracy? [/INST] The problem with democracy is that those who need leaders are not qualified to choose them. </s><s>[INST] Those who need leaders are not qualified to choose them. So that's [/INST] the central problem of democracy. </s><s>[INST] Not all of us need leaders, right? What does it mean to need a leader? Are you saying like people who are actually like free thinkers don't need leaders kind of thing? Sure. But like but like you don't. Okay, so do you acknowledge that there's some value in authority in different subjects? So what that means is I don't mean authority. Somebody who's in control of you, but [/INST] you're doing the definition switch </s><s>[INST] because I am I [/INST] am. You're right. You're right. </s><s>[INST] It's unfair. Okay. That's that's bad. But [/INST] that's what they do. That's their trick. And if this is one of the useful things, by the way, this is total sidebar. If people ask me for advice, I always tell them if you're going to raise your kids, raise them bilingual because I was trilingual by the time I was six and that teaches you to think in concepts. Whereas if you only know one language you fall for things like this because using authority in the sense of a policeman and someone in authority in physics, it's the same word conceptually. They're extremely different. But if you're only thinking in one language your brain is going to equate the two and that's a trap that people who only speak one language have. </s><s>[INST] But even if you know multiple languages, you can still use the trick of usually or see your convenience to manipulate the conversation philosophy. [/INST] You felt I accidentally did it. Yeah, we all tend to do that if you only speak one language and thinking one language. </s>

<s>[INST] well the barbecue and the people are worth [/INST] it now. They are, they're awesome. I was down here for some storm relief a few years ago. Um, and I tell you what, I fell in love with that. The people are great. It's a great state and uh yeah, I'll definitely, definitely be back again for sure. </s><s>[INST] Thanks for talking. [/INST] Thank you sir. Appreciate it. </s><s>[INST] Thanks for listening to this conversation with Niels [/INST] jorgensen. </s><s>[INST] To support this podcast. Please check out our sponsors in the description and now let me leave you some words from franklin. D Roosevelt human kindness has never weakened the stamina or softened the fiber of a free people. A nation does not have to be cruel to be tough. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] in dyke nor mr. con had any involvement in any misconduct by mr. Epstein of any kind at any time the executives the executors lawyer said in a statement it is enormously regrettable that the Attorney General chose to level false allegations and to unfairly malign the co-executors reputation without any proof or factual basis um what no proof has been provided um and what reputation by the way these two dudes have the same reputation as insert any scumbag here and that's the sort of reputational power that these people have if we were doing a social credit score they would have a negative 100 so what do you mean their reputation why is his name all over the Epstein stuff can you answer that for me wise con signature everywhere wise in dike signature everywhere why were they facilitating sham marriages and the real question is why have they not been charged yet if you'd like to contact me you can do that at Bobby kapuchi at protonmail.com that's B-O-B-B-Y-C-A-P-U-C-C-I at protonmail.com you can also find me on Twitter at B-O-B-B-Y underscore C-A-P-U-C-C-I all of the links that go with this episode can be found in the description box [/INST] </s>

<s>[INST] Can I ask you about blind spots? When you reduce stuff to numbers, the cost and benefits, is there stuff you might miss that are important to the flourishing of the human species? [/INST] So everyone will have to say, of course there must be blind spots. But I don't know what they are. Yeah, I'm sure Andy would probably be better at telling me what they are. So we try to incorporate all of it, but obviously we're not successful. You can't incorporate everything, for instance, in the cost benefit analysis. But the point is in some way, I would worry a lot about this if we were sort of close to perfection human race, we're doing almost everything right, but we're not quite right, then we need to get the last digits right. But I think it's much more the point that I tried to make before, that we're all focused on going to an electric car or something else rather than fracking. We're all focused on cutting carbon emissions instead of reducing vulnerability. So we're similarly getting orders of magnitude wrong. And while I'm sure I have blind spots, I think they're probably not big enough to overturn that </s><s>[INST] point. Andy Wise, I don't know what they are, Bjorn and economists are all wrong about everything. [/INST] Well, the models, we could spend a whole day on models. There are economic models, there's this thing called optimization models. There were two big ones used to assess the US plan, this new big IRA, inflation reduction package. And they're fine, they're a starting point for understanding what's possible. But as this gets to the journalism part or the public part, you have to look at the caveats. You have to look at what model... Economists expressly exclude things that are not modelable. And if you look in the fine print on the repeat project, </s><s>[INST] the Princeton version of the assessment of the recent giant legislation the whole day on models, [/INST] the fine print is the front page for me as a deep diving journalist, because it says we didn't include any sources of friction, meaning resistance to putting new transmission lines through your community or people who don't want mining in America, because we've exported all of our mining, we've mined our cobalt in Congo. And trying to get a new mine in Nevada was a fraught fight that took more than 10 years for lithium. So if you're excluding those elements from your model, which on the surface makes this $370 billion package have an emissions reduction trajectory, that's really pretty good. And you're not saying in your first line, by the way, these are the things we're not considering. That's the job of a journalist. </s><s>[INST] Summarize all of human history with that one word friction. [/INST] Well, inertia, friction implies there's a force that's already being resisted, but there's also inertia, which is a huge part of our... We have a status quo bias. The scientists that I... In grappling with the climate problem as a journalist, I paid too much attention to climate scientists. That's why all my articles focused on climate change. And it was 2006. I remember now pretty clearly, I was asked by the weekend review section of the New York Times to write a sort of a weekend thumb sucker, we call them on... Thumb sucker? You know, you sit and suck your thumb and think about something. Why is everybody so pissed off about climate change? It was after the Al Gore movie came out in Convena Truth, Hurricane Katrina's big Senator Inhofe in the Senate from Oklahoma wasn't yet throwing snowballs, but it was close to that. And so I looked into what was going on. Why is this so heated? In 2006, the story is called Yelling Fire on Hot Planet. And that was the first time... This is after 18 years of writing about global warming. That was the first time I interviewed a social scientist, not a climate scientist. Her name was Helen Ingram. She's the UC Irvine. And she laid out for me the factors that determine why people vote, what they vote for, what they think about politically. And they were the antithesis of the climate problem. She used the words, she said, people go into the voting booth thinking about things that are soon, salient, and certain. And climate change is complex, has long timescales. And that really jogged me. And then between 2006, 2010, I started interviewing other social scientists and this was by far the scariest science of all. It's the climate in our heads or inconvenient minds. And in how that translates into political norms and stuff really became the monster, </s><s>[INST] not the climate system. Is there social dynamics to the scientists themselves? Because I've gotten to witness a flocking behavior with scientists. So it's almost like a flock of birds. Within the flock, there's a lot of disagreement and fun debates and everybody trying to prove each other wrong, but they're all headed in the same direction. And you don't want to be the bird that leaves that flock. So there's an idea that science as a mechanism will get us towards the truth, but it'll definitely get us somewhere. But it could be not the truth in the short term. In the long term, a bigger flock will come along and it will get us to the truth. But there's a sense that I don't know if there's a mechanism within science to snap out of it if you're down the wrong track. Usually you get it right, but sometimes you don't. When you don't, [/INST] it's very costly. And there's so many factors that line up to perpetuate that flocking behavior. One is media attention comes in. The other is funding comes in. The National Science Foundation or whatever, European foundations pour a huge amount of money into things related to climate. And then your narrative in your head is shaped by that aspect of the climate problem that's in the spotlight. I started using this hashtag a few years back, narrative capture. Be wary of narrative capture where you're on a train and everyone's getting on the train. And this is in the media too, not just science. It becomes self-sustaining and contrary indications are ignored or down played. No one does replication science because your career doesn't advance through replicating someone else's work. So those contrary indications are not necessarily really dug in on. And this is way beyond climate. This is of many fields. As you said, you might have seen this in AI. And it's another form of path dependency, the term I used before. The breaking narrative capture to me in AI for me has come mostly from stepping back and reminding myself of the basic principles of journalism. Journalism's basic principles are useful for anybody confronting a big, enormous, dynamic, complex thing is who, what, where, when, why. Just be really rigorous about not assuming because there's a fire in Boulder County or a flood in Fort Myers that climate, which is in your head because you're part of the climate team at the New York Times or </s>

<s>[INST] Yeah, I mean there, but I'm sure there's contradictory information coming in from the intelligence apparatus. So it's difficult. I mean, first of all, nobody likes to be disagreed with, especially when you get become more and more powerful and then the intelligence apparatus is probably giving information you don't like. But it's often negative information about basically information that says that the decisions you made in the past are not great decisions and that's the difficult truth to deal with. So there, you know, in the modern times, if we hop around briefly is Vladimir Putin has been um not happy with the intelligence of the Fsb thereby at least if you read the news, choosing to put more priority to the G R. U. For the intelligence in Ukraine. Um but I guess I suppose the same story happens there as a dis throughout history as a paranoia. [/INST] I give you an example that that comes from a very reliable source. Uh and that my my best german friend worked as a chemist in the in the Stasi East german intelligence and </s><s>[INST] he [/INST] eventually he rose to the rank of major and was in charge of the forgery department, it's very likely that he made passports, that I used to travel. He was aware that there was intelligence that that was that was collected. The Stasi was really good. They had about 1000 people in West Germany undercover agents. Uh some of them in government and the Central committee of of the party. The decision makers ignored it because it didn't quite fit in their worldview. It didn't quite fit into their plans. So and and 11 delicious uh thing that I just want to add on to this when when Gorbachev um wrote his book about historical and glasnost, the the East german rulers did not like it. They were much, much more orthodox. So they had to print the books in translation, guess where they wound up. They were piled up in the hallways of the Stasi. They bought the entire print run. </s><s>[INST] It's fascinating. Uh so, but let's backtrack. So Operation Barbarossa, the invasion of the soviet union, and then hopefully that leads us all the way to the east Germany west Germany after the end of the war. [/INST] So what happened was the soviet union rolled into the eastern part of Germany and the Western allies took a larger chunk, Which was eventually it was occupied by the three allies, the french and the the english and the americans. And the eastern part was occupied by the by soviet troops. And the soviet troops actually uh conquered Berlin. But as a and as in a in a contract, they decided that Berlin would be ruled by the four allies. And they all had, you know, had free access to uh that city. I was born in the east german part, which very quickly became ruled by Communists, Socialists, the Communist Party and the Socialist Party United. And but the leaders of that new party were all communists. </s><s>[INST] It's nevertheless called Democratic, [/INST] Yes german Democratic Republic, which uh was formed a couple of months after I was born, I was born into a the remote southeastern corner of of East Germany. And uh interestingly enough, genetically, I'm only half german </s><s>[INST] the other half [/INST] the other half is Czech and polish because and where I grew up? You know, we could walk to the nicer river which was the uh border with Poland and and it was only about an hour by bus to get to the check border. So that's why I'm a mix. </s><s>[INST] So okay, so East Germany after the war was communists. Socialists in the west, Germany was representing the western world with democracy [/INST] and what the United States did. What this was really really very uh forward looking, very strategic. The marshall plan to rebuild the economy in the West as compared to what the soviet union did. Whatever they hadn't destroyed on the way in. They took with them on the way out for reparations because you know, they had every right to do that. But it was not a good idea because you know East Germany was always behind an economic development to to the western counterpart. </s><s>[INST] So when you were young as today, but when you were young you were clearly an exceptional student. You're a brilliant academic superstar. Let's go to your childhood. What's a fond memory from childhood that you have and being woken up to the beauty of this world and sort of being curious about all the mysteries around you that I think ultimately lead to academic um success [/INST] or </s><s>[INST] was it the [/INST] the fondest memory that comes to mind is my first kiss. </s><s>[INST] How's that do you want to go to the details of that? What uh what what what what do you make of that? What do you make of that? Guess what what what that teach you about yourself and human nature and all that. [/INST] It taught me only in hindsight at the time, I was just like, My God, I was head over heels in love, I was 16 years old and I knew in those days I admired girls. I I knew the girls were like sort of magical beings, they were not capable of doing evil things, they were beautiful and they had to be adored. And one of them actually loved me too. She came after me initially, right? </s><s>[INST] And that was like, that was that too, was magical for you? [/INST] Oh my God. </s><s>[INST] Yeah. [/INST] Uh and I literally, I uh dedicated, that's when I started studying up until that point, I just like did whatever I had to do to be in a minus students. And that's when I started studying in every a that I got I dedicated to her sometimes, explicitly because I knew I was going to take care of her. You know, when as I grew </s>

