Genetics, China & Artificial Intelligence | Stephen Hsu

Steve Hsu, a physicist and expert in genetics and AI, joins Siddharth for a wide-ranging discussion. They cover topics like genetic influences on human intelligence, China's rapid development, comparisons of the Chinese and American political systems, the future of artificial general intelligence, the possibility that we are living in an ancestor simulation, and more. Steve provides insightful commentary throughout, drawing on his deep knowledge of physics, biology, technology, and history. An intellectually stimulating conversation about genetics, geopolitics, and the future of intelligence on Earth.

Siddhartha Socials:

• Youtube
• X
• Instagram
• TikTok
• More Socials

Chapters

• 0:00 Introduction
• 2:21 Steve's Encounter with Richard Feynman
• 3:31 Genetics and Human Improvement
• 11:08 Role of Genetics in Disease Prediction
• 18:10 Influence of Genetics on Behaviour
• 21:37 Genetic Selection in Embryos
• 39:24 Role of Genetics in Addiction
• 41:53 Importance of Individual Differences and Success
• 46:36 Value of STEM in Indian Culture
• 48:02 Importance of Non-Academic Skills for Success
• 49:01 Embryo Modification
• 52:13 Quest for Immortality: Brian Johnson's Story
• 58:03 Role of Genetics in Aging
• 1:12:20 Impact of Genetics on Society and Policy
• 1:17:19 Rise of China in the Global Stage
• 1:53:57 Future of AI and the Impact on Jobs
• 2:02:37 Possibility of Living in a Simulation

Transcript

2:01

All right. We are live Mr. Steve. Thank you so much for joining