<s>[INST] and then the robot just does everything in between. And it does everything, all the coordinated movement of all the different actuators that includes the body [/INST] and all of the body and all of the- Yeah, maintains its balance. It walks itself into position so it's within reach and the arm is in a position where it can do the whole task. </s><s>[INST] So it manages the whole body. So how does one become a big enough customer to request features? Because I personally want a robot that gets me a beer. I mean, that has to be like one of the most requests, I suppose, in the industrial setting. That's a non-alcoholic beverage of picking up objects [/INST] and bringing the objects to you. We love working with customers who have challenging problems like this. And this one in particular, because we felt like what they were doing, A, it was a safety feature, B, we saw that the robot could do it because they teleoperated it the first time. Probably took them an hour to do it the first time, right? But the robot was clearly capable. And we thought, oh, this is a great problem for us to work on to figure out how to automate a manipulation task. And so we took it on not because we were gonna make a bunch of money from it in selling the robot back to them, but because it motivated us to go solve what we saw as the next logical step. But many of our customers, in fact, we try to, our bigger customers, typically ones who are gonna run a utility or a factory or something like that, we take that kind of direction from them. And if they're, especially if they're gonna buy 10 or 20 or 30 robots, and they say, I really need it to do this, well, that's exactly the right kind of problem that we wanna be working on. </s><s>[INST] And so- Note to self, by tense thoughts, and aggressively pushed for beer manipulation, I think it's fair to say it's notoriously difficult to make a lot of money as a robotics company. How can you make money as a robotics company? Can you speak to that? It seems that a lot of robotics companies fail. It's difficult to build robots. It's difficult to build robots at a low enough cost where customers, even the industrial setting, want to purchase them. And it's difficult to build robots that are useful, sufficiently useful. So what can you speak to? And Boston Dynamics has been successful [/INST] for many years of finding a way to make money. Well, in the early days, of course, the money we made was from doing contract R&D work. And we made money, but we weren't growing and we weren't selling a product. And then we went through several owners who had a vision of not only developing advanced technology, but eventually developing products. And so both Google and SoftBank and now Hyundai had that vision and were willing to provide that investment. Now, our discipline is that we need to go find applications that are broad enough that you could imagine selling thousands of robots because it doesn't work if you don't sell thousands or tens of thousands of robots. If you only sell hundreds, you will commercially fail. And that's where most of the small robot companies have died. And that's a challenge because, you know, A, you need to field the robots. They need to start to become reliable. And as we've said, that takes time and investment to get there. And so it really does take visionary investment to get there. But we believe that we are going to make money in this industrial monitoring space because, you know, if a chip fab, if the line goes down because a vacuum pump failed someplace, that can be in a very expensive process. It can be a million dollars a day in lost production. Maybe you have to throw away some of the product along the way. And so the robot, if you can prevent that by inspecting the factory every single day, maybe every hour, if you have to, there's a real return on investment there. But there needs to be a critical mass of this task. And we're focusing on a few that we believe are ubiquitous in the industrial production environment. And that's using a thermal camera to keep things from overheating, using an acoustic imager to find compressed air leaks, using visual cameras to read gauges, measuring vibration. These are standard things that you do to prevent unintended shutdown of a factory. And this takes place in a beer factory where we're working with AB InBev. It takes place in chip fabs. We're working with global foundries. It takes place in electric utilities and nuclear power plants. And so the same robot can be applied in all of these industries. And as I said, we have about, actually it's 1100 spots out now to really get profitability. We need to be at 1000 a year, maybe 1500 a year for that sort of part of the business. So it still needs to grow, but we're on a good path. </s><s>[INST] So I think that's totally achievable. So the application should require crossing that 1000 robot barrier. [/INST] It really should. 1000 robot. </s><s>[INST] I want to mention our second robot, Stretch. Yeah, tell me about Stretch. What's Stretch? [/INST] Who's Stretch? Stretch started differently than Spot. Spot we built because we had decades of experience building quadrupeds. We just, we had it in our blood. We had to build a quadruped product, but we had to go figure out what the application was. And we actually discovered this factory patrol application, basically preventative maintenance by seeing what our customers did with it. Stretch is very different. We started knowing that there was warehouses all over the world. There's shipping containers moving all around the world full of boxes that are mostly being moved by hand. By some estimates, we think there's a trillion boxes, cardboard boxes shipped around the world each year. And a lot of it's done manually. It became clear early on that there was an opportunity for a mobile robot in here to move boxes around. And the commercial experience has been very different between Stretch and with Spot. As soon as we started talking to people, potential customers about what Stretch was going to be used for, they immediately started saying, oh, I'll buy, I'll buy that robot. You know, in fact, I'm going to put in an order for 20 right now. We just started shipping the robot in January after several years of development. This year. This year. So our first deliveries of Stretch to customers were DHL and Marisk in January. We're delivering the gap right now. And we have about seven or eight other customers, all who've already agreed in advance to buy between 10 and 20 robots. And so we've already got commitments for a couple hundred of these robots. This one's going to go, right? It's so obvious that there's a need and we're not just going to unload trucks. We're going to do any box moving task in the warehouse. And so it too will be a multi-purpose robot and we'll eventually have it doing palletizing or depalletizing or loading trucks or unloading trucks. There's definitely thousands of robots. There's probably tens of thousands of robots of this in the future. So it's going to be profitable. </s>

<s>[INST] And the core to you making a good effective decision on what you should do with your life, lifestyle, health, diet, whatever, is to ground that in data. So the core to the allergy thing is data. The core to everything inside tracker does is data. I'm really excited about personalized data, driven advice for how to make your life better. Get special savings for a limited time when you go to inside tracker.com slash Lex. This is a Lex Friedman podcast to support it. Please check out our sponsors in the description. And now, dear friends, here's Chris Voss. What is it like negotiating for a hostage with a kidnapper? [/INST] What is the toughest part of that process? The toughest part is if it looks bad from the beginning and you gotta engage in a process anyway. </s><s>[INST] What are the factors that make it bad? That makes you nervous that if you were to observe a situation where there's general negotiation or it's a hostage negotiation, [/INST] what makes you think that this is going to be difficult? If they wanna make it look like they're negotiating, but they're not. Like in the 2004 timeframe, Al Qaeda in Iraq was executing people on camera for the publicity. And they wanted to make it look like they were negotiating. So they'd come on and they say, if you don't get all the women out of, Iraqi women out of the jails in Iraq in 72 hours, we're gonna kill a hostage. That was one of the demands in one of the cases in that timeframe. Now, first of all, even if we'd have been willing, the US government coalition would have been willing to do that, it wouldn't have been able to happen in 72 hours. So is it an impossible ask from the beginning? And so then that looks really bad. Like they're trying to make it look like they're talking reasonably, but they're not. So your hostage is in bad shape there. If they've made a demand that you just, even if you wanted to do, you couldn't do. So then what makes that very difficult is, </s><s>[INST] in kidnappings especially, [/INST] but they're not, you're working with family members, you're coaching people. Bad guys are in touch with family members, or if they're not directly in touch with family members, the other thing that Al Qaeda was doing at that time was, they didn't give us a way to talk to them. They're making statements in the media, but then not leaving their phone number, if you will. So that's one more thing. They're intentionally blocking you. They're asking you to do something you can't do. They're not giving you a way to talk to them. So you gotta get with the family and discuss with the family how you're gonna approach things. Now the family definitely wants to know, is this gonna help? So a bunch of cases like that in that timeframe. And you gotta be honest with them. It's a long shot. Our chances here are slim and none. And when it's slim and none, I'll take slim, but it's still very, very slim. And there were a number of people that were killed in that timeframe before the tide finally got turned. </s><s>[INST] And it was hard dealing with families at the time. Can you negotiate in public versus like a direct channel in private? [/INST] Oh yeah. Bad guys pick the media. They're making statements in the media. And that's a big clue. Their channel of choice tells you an awful lot. And if they're choosing the media, then that means people are trying to appeal to, that means there, in their view, there's such a thing is good media. So if there's good media, there's bad media. How do you make it bad? And we made a bad form. That just, unfortunately, it had to go through a number of iterations </s><s>[INST] before they got the message and quit. In that negotiation, do you think about the value of human life? Is there a dollar figure? Is there, how do you enumerate, not enumerate, quantify the value of human life? [/INST] Yeah, that's like beauty, it's any other beholder. So that was the first lesson on any hostage negotiation, really any negotiation. Like it doesn't matter what it is to you, matters what it is to the other side. One of the things, especially in your conversation I listened to with Andrew. By the way, you guys, another thing I really liked about that conversation, first of all, I think the world of him. Andrew Huberman. Yeah, Andrew Huberman. And he released it on my birthday. I appreciate that. It was a nice birthday for me. Tried to tell him perfectly just for you, yeah. Yeah, nice job, thank you. But empathy is in the eye of the beholder, in every negotiation, whether it's over a car, a house, collaboration in your company with the bad guys. How does the other side see it? Now, the nice thing about kidnapping for ransom, if there's an actual ransom demand, </s><s>[INST] an actual demand, there's a nice birthday present to me. [/INST] Perfectly just for you, yeah. It's a mercenary's business. They're gonna take what they could get. And they tend to be really good at figuring out how much money somebody has. So, and again, I'll keep drawing business analogies. You're looking for a job with an employer. There's a market price of the job, and then there's what the employer can pay you. Now, maybe the market price of the job, the market's 150 grand. Employer can pay 120, but it's a great job. We were talking about Elon a minute ago. I'd work minimum wage to follow him around. That would be worth it. What are the value other than the dollars? And how hard is it to get the dollars? And how quickly can you get to them? Those are all things that the bad guys are good, and kidnapping are good at figuring out. So, the value of human life to them is gonna be what can they get. A crazy thing in the kidnap business. We used to get asked by FBI leadership, when is this gonna be over? And the answer would be, when the bad guys feel like they've gotten everything they can, by dissecting that statement, you're talking about when they feel like they got everything they can. So, the key to kidnapping negotiations are the feelings. Are the bad guys, we're talking about feelings! </s>

<s>[INST] you know, with human life, including often our own. So you think this, there's an underlying creative force to the universe that might even have a kind of built-in morality to it, where creating is better than destroying. And then that somehow maps on onto our society, where we kind of try to figure out what that actually means in terms of good and evil. So there, that's something is there like that. It's like, but it has to be, it's like so nice. It's so perfect, because it's rare. It's sufficiently rare when we have our own space. Like you can like close the door and they're like, I need to be alone right now as our human civilization to work on my thing. So it's sufficiently rare that there's not other alien civilizations. They're just like constantly knocking on our door, destroying us, but it still exists. [/INST] That's weird. Right, right. It's so fantastically improbable that I think we should be very respectful of it. And I think you said there's a creative force that values creativity. This would be, well, sure, it's a creative force, right? </s><s>[INST] It's existence, its ability to exist and to create, [/INST] it's so fanta comes from something other than entropy, right, something other than so much dispersion that there's nothingness, right. So the creative force will value the sanctity of things, right, you know, keeping things together, not destroying things, right? Building novelty, including novelty of knowledge, novelty of tension. I mean, it fits with the idea that we're not nothing, that that's incredibly improbable. and that there are these many, many layers of emergence that we're standing upon. And I think it tells us something that we're not doing ourselves a service to ignore, right? It's not just a jump to saying, oh, there's a religious answer to everything. It's just, no, it's saying science isn't a God either, right? So if we think of science as a tool and not as an endpoint in and of itself, what is the science telling us? Like, I remember showing up at medical school and it really is true. I mean, I knew so little about the human body. I'd only been in hospitals to visit people. I'd taken pre-med classes, but sort of intensely at once after I didn't take any and I was working in business. I knew next to nothing. And I had this idea that was so naive in retrospect that I was gonna learn so much, right? I was gonna answer these questions because I was gonna know what's going on in the body. What are these organs doing? What are these cells? And what I learned was there was so much more that was amazing and mysterious and seemingly impossible. Like even how a cell functions, right? Like what is going on inside of a cell? The transport mechanisms and energy functions and diffusion functions. And then you can go down to smaller levels than that. But when you come back out and you say, how will those cells make a kidney? It's not explanatory. I remember asking the OB who had delivered my first child. I was so amazed. And I asked him like, what do you think? Like, what do you know? You do this, right? You're seeing this life created. And his thought was nothing, I just marvel. I mean, I get to do this, but I just marvel at it. And I think the more we know about us, </s><s>[INST] the more we respectfully marvel. And we should do that. We should proactively marvel at every aspect, at every layer that where the novelty emerges. [/INST] Yes, we'd be a lot less likely to say, hey, I don't like you because of something, whatever it is, race, religion, culture, sexuality, gender identity, whatever it is. Or I want to say, I want rights that you don't have, right? Or I want what you have, right? I mean, there's so much of this. And I understand it's driven by scarcity and by human insecurity and envy and all of these things that I think could drive us towards destruction. But all of that recklessness comes from not having this initial appreciation and respect that you're referring to and just marveling at, like, wow, okay, we're here. That's amazing. </s><s>[INST] Let's start with that. But if we marvel at this whole thing, the human project, the human condition, all the different kinds of human beings that are possible, what do you then make of that some humans do evil onto the world? First of all, are all human beings capable of evil? If we're in the process, now we've got a little bit of momentum in terms of marveling the layers of the cake. Should we also marvel at the capacity for evil in all of us? Yeah. [/INST] Is that capacity there? Yeah. </s><s>[INST] I believe that it is, yes. So what do we understand about the psychology of evil? Where does that originate in the human mind? Is it there in the neurobiology? [/INST] Is it there in the environment and the upbringing of evil? Can I clarify first? I think the capacity for evil, I do believe is in all of us. There's a difference between enacting evil and a sort of preset followed developed plan of evil, right? I don't believe that all of us are capable of doing what the people who perpetrate the most evil do, right? But I do believe that we're capable of perpetrating evil, right? And the thought, one thought would be that there are drives in us. I mean, there certainly seem to be drives in us towards survival, towards gratification, in some ways towards pleasure. And that can get very complicated because pleasure inside can be relief of distress. So if I feel very badly about myself and I can feel a little better about myself by making you feel worse about yourself, right? Which that plays out in a lot of human beings. Is that an indirect way of bringing pleasure, right? So it gets very complicated what's going on inside of us. And sometimes the perpetration of evil things can be through misunderstandings, anger, impulsivity. I mean, there are things that we can have in us and other times there can be other things going on which are through the lens of unhealthy human psychology. So for example, the psychology of envy, which I think drives the lion's share of the orchestrated evil, right? There's a difference between impulsive, reflexive evil and highly orchestrated evil, </s>

<s>[INST] So it's interesting to ask you just your thoughts? Uh, Cosmo still lives on under dream labs, I think, uh, is that are you tracking the progress there or is it too much pain? Uh, is it, are you, is that something that you excited to see where that goes? [/INST] So keeping an eye and of course, just out of curiosity, and obviously just kind of care for a product line, I think, um it's deceptive how complex it is to manufacture and evolve that product line. Um and the amount of experiences that are required to complete the picture and be able to move that forward, and I think that's gonna make it pretty hard to do something really substantial with it. It would be cool if, like, even the product in the way it was, was able to be manufactured. Again, that was just the current, which will be need. Um but uh, it's, I think it was deceptive how tricky that is on, like everything from the quality control, the details and um and then, like, technology changes that forces you to rick, reinvent and update certain things. Um so I haven't been super close to it, but just kind of keeping it on </s><s>[INST] it. Yeah, it's really interesting how it's deceptively difficult, just as you're saying, for example, uh those same folks, uh and I've spoken with them there, they partnered up with rick and morty uh creators to do the butter robot. I love the idea just recently, I kind of half fast watch rick and morty previously, but now I just watched, like the first season, it's such a brilliant show. I'd like, I did not understand how brilliant that show is, and obviously, I think in season one is where the butter robot comes along for just a few minutes or whatever. But I just fell in love with the butter robot, the sort of the that particular character, just like you said, there's characters, you can create personality, you can create in that particular a robot who's doing a particular task [/INST] realizes, </s><s>[INST] you know, just like realized, asked the existential question, the myth of Sisyphus question that uh camus writes about like, is this all there is is he moves butter. But you know that realization, that's a that's a beautiful little realization for a robot [/INST] that I'm my purpose </s><s>[INST] is very [/INST] limited to this particular task. </s><s>[INST] It's abuse, it's humor of course, is darkness. It's a beautiful mix. But so they wanna release that butter robot. But something tells me that to do the same depth of personalities Cosmo had the same richness. It would be on the manufacturing on the ai on the storytelling and the design is going to be very, very difficult. It could be a cool sort of toy for rick and morty fans, but to create the same depth of existential angst that the butter robot symbolizes is really, that's the brave effort you succeeded at with Cosmo. But it's not easy. [/INST] It's really does. And you can fail in almost any one of the kind of dimensions and like uh and yeah, I think, you know, unique convergence of a lot of different skill sets to try to pull that off </s><s>[INST] on this topic, let me ask you for some advice because as I've been watching rick and morty, I told myself I have to build the butter robot, just this is a hobby project. And so I got a nice platform for it with treads and, and there's a camera that moves up and down and so on. Um, I'll probably paint it. Uh, but the question I'd like to ask the obvious technical questions. I'm fine with communication, the personality storytelling, all those kinds of things. Um, I think I understand the process of that. But how do you know when you got it right. So with, with Cosmo, how did you know this is great like, or something is off? Like is this brainstorming with the team? Do you know when you see it, is it like love at first sight, it's like this is right or like, I guess if we think of it as an optimization space is their uncanny valley where you're like, that's not right or this is right or are a lot of characters [/INST] right? We stayed away from uncanny valley just by having such a different like mapping where it didn't try to look like a dog or a human or anything like that. And so, uh, you avoided having like a weird pseudo similarity but not quite hitting the mark. Um, but you could like just fall flat, we're just like a personality or character emotions just didn't feel right. And so it actually mirrored very closely to kind of the iterations that are character director of Pixar would have, where you're running through it and you can virtually kind of like see what it'll look like. We, we created a plug into where we actually used like like Maya, the, you know, the animation tools and then we created a plug in that Perfectly matched it to the physical one. And so you could like test it out virtually and then push a button and see it physically play out and there's like subtle differences and so you want to like make sure that that feedback loop is super easy to be able to test it live. Um and then sometimes like you would just feel it that it's right and intuitively know and then you'd also do, we did user testing but it was very, very often that like the into like if we found it magical, it would scale and be magical more broadly. There were not too many cases where like, like we were pretty decent about not like getting to it, you know, getting out or getting too attached to something that was super unique to us. Um but trying to kind of like, you know, put a customer hat on and does it truly kind of feel magical and so in a lot of ways we just give a lot of um autonomy to the character team to really think about the, you know, character board and mood boards and storyboards and like what's the background of this character and how would they react. Um And they went through a process that's actually pretty familiar, but now had to operate under these unique constraints. Um but the moment where it felt right um kind of took a fairly similar journey than like as a character in an animated film, actually, it's quite cool. </s>

<s>[INST] What percent of the day would you say you spend as a physical ist um sort of living life pretending your car matters, your reputation matters. Like like how much was that? Tom waits song. I like my town was a little drop of poison. How much poison do you allow yourself to have? [/INST] I think my default mode is physical ist. I think that that's just the default I when I'm not being conscious. Yeah, consciously attentive </s><s>[INST] intellectually, consciously intensive, [/INST] because if you're just </s><s>[INST] you're still, if you're tasting coffee and not thinking or drinking or just taking in the sunset, you're not being intellectual, you're but you're still experiencing it. So it's when you turn on the like the introspective machine, that's when you can start [/INST] and turn off the thinker when I actually just start looking without thinking. So that's that's when I feel like I all of a sudden I'm starting to see through, sort of like, okay, part of, part of the addiction to the interface is all the stories I'm telling about. It's really important for me to get that really important to do that. So I'm telling all these stories and so I'm all wrapped up almost all of the mind stuff that's going on in my head is about attachment to the interface. And so what I found is that the essentially the only way to really detached from the interface is to literally let go of thoughts altogether. And then all of a sudden, um, even my identity, you know, my whole history, my name, my education and all this stuff is almost irrelevant because it's just now, here is the present moment and this is this is the reality right now. And all of that other stuff is an interface story. But this conscious experience right now, this is the only, this is the only reality as far as I can tell, thrust of, it's a story and but that is again not my default, that is I have to make a really conscious choice to say, okay, I know intellectually this is all an interface, I'm going to take the headset off and so forth and then immediately sink back into the game and just be out there playing the game and get lost. And so I'm always lost in the game unless I literally consciously choose to stop thinking. </s><s>[INST] Isn't it terrifying [/INST] to </s><s>[INST] acknowledge that, to look beyond the game, isn't it? Uh [/INST] scares the hell out of me. It's really scary because I'm so attached, I'm attached to this body, I'm attached to the interface. </s><s>[INST] Are you ever worried about breaking your brain [/INST] a bit, </s><s>[INST] meaning? Like, it's um I mean some of these ideas when you think about reality, even with, like Einstein just realizing you said interface, just realizing that light, you know, there's a speed of light and you can't go fast in the speed of light and like what kind of things black holes and can do with light, even that can mess with your head. [/INST] Yes, </s><s>[INST] but that's still space time, [/INST] That's a big mess, But it's still just space time is still a property of our interface. That's right. </s><s>[INST] But it's still like even even Einstein realized that this particular things, some of the stories we tell ourselves is constructing interfaces that are oversimplifying the, the way things work because the it's nice, the stories are nice, [/INST] stories </s><s>[INST] are nice, that's what I mean, just like video games, they're nice, [/INST] right? But Einstein was a realist, right? He was a famous realist in the sense that he It was very explicit in 1935 paper with Podolski and Rosen the DPR paper where he they said if without in any way disturbing the system, I can predict with probability one the outcome of a measurement, then there exists in reality that element write that value that and we now know from quantum theory that that's false, That Einstein's idea of local realism is strictly speaking false. Yeah. And and so we can predict, we can set up in quantum theory, you can set up and there's a there's a paper by chris Fuchs quantum basin is um where he scouts this out that it was done by the people. But he gives a good presentation of this where they have a sequence of something like nine different quantum measurements that that you can make and you can predict with probably one what a particular outcome will be but you can actually prove that it's impossible that the value existed before you made the measurement. So you know with probability one what you're going to get. But you also know with certainty but that value was not there until you made the measurement. So we know from quantum theory that the act of observation is an act of fact creation. And that is built into what I'm saying with this theory of consciousness. If consciousness is fundamental, space time itself is an act of fact creation, it's an interface that we create. Consciousness creates plus all the objects in it. So local realism is not true. Quantum theory is established that also non contextual realism is not true and that that fits in perfectly with this idea that consciousness is fundamental. These things are these exist as data structures when we create them as as chris Fuchs says the act of observation is an act of fact creation. But I must say on a personal level I'm having to spend I spent a couple hours a day just sitting in meditation on this and facing the rebellion in me that goes to the core. It feels like it goes to the core of my being rebellion against these ideas. So so here is very, very interesting for me to look at this because so here I'm I'm a scientist and I'm a person. The science is really clear. Local realism is false. Non contextual realism is false space time is doomed. It's very, very clear, it couldn't be clearer and my emotions rebel left and right. When I sit there and say, okay, I am not something in space and time. Something inside of me says you're crazy. Of course you are. And I'm completely attached to it. I'm completely attached to all this stuff. I'm attached to my body. I'm attached to the headset attached to my car. Touch two people I'm attached to all of it. And yet I know as an absolute fact, I'm going to walk away from all of it. I'm going to die. You know it in fact almost died last year. Covid almost killed me. I I sent a goodbye text to my wife. So I was I thought </s>

<s>[INST] How do you generate the different simulations? [/INST] So you know there's plenty of parameters out there, we just pick them randomly And uh in simulation model just goes for thousands of years and keeps on solving Rubik's cube in each of them. And the thing is that neural network that we used, it has a memory and as it presses for instance the side of the of the cube, it can sense, oh that's actually this side was difficult to press, I should press it stronger and throughout this process kind of learns even how to how to solve this particular instance of the Rubik's cube. Thank you very much. It's kind of like an you know, sometimes when you go to the gym and half there after bench press, you try to lift the glass and you kind of forgot and your hand goes like right away because kind of you've got to maybe different weight and it takes a second to adjust and this kind of of memory model going through the process of interacting with the Cuban, a simulation. </s><s>[INST] I appreciate you speaking to the audience with a bench press all the boroughs in the audience probably working out right now. There's probably somebody listening to this actually doing bench press. Um so maybe uh put the bar down and pick up the water bottle and you'll know exactly what what is talking about. Ok, so what what was the hardest part of getting the whole thing to work? So [/INST] the hardest part is at the moment when it comes to a physical world, when it comes to robots, uh they require maintenance. It's hard to replicate a million times. It's also it's hard to replay things exactly. I remember this situation that one guy at our company, he had like a model that performs way better than other models in solving Rubik's cube. And uh you know, he kind of didn't know what's going on. Why is that? And it turned out that you know he was running it from his laptop that had better CPU or uh better maybe local Gpu as well. And because of that there was less of a latency and the model was the same. And that actually made solving Rubik's cube more reliable. So in some sense there might be some saddlebags like that when it comes to running things in the real world. Even hinting on that you could imagine that the initial models you would like to have models which are insanely huge neural networks and you would like to give them even more time for thinking. And when you have these real time systems uh then you might be constrained actually by the amount of latency. And uh ultimately I would like to build a system that it is worth for you to wait five minutes because it gives you the answer that they are willing to wait for five minutes. </s><s>[INST] So latency is a very unpleasant constraint under wish to [/INST] operate correct. And also there is actually one more thing which is tricky about robots. Uh There is actually uh no uh not much data. So the data that I'm speaking about would be a data of uh first person experience from the robot like gigabytes of data like that if we would have gigabytes of data like that of robots solving parties problems. It would be very easy to make a progress on robotics and you can see that in case of text or code. There is a lot of data, like a first person perspective data on writing coat. </s><s>[INST] Yeah. So you had this, you mentioned this really interesting idea that if you were to build like a successful robotics company, so open as mission is much bigger than robotics, this is one of the one of the things you worked on. But if it was a robotics company that you wouldn't so quickly dismiss supervised learning, correct? That you would build a robot that was perhaps what like um an empty shell like dumb and they would operate under tell the operation. So you would invest that's just one way to do it, invest in human super like direct human control of the robots as its learning and over time add more and more automation. [/INST] That's correct. So let's say that's how I would build a robotics company today. If I would be building robotics company which is you know spend $10 million dollars or so recording human trajectories controlling a robot </s><s>[INST] after you find a thing that the robots should be doing that. There's a market fit for like that. You can make a lot of money with [/INST] that product, correct? Correct. Uh So I would record data and then I would essentially train supervised learning model on it. That might be the path today long term I think that actually what is needed is to train powerful models over video. So you have seen maybe models that can generate images like dolly and people are looking into models generating videos there like a party's algorithmic questions, even how to do it. And it's unclear if there is enough compute for this purpose. But I suspect that the models that which would have a level of understanding of video, same as GPT, has a level of understanding of text could be used uh to train robots to solve tasks. They would have a lot of common sense </s><s>[INST] if one day, I'm pretty sure one day there will be a robotics company by robotics company. I mean, the primary source of income is from robots That is worth over $1 trillion. What do you think that company will do? [/INST] I think self driving cars? No, </s>

<s>[INST] I would have to look back into evolutionary history when the gender has emerged. But yeah, most [/INST] of the faces of humans are quite different from the faces of great apes. Great apes. Right? Yeah. You think you couldn't tell, you couldn't tell uh the face of a female chimpanzee from the face of female chimpanzee? Probably. Yeah, I don't think humans of all </s><s>[INST] that. So [/INST] we do have innate knowledge of what makes the face. But it's actually impossible for us to have any DNA encoding knowledge of the difference between a female human face and male human face. Because the that knowledge that information came up into the world actually very recently. If you look at the at the at the slowness of the process of fun coding knowledge into DNA. </s><s>[INST] Yeah, so that's interesting. That's a really powerful argument that DNA is a low bandwidth and it takes a long time to encode here. That naturally creates a very efficient encoding. [/INST] But yeah, 1 1 important consequence of this is that So yes, we are born into this world with a bunch of knowledge sometimes the high level knowledge about the world, like the shape, the rough shape of a snake of the rough shape of face. But importantly because this knowledge takes so long to write almost all of these innate knowledge is shared with our cousins with great apes, right? So it is not actually this innate knowledge that makes us special </s><s>[INST] but to throw it right back at you from the earlier on in our discussion, it's that encoding might also include the entirety of the environment of earth [/INST] to some extent. So it can it can include things that are important to survival production. So that from which there is some evolutionary pressure and things that are stable constant over a very very very long time periods. And honestly, it's not that much information. There's also besides the bandwidths constraint and uh concerns of the writing process. There's also memory constraints like D. N. A. The part of DNA that deals with the human brain is actually fairly small. It's like, you know, on the order of megabytes, right? It's not that much high level knowledge about the world you can encode. </s><s>[INST] That's quite brilliant and hopeful for benchmark of uh that you're referring to of encoding prior is actually look forward to. Uh I'm skeptical whether you can do in the next couple of years, but hopefully [/INST] I've been working. So honestly it's a very simple benchmark and it's not like a big Brexit or anything. It's more like a fun a fun side project. </s><s>[INST] These fun so is image that these fun side projects could launch entire groups of efforts towards um towards creating reasoning systems and so on. [/INST] And I think that's the goal. It's trying to measure strong generalization to measure the strength of abstraction in our minds, in our minds and in a in an artificially intelligent agent. And if </s><s>[INST] there's anything true about this science organism is its individual cells love competition. So in benchmarks encourage competition. So that's a that's an exciting possibility if you uh do you think an Ai winter is coming and how do we prevent it? [/INST] Not really. So, an air winter is something that would okay when there's a big mismatch between how we are selling the capabilities of AI and the and the actual capabilities of AI and today. So deep learning is creating a lot of value and will keep creating a lot of value in the sense that this is models applicable to a very wide range of problems that are driven today. And we're only just getting started with the primes algorithms to every problem that could be solving. So the planning will keep creating a lot of value for the time being. What's concerning, however, is that there's a lot of hype around depleting around Ai. There lots of people are over selling the capabilities uh these systems, not just the capabilities but also over selling the fact that there might be more or less, you know, brain like like giving a kind of a mystical aspect. These technologies and also overseeing the pace of progress, which, you know, it might look fast in the sense that we have this exponentially increasing number of papers. But again, that's just a simple consequence of the fact that we have ever more people coming into the field doesn't mean the progress is actually explanatory fast. Let's say you're trying to raise money for your startup or your research lab, you might want to tell, you know, uh, grandeur stories to investors about how deep learning is, just like the brain and hide can solve all these incredible problems like self driving and robotics and so on. And maybe you can tell them that the fill is progressing so fast and we're going to g within 15 years or even 10 years and none of this is true. And every time your life saying these things and uh, an investor or, you know, a decision maker believes them. Well, your this is like the equivalent of taking on credit card depth, but for for trust, right? And um, maybe this, this will, you know, uh uh, this will, this will be what enables you to raise a lot of money, but ultimately, you are creating damage your dimension in the field. </s><s>[INST] That's the concern is that that that that's what happened to the other Ai winters. As the concern is you actually tweeted about this with autonomous vehicles, right? There's uh, almost every single company now have promised that they will have full autonomous vehicles by 2020 [/INST] one, That's a good example of the the consequences of overhyping the capabilities of AI and the pace of progress </s>

<s>[INST] Yeah, I mean, but nevertheless there's, there's a lot you can do on social networks, I mean, I I enjoy tweeting and consuming twitter, it's just I I apply the exact same principle that you just said, which is uh free will discussion, which is like I approach it in a way that I don't get stuck in this loop. That's counterproductive, I try to do things that are productive and like it's just like you said, that's like uh like what kind of things can I do in this world, whether that's tweeting or building things, those are low effort tweeting or actually building businesses or building ideas out as high effort, what can I do that will actually solve problems and that's that's the way I approach it and I do wonder if it's possible to at scale encourage each other to approach like social media and communication with fellow humans in that way, I don't know [/INST] how do you think that would be done? I guess like to improve the, improve the quality of discourse. Maybe like even like you said the empathy or the decency of discourse, </s><s>[INST] I think people should be, you know uh incentivized, encouraged to do that. I think most of what's we see happening on twitter and facebook and so on has to do with very small but very powerful implementation details. It goes down to like what is the source of the dopamine rush, the like button uh sharing mechanisms just even small tweaks and those can fix a lot really, I I believe so, so like a lot a lot of the stuff we see now is the result of just initial implementations of these systems that we didn't anticipate. So the modernization comes from engagement and the tools we have is clicking like and sharing, it was not always obvious, it was not obvious from the beginning, it was an obvious while twitter and facebook grew that there's a big dopamine rush from getting more followers and likes and shares. So we've gotten addicted to this feeling like how many people are commenting, how many people are saying like uh clicking like and so on. So that's that dopamine rush. So we want to say the thing they'll get the most likes on and like unmasked in society. And then the other thing that was expected is the controversial, the divisive will get the most likes. So we it had to do with the initial mechanisms of likes and shares, resulting in an outcome that was unpredicted, which is huge amounts of division irrespective of like any of the basics of human connection that we've actually all coming to understand the society is valuable at the individual level that we're saying but on mass what results is like you throw all that out and it's all just divisive at scale, ah discourse, I think it could be fixed by incentivizing personal growth, like incentivizing you to challenge yourself to grows individual and most importantly to be happy at the end of the day? So, uh, feed, like incentivize you feeling good has in a way that's long lasting long term. I think what makes people actually feel good is being kind to others long term in the short term, what feels good is getting a lot of likes and I think those are just different incentives that if implemented correctly, you can just build social networks that would do much [/INST] better. So, do you think it comes from a structural perspective? I guess at what point does you mention, like you mentioned free will and also you mentioned, you know, feeling good like and again, working hard. You know, you, I know that you have a, I guess the it was a race or </s><s>[INST] no, it's the Goggins uh, for about 44 by four by 48 challenge where you run four miles every four hours for two days. That's awesome. It's a bunch of its uh, the challenge of it isn't just the running, the running is very tough, but it's mostly the sleep deprivation, right? You're just training every four hours, [/INST] but it's a struggle, right? But the struggle gives meaning and ultimately, I guess so how can we, because you mentioned, like you said, adjusting things on like a, I guess like a programming level. Almost based programming level. So that the interface is different for the user. But at what point does the user have a responsibility to, you know, as as a man or a woman or a person to just behave more decently. How can we, I guess utilize what can we do? It seems like, you know, we're like our society is so grossly missing like a martin Luther king right now, like the great inspiring characters throughout american history throughout world history. Where are the great leaders? </s>

<s>[INST] Of course, the flip side of that is sometimes you need that to have the strength to carry a belief against the entire scientific community that resists your beliefs. And so it's a double-edged [/INST] sword. That can happen, but I guess the distinction here is evidence. So in this case, that the evidence was so overwhelming, it wasn't really a matter of interpretation. You would observe this star with the same star, but with maybe 10, even 100 times greater precision for much longer periods of time. And there was just no doubt at this point this planet was a mirage. And so that's why you have to be very careful. I always say, don't ever name my wife and my daughter. Name this planet after me that you discover. And I can't ever name a planet FDU, because I won't be objective anymore. How could I ever turn around to you and say that planet wasn't real that I named FDU? </s><s>[INST] So you're somebody that talks about and it's clear in your eyes and in your way of being that you love the process of discovery, that joy, the magic of just seeing something, new observation, a new idea. But I guess the point is when you have that great feeling is to then switch on the skepticism. To start testing, what does this actually mean? Is this real? What are the possible different interpretations that could make this a lot less [/INST] grand than I first imagined? So both have the wonder and the skepticism on one brain. Yeah. I think generally the more I want something to be true, the more I inherently doubt it. And I think that just comes from I grew up with a religious family and was just sort of indoctrinated to some degree, like many children are, that, okay, there's just normal that there's a garden. This is the way the world is. God created the earth. And then as I became more well-read and illiterate of what was happening in the world scientifically, I started to doubt. And it really just struck me that the hardest thing to let go of when you do decide not to be religious anymore, and it's not really like a light bulb moment, but it just kind of happens over 11 to 13, I think for me it was happening. But it's that sadness of letting go of this beautiful dream, which you had in your mind of eternal life, for behaving yourself on earth, you would have this beautiful heaven that you could go to and live forever. And that's very attractive. And for me, personally, that was one of the things that pulled me against it, was it's like it's too good to be true. And it's very convenient that this could be so. And I have no evidence directly in terms of a scientific sense to support this hypothesis. And it just became really difficult to reconcile my growth as a scientist. And I know some people find that reconciliation. I have not, maybe I will one day. But as a general guiding principle, which I think I obtained from that experience, was that I have to be extremely guarded about what I want to be true, because it's going to sway me to say things which are not true, if I'm not careful. And that's not what </s><s>[INST] we're trying to do as scientists. So you felt from a religious perspective that there was a little bit of a gravitational field in terms of your opinions, like it was affecting how you could be as a scientist. Like as a scientific thinker, obviously, you were young. [/INST] Yeah, I think that's true, that whenever there's something you want to be true, it's the ultimate seduction intellectually. And I worry about this a lot with UFOs. It's true already with things like Venus, Phosphine, and searching for astrobiological signals. We have to guard against this all the way through from however we're looking for life, however we're looking for whatever this big question is. There is a part of us I think that I would love there to be life in the universe. I hope there is life in the universe, but I've been on record several times as being fairly firm about trying to remain consciously agnostic about that question. I don't want to make up my mind about what the answer is before I've collected evidence to inform that decision. That's how science should work. If I already know what the answer is, then what am I doing? That's not a scientific experiment anymore. You've already decided, so what are you trying to learn? What's the point of doing the </s>

<s>[INST] Yeah, [/INST] right. I'm talking to the guy who, who quit imax so I don't know there's discord servers and in these like there's like there's like seven uh different rooms you can go in or there's several rooms and each one is like a different field of making money. E commerce trading, Cryptocurrency um I think fulfillment by amazon like copyrighting okay, so I went to all of these, I checked him out, I checked all their money making tools. The first funny thing is that Andrew Tate is nowhere to be found. The supposed successful guy that you like bought into is nowhere to be found. It's these professors that you have now he has hired and said these guys are super qualified so like looked up some of what some of these guys have done and some of them have launched like scam e crypto coins. The Cryptocurrency professor was like shilling a bunch of coins that did bad and then like deleted the tweets. I mean just completely exactly what you'd expect behind the paywall, it's nothing of substance. You're not gonna learn to get rich by escaping the Matrix and going to work for Jeff Bezos fulfillment by amazon is not escaping the Matrix, right? Like that's not the way to hustle to the top, it's literally a field of making money that everyone in the world has access to if you want to differentiate yourself and make money. The first thing you realize is going into skill sets that literally anyone with an internet account can do is a bad way to do that because you have to have some differentiating factor to add value. So it's just such a obvious and complete scam because there is no value to this like so called education, the professors are crap, the advice there, like hiding some of the bad things they've done and agitates nowhere to be found. Ultimately, that's why everyone joins what he's done is very interesting because, and I'll give him credit in his marketing. He's been very savvy to like make the reasons you admire him, not the thing he's sort of selling, which is weird, like he's selling get rich quick, which seems like it relates to his persona but is actually very orthogonal to it. His persona is like the tell it how it is like tall buff rich guy. It's like actually his persona that you're buying into and then he's selling you this thing to the side which when people get in there and they're not delivered on the product, he still is those things that you first thought he was. So it's like, I think it's to some extent he's made a lot of money by making the thing that he makes money on, not the thing he gets so much pushback online for and what he's also loved for. So people will push back for his misogyny, but the real way he's making money is just like basic get rich quick schemes that are super obvious to spot, but everyone's distracted by like, oh he said some crazy stuff about women or you know, all these various other scandals he's gotten himself in to </s><s>[INST] get more and more and more attention. So with the persona, uh, is the hustler universities still operating? [/INST] I think they've rebranded it, I'm part of their pitch now. I'm like, they put me on is like, I mean, as like the Matrix is trying to take us out lex and then it's me saying like, you know, they put me in like saying I'm part of the Matrix, they put me in saying, oh this guy sucks, you know, I joined, it sucks and so they'll play that and they do like a bass drop and it's like, you know, don't listen to people like this da da da da. I mean, it's basically like co opted by </s><s>[INST] the Matrix to attack [/INST] your </s><s>[INST] an insider threat that infiltrated now is being used by the Matrix to attack him. Well to [/INST] everyone who criticizes him as part of the Matrix, he won't say who the Matrix is. It's just, it's just the shadowy cabal of rich powerful people. It's just like the easy narrative for people who are disaffected and who feel cheated by the system, You just collectivized that system and you make it the bad guy and you go look, look, those guys, those guys have been cheating you, they're the bad guys. They want me shut </s><s>[INST] up and [/INST] then now the person that the people who harmed you, they want this guy shut up, you're gonna listen to him. So that's like, it's like the most basic psychological manipulation that everyone seems to constantly fall for. It's it's really a trivial and stupid, but </s><s>[INST] can you steal man the case for hustlers University where it's actually giving young people confidence teaching them valuable lessons about like actually incentivizing them to do something like fulfillment by amazon. The basic thing to [/INST] try to </s><s>[INST] learn about it to fail or maybe see like to try to give a catalyst and incentive to try [/INST] as much as it pains me. I will try to give a a succinct, maybe steel man of it as best as I can thinking that it's such a grift. But um I think what you would say is that some people in order to make a change in their life need a someone who they can look up to and men don't have a lot of like strong role models like big male presences in their life who can serve as a proper example. So the most charitable interpretation would be Andrew Tate would encourage you to go reach for the stars. I guess my problem is I have a deep, like I have a deep issue with the like lust and greed that centers all these things. It's like this glorification of wealth equal status, wealth equals good person wealth and Bugatti equals you are meaningful, you matter. And like the dark underlying thing is that none of that, none of that matters like it matters that you make a decent living but passed just like that. I think the like lust for more stuff and the idealization of these people that is just like opulence is a net bad. So that's like my steel man has to stop there because I really disagree with like the values that are pushed by people like that. </s>

<s>[INST] So you mentioned one of the ideas in the vision context to uh that works, is to have different crops. So you can think of that as a way to sort of manipulating the data to generate examples that are similar obviously uh there's a bunch of other techniques, you mentioned lighting as a very, you know, in images, lighting is something that varies a lot and you can artificially change those kinds of things. There's the whole broad field of data augmentation which manipulates images in order to increase arbitrarily the size of the data set. First of all, what is data augmentation and second of all, what's the role of data augmentation in self supervised learning and contrast of learning? [/INST] So data augmentation is just a way, like you said, it's basically a way to augment the data. So you have send samples and what you do is you basically define some kind of transforms for the sample. So you take your image and then you define a transform where you can just increase the color, like the color or the brightness of the image or increase or decrease the contrast of the image for example or take different crops of it. So, data augmentation is just a process to basically put or the data or augment the data. Right? And so it has played a fundamental role for computer vision for self supervisor, especially the way most of the current methods were contrasted or otherwise, is by taking an image in the case of images is by taking an image and then computing basically to perturbations of it. So these can be two different crops of the image with like different types of lighting or different contrasts or different colors. So you did all the colors a little bit and so on. And now the idea is basically because it's the same object or because it's like related concepts in both of these perturbations. You want the features from both, both of these perturbations to be similar. So now you can use a variety of different ways to enforce this constraint. Like these features being similar, you can do this by contrast of learning. So basically both of these things are positives. A third sort of images negative, you can do this basically by like clustering for example, you can say that both of these images should the features from both of these images should belong in the same cluster because they're related. Whereas image, another image would belong to a different cluster. So there's a variety of different ways to basically enforce this particular constraint </s><s>[INST] by the way, when you say features, it means there's a very large neural network that extracting patterns from the image and the kind of patterns that extracts should be either identical or very similar. That's what that means. [/INST] So, the neural network basically it takes in the image and then outputs a set of basically vector of like numbers and that's the feature and you want this feature for both of these different crops that your computer to be similar. You want this vector to be identical and it's like entries for example, </s><s>[INST] be like literally close in this multidimensional space to each other and like you said, close can mean part of the same cluster or something like that in the in this large space first of all that, I wonder if there is connection to the way humans learn to this almost like maybe subconsciously in order to understand the thing, you kind of have to see it from 23 multiple angles. I wonder there's a lot of friends for neuroscientists, maybe orc and cognitive scientists, I wonder if that's in there somewhere. Like in order for us to place a concept in its proper place, we have to basically crop it in all kinds of ways uh do basic data augmentation on it and whatever. Very clever ways that the brain likes to [/INST] do </s><s>[INST] um like spinning around in our minds somehow. That is very effective. [/INST] So I think for some of them we need to do it. So like babies for example pick up objects like move them and put them go to their eye and whatnot. But for certain other things actually we are good at imagining it as well. Right? So if you, I have never seen for example an elephant from the top, I've never basically looked at it from the top down, but if you showed me a picture of it, I could very well tell you that that's an elephant. Uh I think some of it, we're just like we naturally build it or transfer it from other objects that we've seen to imagine what it's going to look like. Has anyone </s><s>[INST] done that with the augmentation. Like imagine all the possible things that are occluded or not there, but not just like normal things like wild things, but that are nevertheless physically consistent. [/INST] So I mean people do kind of like occlusion based augmentation as well. So you place in like a random like box grey box to sort of mask out a certain part of the image. And the thing is basically you're kind of including for example, you place it on half of a person's face. So basically saying that you know something below their noses occluded because it's great out. So you know, I </s>

<s>[INST] off the engine. [/INST] Yeah he shut off the engines and landing flying side by side like this. </s><s>[INST] How do you explain that? Was he just a wild man or was he sufficiently skilled to [/INST] work? Sufficiently skilled? Absolutely. He knew what he was doing. I can get a plane from one spot to another and I guess I'm known as a good pilot but that guy it was an aerobatic uh so anyway he stayed with a couple of days and then I told him I said this plane needs uh need's tank and I said I got some work down to Colombia needs to come back to Louisiana and I need 2500 mile range. He said I got somebody in Mena Arkansas do that and keep your mouth shut. So I gave him $10,000 and he flew away and in a few days he called me and says come to my house in baton rouge. So I went out to his house in baton rouge and I stayed with him for a few days and that plane was tanked, I mean beautiful from stem to stern. I could went from Bolivia to Canada with. So uh he was uh, then I hired him to fly. And uh, he was funny, I paid him a million dollars a trip of him. $2000 a kilo. So About $1 million New York and then the money come back. So it was a couple of two or three weeks pipeline. Well I always had to pay him before before he'd go again. I mean, and he belly ache. I mean he had my own room. So one time I, I gave him $1 million dollars and I put it in a box real nice. </s><s>[INST] So how big is a box that contains $1 million. So we're talking about $100 bills. one [/INST] $100. It's not very big. You can put it in a large briefcase, It weighs exactly 10 kg. H bill weighs the ground so you can wear your money and almost get exactly </s><s>[INST] 20 some pounds is a million dollar, £22 [/INST] £20. 100 </s><s>[INST] dollar bills. But [/INST] $101 bills. It's one ton, £2,200. We didn't accept them. </s><s>[INST] Were you the one that introduced barry seal to Pablo escobar? [/INST] No, I didn't introduce him at all. And uh, and I, our deal was that you don't meet my people. I mean we just kind of cross, you working for me to fly the airplanes. So he wanted these panther conversions cost $400,000 each with storm scope radar on anything you want. </s><s>[INST] What's that mean? Sorry to interrupt panther conversions. [/INST] Panther conversion was uh, these people called panther the took everything out from the firewall and instruments and all and converted them and put Q tip propellers on the full bladed and you're very quiet. And the CIA developed those in Southeast Asia for running behind the lines. And that's where barry had flung on those things. So he knew about. So hey, that's what he wanted and that's what we got him. </s><s>[INST] How does that connect to Pablo And so he worked for you and you got those upgrades. [/INST] I think he flew about 30 loops for me and then I got arrested and was better for everything in the world. Got 35 year sentence. But let me back up a little bit. Barry. Uh, was our friend Mari and I both friends. We </s><s>[INST] should pause real quick and say mari is, uh, your wife and well hopefully she'll will convince her to join us in a little bit. She's the love of your life and sort of, she uh, weaves in and out of many of these stories that you tell. Yes, [/INST] she was there. She was behind the scenes, but I kept her out of it completely. </s><s>[INST] And then also you mentioned Merriam as uh, your [/INST] daughter. Yes. Read our son was a, was a baby. And I remember we went out to the festival was my favorite restaurant and carl Gables. Oh God it was good and barry knew about it anyhow. We went out to dinner and uh, so we came back and there was no rooms so very well in the night with us. So he goes to our hotel room with him, we got to two big beds and uh, the Omni hotel and he lays over and gets down to stripe under shorts in his t shirt and he put the baby up on his belly and give him the bottle. Mm ain't that good red. Oh my mind, he just leads the baby. We laugh and talk. That's how close we were that we could all stay in a hotel room together </s><s>[INST] and we just say he's a good man, [/INST] a wonderful man, a gentleman Southern gentlemen just just looked after his mother, his family, everybody around him, everybody loved barry. He just had a, he had a little little smile on his face always. </s><s>[INST] So you got arrested. And then what happened to Barry? [/INST] Well Barry knew the people that unloaded of course he sent the cars down all that. So he met the unloaded guy named lito Luis Carlos, Bustamante, Venezuelan. And uh, so he just kept on flying. But he, uh, I believe you had three of my airplanes and $400,000 apiece and they owed me some money when he collected a lot of that and gave Murray the money and put it in his safe and took her to his house and all after I got arrested. Single lawyer in you got me the best lawyer in the country Albert Krieger. He was head of the defense team for all. America wonderful man. </s><s>[INST] Can you tell the story of the months that led up to berries assassination. What what what what did you know? What did you sense? What did you think? [/INST] Okay when I got a prison I hadn't been out long. I was watched eating breakfast in there was Ronald Reagan's face right in the television we have absolute proof that the communist Sandinista government is in the cocaine running business. And there was a fat lady the C 21 26 on the runway would bellied in and I thought oh God he had done it. So I had heard that barry might been working with him. So it wasn't mom </s>

<s>[INST] emergent in this kind of situation. It's so fascinating that on the individual level, we have our own complex cognition, we have consciousness, we have intelligence, we're trying to figure out little puzzles. And then that somehow creates this graph of collective intelligence. We figure out and then you throw in these viruses of which there's millions of different, you know, there's entire taxonomy. And the viruses are thrown into the system of collective human intelligence and little humans figure out what to do about it. We get like, we tweet stuff about information. There's doctors as conspiracy theorists and then we play with different information. I mean, the whole of it is fascinating. I'm like you also very optimistic, but there's a, just you said the computational reducibility, you know, there's always a fear of the darkness of the uncertainty before us. [/INST] It's scary before us, it's scary. I mean, the thing is, if you knew everything, it would be boring. And it would be, and then, and worse than boring, so to speak. It would be, it would reveal the pointlessness, so to speak. And in a sense, the the fact that there is this computational reducibility, it's like, as we live our lives so to speak, something is being achieved. We're computing what happens in our lives. </s><s>[INST] Yeah, that's funny. So the computational reducibility is kind of like, it gives the meaning to life. It is the meaning of life. Computational reducibility is the meaning of life. There you go. It gives it meaning, yes. [/INST] It gives it meaning. Yes, I mean, it's what causes it to not be something where you can just say, you know, you went through all those steps to live your life, but we already knew what the answer was. Hold on one second. I'm going to use my handy Wolfram Alpha sunburn computation thing, so long as I can get network here. There we go. Oh, actually, you know what? It says sunburn unlikely. This is a QA moment. </s><s>[INST] This is a good moment, okay. [/INST] Okay, well, let me just check what it thinks there. I see why it thinks that. It doesn't seem like my intuition. This is one of these cases where we can, the question is, do we trust the science </s><s>[INST] or do we use common sense? [/INST] The UV thing is cool. Yeah, yeah, well, we'll see. This is a QA moment, as I say. And so, do we trust the product? Yes, we trust the product, so. And then there'll be a data point either way. If I'm desperately sunburned, </s><s>[INST] I will send in an angry feedback. Because we mentioned the concept so much, and a lot of people know it, but can you say what computation reducibility is? [/INST] Yeah, yeah, right. So it's only the question is, if you think about things that happen as being computations, you think about some process in physics, something that you compute in mathematics, whatever else, it's a computation in the sense it has definite rules. You follow those rules. You follow them in many steps and you get some result. So then the issue is, if you look at all these different kinds of computations that can happen, whether they're computations that are happening in the natural world, whether they're happening in our brains, whether they're happening in our mathematics, whatever else, the big question is how do these computations compare? Is, are there dumb computations and smart computations, or are they somehow all equivalent? And the thing that I was sort of surprised to realize from a bunch of experiments that I did in the early 90s, now we have tons more evidence for it, this thing I call the principle of computational equivalence, which basically says when one of these computations, one of these processes that follows rules, doesn't seem like it's doing something obviously simple, then it has reached the sort of equivalent level of computational sophistication of everything. So what does that mean? That means that you might say, gosh, I'm studying this little tiny program on my computer, I'm studying this little thing in nature, but I have my brain, and my brain is surely much smarter than that thing. I'm going to be able to systematically outrun the computation that it does, because I have a more sophisticated computation that I can do. But what the principle of computational equivalence says is that doesn't work. Our brains are doing computations that are exactly equivalent to the kinds of computations that are being done in all these other sorts of systems. And so what consequences that have? Well, it means that we can't systematically outrun these systems. These systems are computationally irreducible in the sense that there's no shortcut that we can make that jumps to the answer. In a general case. Right. So what has happened, what science has become used to doing is using the little pockets of computational reducibility, which by the way, are an inevitable consequence of computational irreducibility, that there have to be these pockets scattered around of computational reducibility to be able to find those particular cases where you can jump ahead. I mean, one thing sort of a little bit of a parable type thing that I think is, is fun to tell, you know, if you look at ancient Babylon, they were trying to predict three kinds of things. They tried to predict, you know, where the planets would be, what the weather would be like, and who would win or lose a certain battle. And they had no idea which of these things would be more predictable than the other. That's funny. And, you know, it turns out, you know, where the planets are is a piece of computational reducibility that, you know, 300 years ago or so, we pretty much cracked. And it's been technically difficult to get all the details right. But it's basically, we got that, you know, who's going to win or lose the battle? No, we didn't crack that one, that one, that one. </s>

<s>[INST] So I can't make sense of new york in this physical reality, but yeah, I mean, but you said pain and walking into things, well you can create a pain signal and as long as it's consistent that certain things result in pain, you can start to construct a reality. There's some, maybe, maybe you disagree with this, but I think we are born almost with a desire to be convinced by our reality, like a desire to make sense of our reality. So there's an imperative, so whatever that reality is given to us, like the table hurts, fire is hot, I think we want to be deluded in a sense that we want to make a simple, like Einstein simple theory of the thing around us, we want that simplicity And so um maybe the hunger for the simplicity is the thing that could be used to construct a pretty dumb simulation that that tricks us, so maybe tricking humans doesn't require building a universe. [/INST] No, I don't, I mean, I know this is not what I work on, so I don't know how close to it we </s><s>[INST] are, [/INST] but I I agree with you. Yeah, I'm not sure that it's a morally justifiable thing to do, but it's it's is it possible in principle? Um I think it'd be very difficult, but I don't see why in principle it wouldn't be possible. And I agree with you that it's it's that we try to understand the world, we try to integrate the century inputs that we have and we try to come up with a hypothesis that explains what's going on. I think though that we have huge input from the social context that we're in, we don't do it by ourselves, we don't kind of blunder around in a in a universe by ourselves and understand the whole thing. We're told by the people around us what things are and what they do and languages coming in here and so on, so it would have to be an extremely impressive simulation to simulate all of that. </s><s>[INST] Yeah, simulate all of that, including the social construct the thing that the spread of ideas and the exchange of ideas, I don't know, and but those questions are really important to understand as we become more and more digital creatures. It seems like the next step of evolution is us becoming partially all the same mechanisms we've talked about are becoming more and more plugged in into the machine were becoming cyborgs and there's an interesting interplay between wires and biology um you know, zeros and ones and the biological systems. And I don't think you can just I don't think we'll have the luxury to see humans as this joint from the technology we've created for much longer. We are in organisms. That's um [/INST] Yeah, I mean I agree with you but we come really with this too consciousness. And is there a distinction there? Because what you're saying, the natural end point says we are indistinguishable that if you are capable of building a an ai which is sufficiently close and similar that we merge with it then then to all intents and purposes that ai is conscious as we know it. Um and I don't I don't have a strong view but I have a view. Um and I wrote about it in the epilogue to my last book because 10 years ago I wrote a chapter in in a book called life ascending about consciousness and the subtitle of life ascending was was the 10 great inventions of evolution and I couldn't possibly write a book with a subtitle like that that did not include consciousness and specifically consciousness as one of the great inventions. And it was in part because I was just curious to know more, and I read more for that chapter. I never worked on it, but I've always how can anyone not be interested in the question? Um and I was left with the feeling that A nobody knows and B. There are two main schools of thought out there with a big kind of skew in distribution. One of them says, oh, it's a property of matter is an unknown law of physics. Panpsychism, Everything is conscious, the sun is conscious, it's just a matter or a rock is conscious. It's just a matter of how much and I find that very unpersuasive. I can't say that it's wrong. It's just that I think we somehow can tell the difference between something that's living and something that's not. And then the other, the other end is it's a it's an emergent property of a very complex central nervous system. Um and I am I never quite understand what people mean by words like emergence. I mean, there are genuine examples which I think we very often tend to um use it to to plaster over ignorance </s><s>[INST] as [/INST] a biochemist. The question for me then was, okay, it's a it's a concoction of a central nervous system, a deep polarizing neuron gives rise to a feeling, to a feeling of pain or to a feeling of love or anger or whatever it may be. So, what is then a feeling in biophysical terms in the central nervous system which bit of the wiring gives rise to? I've never seen anyone answer that question in a in a way that makes sense to me </s><s>[INST] and that's an important question to answer. [/INST] I think if we want to understand consciousness that's the only question to answer. Because you know I certainly an Ai is capable of out thinking and it's only a matter of time. Maybe it's already happened in terms of just information processing and computational skill. I don't think we have any problem in designing a mind which is at least the equal of the human mind. But in terms of what we value the most as humans, which is to say our feelings, our emotions are sense of what the world is in a in a very personal way. Um That's I think means as much or more to people than their information processing. And that's where I don't think that ai necessarily will become conscious because I think it's the property of life. </s>

<s>[INST] also for your mind to yeah, just [/INST] like </s><s>[INST] yeah, yeah, this is something about clearing your, I think you've talked about this is like uh you're basically taking steps back before you take steps forward is I forget how you [/INST] call the wave. Yeah, under, you know, you have to go under, you gotta, you know, if you want to go above the line, you have to spend some time beneath it and yeah, I was definitely beneath the line for a long time but mountain, I mean like the interesting thing was as incredible as he is uh you know like what a monster. And I think if you had had him training and boxing you know for a long time and like from his youth, I think you know, the guy could be world champion but you know to be so specialized and then to switch your at a disadvantage and and also like I know from just fighting guys in the gym in tristar, some of those guys were way scarier for real like as scary as thor is like there's guys in that tristar gym that don't look like anything that would murder me much worse, much </s><s>[INST] worse. Yeah, but also you know that's the difference from being in the gym and under the lights to I mean gSB is an example of George ST Pierre is an example of somebody that maybe doesn't look. Yeah, terrifying, yep. He's he trains the tristar but he's quite he's super nice, super humble but it's terrifying when he's fighting is dominating people. You mentioned death and you're Canadian special forces and in general thinking about mortality, do you think about your death? Do you do you contemplate the the end that this thing that this ride ends [/INST] all the time? Yeah. From I've thought about death from a young age, </s><s>[INST] are you afraid of it? [/INST] Yeah I hate it. Yeah I don't want to die. Yeah definitely, definitely don't want to die. Um But there's times when I can rid myself of it. Yeah but for sure, I mean I'm not I'm not happy that you know death is inevitable and I'm not happy that potentially it's inevitable for all of us. Um But it does you know I like to to fight against it </s><s>[INST] that you if you if you could be immortal would you that's my only wish. Oh I see but here's the thing that's but the point is to have that wish it's uh it's like the all you can eat buffet at uh sushi it's that sushi is more delicious if you have a limit. Do [/INST] you have a stone. Well I don't know, I don't think I get sick of stuff. I'm very simple. Yeah. I don't think I would get tired of it. I really don't. Um, I mean if someone would pose it to you, do you want to live forever? You would choose No. </s><s>[INST] Yeah. Which is [/INST] no shoes. No. </s><s>[INST] Well, my answer is probably yes. Like no, I would uh it's more like the snooze button. Do you want to do, do [/INST] you you want to go to sleep? </s><s>[INST] But it's very difficult in the moment to go to sleep. But if if I'm allowed to live forever, I'm going to delay all the crazy. Like um, all the ambitious goals, all the, all the thing because there's always time. [/INST] But there is tomorrow </s><s>[INST] then, but there is tomorrow. But see I think that takes away from the richness of like the richness of the lived experience of just each moment. I think the richness of each moment comes from saying like I could die tonight. Like that, that it tastes delicious because you're going to die. I'm, I'm afraid if if you're not, I'm afraid all that goes away, all that magic goes away if you can live [/INST] forever. I don't know. But I'll tell you every time I have a near death experience, I think I'm going to die. I definitely live better afterwards. Yeah. Like uh it's always been that way. But yeah, no, </s><s>[INST] um that's why the stoics, you know, they really preach contemplating your mortality often. It kind of reminds you it get this whole thing could just end any moment and it makes you really appreciate Yeah, I don't know. I don't know. Certainly improving the quality of life. It is important. But part of me thinks that immortality is not as fun as um I would like to imagine. [/INST] Do you think that maybe you're and what you're building potentially is immortal? </s><s>[INST] Well that's what I definitely think about with robots. If they were to have a human life experience and be able to interact with humans in a deep meaningful way, I think they too have to be mortal in some fundamental way that means mortal. Like their ride has to end as well because they won't be able to interact with humans deeply unless that's the case. Um I like to have to have fear to have love the ability to suffer. The ability to miss somebody. I think scarcity is important. You have to be able to truly lose somebody. You have to be to fear things. You have to truly have the risk of destroying yourself and to have a sense of what it means to be a self. You have to be able to lose it. So if you're immortal you're just going to be, I feel like you're going to be like a toaster an intelligent toaster that just serves such [/INST] a negative perspective </s><s>[INST] on it, On the mortality. [/INST] Yeah. Just think well now you just you can get all those things done that you </s><s>[INST] want to do? I hope you're [/INST] right, I hope you're right. Yeah I I mean potentially you could invest even harder because you're like wow I'm actually gonna be able to get all this stuff done. </s>

<s>[INST] You're looking forward for the, what did he say? Let's make twitter more fun. I'm looking forward to to the fun [/INST] you've talked </s><s>[INST] about. You are at a prestigious university, [/INST] Brown </s><s>[INST] University, Brown University ah [/INST] and </s><s>[INST] you've mentioned that universities might be in trouble. I think it's with Jordan's [/INST] but everywhere </s><s>[INST] else, the barbarians are at the gate or the barbarians at the gate of the university. So, first of all, [/INST] what is </s><s>[INST] to you? Beautiful [/INST] about </s><s>[INST] the idea of the university [/INST] in America </s><s>[INST] of [/INST] academia </s><s>[INST] and what is a threat? [/INST] Well, you know, the university is dedicated to </s><s>[INST] the pursuit of truth [/INST] um and to </s><s>[INST] the education [/INST] and nurturing of young people as they enter into the pursuit of truth, to doing research into </s><s>[INST] teaching in [/INST] a environment </s><s>[INST] of free [/INST] inquiry and civil </s><s>[INST] discourse. [/INST] So, free inquiry means you go wherever </s><s>[INST] the evidence and [/INST] your imagination may lead </s><s>[INST] you and civil [/INST] discourse means that you exchange arguments with people when you don't agree with them on behalf of trying to get to the bottom of things. Um I think the university is a magnificent institution. It is a relatively modern uh institution, I mean, last five </s><s>[INST] 100 years or so. [/INST] I mean there are universities that are older than that, </s><s>[INST] but the [/INST] great research universities of the world. And they're not only here in the United States are places where human ingenuity is nurtured, where new lot knowledge is created. </s><s>[INST] Um and [/INST] where young people are equipped </s><s>[INST] to uh answer [/INST] questions that are open questions about our existence in the world that we live in. You can trace to the university much, if not most, of the advances in technology </s><s>[INST] and resourcefulness [/INST] and our understanding of the origins of the species, of the nature of the universe, cosmology, etcetera, of science, of the </s><s>[INST] pursuit of [/INST] humanistic, understanding, the nurturing of traditions of inquisitive. Also, that's the university barbarians at the Gates. </s><s>[INST] The [/INST] people who are trying to shut down open inquiry at the university on behalf of their particular view about </s><s>[INST] things [/INST] are a threat to what the university stands for and they should be resisted. So </s><s>[INST] if [/INST] I'm enquiring about the nature of human intelligence </s><s>[INST] and [/INST] I want to study differences between human populations and their acquisition of or their expression of cognitive ability. That's fair game. It's an open question if I want to know something about the nature of uh gender affiliation and identity and gender dysphoria and whatnot. That's fair game to study in the university, you can't shut that down. You shouldn't be </s><s>[INST] able to [/INST] by saying, I have a particular position </s><s>[INST] here, I'm a [/INST] member of a particular identity group. </s><s>[INST] Suppose I want to study [/INST] the history </s><s>[INST] of colonialism. [/INST] Uh and there's a narrative on the progressive side, which is colonialism is about Europeans dominating and stealing or whatever, whatever. And I happen to think, well, there's another aspect to the story about colonialism, to which is that it's a mechanism for the diffusion of the best in human civilization to populations that were significantly lagging behind with respect to that. It brought literacy through the southern hemispheric populations that were dominated in the process of the colonizing thing. It's complicated. I'm not taking that position by the way, I'm just saying somebody at a university should be </s><s>[INST] able to [/INST] take it up and pursue it and engage in arguments with people about it. I'm talking about race and ethnicity, but this extends to a wide range of things supposedly talking about climate. And one person says the earth is endangered because carbon in global warming, etcetera. Seven and another person says, no, wait, no, wait, </s><s>[INST] look at where we [/INST] Stand in the 21st century were vastly richer than our ancestors. Just 250 years ago, we have much more knowledge about that and so forth, and so on. 250 years from now, human </s><s>[INST] ingenuity will have [/INST] devised in ways that we cannot even begin to anticipate all manner of technological uh means for managing the problem. There's no reason that we should shut down industrial civilization today, because we fear the consequences of it when in fact we are vastly richer than our ancestors and those who come up two centuries after us will be vastly more effective at </s><s>[INST] dealing with problems than we [/INST] are. Now. Let's it's, you know, etcetera. I'm not actually making that argument, I'm just saying the tendency </s><s>[INST] to try to say, oh [/INST] no, that person is a climate denier. They can't pursue that area of inquiry uh, is against the spirit of the university. I think the barbarians at the gates has to do with the people who think they know what the right side of history is and try to make the university stand on the right side of history. My position is, you don't know what the right side of history </s><s>[INST] is [/INST] and the purpose of a university is to equip you to be able to think about what is the right side of history, What is the solution to the dilemmas that confront us as human beings living on this planet with the billions that we are in the condition that we are. So um, the identitarians, the ones who want to make the university called </s><s>[INST] tao [/INST] to their particular understandings about their own identity. Um, we now have </s><s>[INST] at the at [/INST] Brown University and various other places. We don't do columbus day anymore. We do indigenous peoples day when that day comes up in </s><s>[INST] october, we [/INST] don't talk about columbus, they're taking down statues of columbus all across the country and so forth and so on. I'm not arguing anything here other than that, the latter day position um </s><s>[INST] by [/INST] parks black indigenous and other people of cover the latter day position that the university has to reflect a particular sensibility about these identity </s><s>[INST] questions. [/INST] I think it's a threat to the integrity of the enterprise. </s>

<s>[INST] You mentioned. This is what made you want to get into family medicine As a Dr. What do you think about the saying? Save one life, saved the world? [/INST] Um, </s><s>[INST] this was always moving to me about doctors because you have like this human in front of you and your time is worth money. You're what you prescribe and your efforts after the visit are worth money. And uh, it seems like the task of the doctors to not think about any of that or not, not the task, but it seems like a great doctor despite all that, just forgets it all and just cares about the one human. And somehow that feels like the love and effort you put into helping one person is the thing that will save the world? It's not like some economic argument or some political argument or financial argument. It's a very human dr that ultimately is behind all of this that will do good for the world. [/INST] Yes. I think that's true. And at the same time, I think it's equally true that all physicians need to have a sense of responsibility about how the common resources are allocated to serve all the whole populations interests best. That's attention that you have as a physician. Let's take the extreme example. Let's say you had a patient in front of you who if you gave a 10 1 $10 billion pill too, you would save their life. I would just be tortured by that as a physician Because I know that $10 billion dollars spent properly in an epidemiologically guided way is going to save a whole lot more lives than one life. </s><s>[INST] So it's also your responsibility as a physician to walk away from that patient. [/INST] I I wouldn't say that. I think it's your responsibility to be tortured by it. That's exactly </s><s>[INST] right. Uh, the human condition. Uh, [/INST] that's a tough </s><s>[INST] job. But yeah, yeah. To maintain your humanity through it [/INST] all. But you've been asking at different points in this conversation, why are doctors so complacent about the tremendous amount of money we're spending? Why do they accept knowledge from different sources that may not pan out when they really know the truth? And the answer is that they're trying to do their best for their patients. Yeah. And um there's this, it's the same kind of torture to figure out what the hell is going on with the data and that's a sort of future project. And maybe people read my book and maybe they'll get a little more excited about it and become more legitimate and practice. I would feel like my life was worthwhile if that happens. But at the same time they've got to do something with the patient in front of them. They've got to make a decision. And they probably, there are not many weirdos like me who invest their life in figuring out what's behind the data. They're trying to get through the day and do the right thing for their patient. So they're tortured by that decision </s><s>[INST] to And so if you're not careful, big pharma can manipulate that that drive to try to help the patient that um that humanity of dealing with the uncertainty of it all. Like what is the best thing to do? Big pharma can step in and use money to manipulate that humanity. [/INST] I would state it quite differently. It's sort of an opt out rather than an opt in? Big pharma will do that And you need to opt out of it? </s><s>[INST] What advice would you give to a young person today in high school or college stepping into this complicated world full of advertisements, of big powerful institutions, of big rich companies, how to have a positive impact in the world, how to live a life they can be proud of, [/INST] I would say should that person um who has only good motives go into medicine, they have an inclination to go into medicine. And they've asked me what I think about that, given what I know about the undermining of american health care at this point. And my answer is if you got the calling, you should do it, you should do it because nobody's going to do it better than you. And if you don't have the calling and you're in it for the money, you're not going to be proud of yourself. </s><s>[INST] How do you prevent yourself from doing from letting the system change you over year, over years and years like letting, letting the game pharmaceutical influence affect you? [/INST] It's a very hard question because the sociological norms are to be affected and two trust the sources of information that are largely controlled by the drug industry and that's why I wrote sickening is to um to try and help those people in the medical profession to understand that what's going on right now looks normal, but it's not the health of americans is going downhill. Our society is getting ruined by the money that's getting pulled out of pulled out of other social, socially beneficial uses to pay for health care that is not helping us. </s><s>[INST] So fundamentally the thing that is normal. My question, the normal don't uh if you conform conform hesitantly [/INST] well you have to conform. You can't become a doctor without conforming. Um I I just made it through. Uh but there are many and it's hard work. Um but you have to conform. And even with my colleagues in my own practice, I couldn't convince them that some of the beliefs they had about how best to practice weren't accurate. There's one scene uh younger physician had prescribed hormone replacement therapy. This is back in 2000 2001 had prescribed hormone replacement therapy for one of my patients who happened to be a really good personal friend. And I saw that patient covering for my colleague at one point and I saw that her hormone replacement therapy had been renewed and I said, are you having hot flashes or any problem? No, no, no, no. But uh dr so and so said it's better for my health. And I said, no, it's not. The research is showing that it's not it's harmful for your health. And I think you should stop it. So my colleague approached me when she saw the chart and said, wait a minute, that's my patient, Maybe your friend, but it's my patient. And I went to a conference at my from my Alma Mata Medical School and they said that healthy people should be given hormone replacement. And I said there's got to be drug companies involved in this. And she said no, no, no it was at my university, it was it was not a drug company thing. We didn't go to a caribbean island. I said do you have the syllabus? She said yeah and she went and got the syllabus. And sure enough it was sponsored by a drug company. </s>

<s>[INST] Yeah. And so we're not talking. I guess this conversation is several layers. One is for the olympics and one is for like, what is it? Active athletes? They're like amateurs, whatever, whatever category. I put myself into like, people that exercise regularly. And then, um, maybe people and then there's people who like exercise rarely. So on all of those fronts. I mean, do you think it's possible to live a happy, uh, active life eating meat only, Or mostly meat. Yeah. What have you learned about [/INST] This? Yeah, I think so, for some context, like I followed what I would call a low carbohydrate diet for the last 10 years and just like kind of the training, I paradise it to a degree where there are parts of my training where I do bring back a little more carbohydrate and there's periods of my training, especially in the off season where I'm like very low and I might be like kind of in that ballpark of like ketogenic, strict hygienic or no carbohydrates for periods of time. </s><s>[INST] What kind of food are we talking about? What's the strict low carb diet? [/INST] I've arranged everywhere from like mostly plant based, low carb keto to like mostly animal based. I very rarely gone much more than like two weeks strict, where it's like I'm strict carnivore or strict plant based or anything like that, we're talking probably more like 95% at the, at the peak. Um in terms of any type of like longer lasting uh for my personal experience of like being like either in like the animal food camp or her, like the plant based camp kind of of a process. Um so I've tried all of them things that stayed consistent over the 10 years as a kind of macro nutrient profile that I've done throughout the course. So </s><s>[INST] one didn't win over the other until the meat based versus plan based. [/INST] So for me meat based, definitely, what was, I mean, I was, I was my highest meat consumption in 2019. And that was by far my best racing season. </s><s>[INST] Yeah. Keep, keep coming back to that year. That was a good year. For many reasons, philosophically, nutrition. [/INST] Yeah. 2020 happened and now I haven't had a really good chance to uh improve. We'll see. Hopefully I've got some more, some more in the tank. </s><s>[INST] That's strange. There's most athletes that competed your level have more carbs integrated into their diets. So what have you learned about using me in high performance? [/INST] I think it's maybe less about the meat and it's more about like, what are you, what is it replacing? So if we go if we step away from like me specifically and just like the people that, because I mean, we're getting to the point, I get it's anecdotes but like, like that's what we have at the moment because I mean, there is actually a study being done on like, I think, I guess they call hyper carnivore where they're like, I think above 80% of the intake from </s><s>[INST] meat, um [/INST] and they're looking at a few different things there, but </s><s>[INST] it's so weird and keep interrupting, but it's so weird that it sounds unhealthy hyper carnivore, but it makes me feel really good. So the individual [/INST] thing, right? Like </s><s>[INST] individuals, [/INST] There's countless people now who like, and I'm not saying that they could not have found another route myself included. Like in 2011 when I switched from moderate to high carbohydrate, low carbohydrate and saw some very noticeable differences in the way I felt the way I performed in all this stuff, that doesn't mean that there wasn't another path, I just did not find that path. And the, the, the fact that I found a path that was producing the results I was looking for is really all that matters in my mind, you know, like, I don't really care if there was a parallel path that works just as well or, you know, something like that, because ultimately we only have one shot at everything in doing so, like, It would be great if I could go back and try four or 5 different </s><s>[INST] things. Well, the annoying thing is that the body adjusts to whatever the heck you're doing, so it's hard to do good science, even on yourself, [/INST] I've referenced my 2,019 racing season a few times and it's like, it would be silly for me to put all of the emphasis on my nutrition plan for that, because it's also comes with two decades of endurance training. So it's possible and it's very likely that a huge portion of that success was just the culmination of a lot of work overtime from the training side of things. I just think like anytime you hyper focus on one area or pick a couple variables and just target those, You find yourself in a position where you are, you're putting other things in the most un charitable light possible. So, so then you have this situation where like it's actually a combination of a variety of different things, so where are the big movers? And you know, for me, nutritional shift was pretty clear that that improved my sleep in my recovery. And I mean, people can say, well there's the placebo effect, which is a very real concern, but, you know, for me personally, a 10 year placebo effect would be a quite lengthy placebo effect. Um and I do think it's individual, that I I emphasize that a lot because I mean, I've worked with tons of people with this and I do see a range from person to person, I've worked with people who come to me and they're like strict keto, and we raise up their carbohydrates a bit and they're like, okay, I feel way better doing it this way, and I've worked with people who, they come to me moderate carbohydrate, but they're interested enough, um they want to try a lower carb, so we, you know, we titrate them down and I've had clients where I'm like, okay, I'm gonna give him this workout and they're gonna wish they brought back a little bit of carbohydrate and then they go, and they nail the workout and I'm just like, baffled that because, because they're different from me and every time, you know, when you have your own personal experience the first kind of guttural responses if I had done it would have gone this way, why did it go complete opposite way for them? And you kinda have to just kind of step out of your own perspective a bit and say like, okay, well they're different, you know, for whatever reason they're getting getting along like this, </s>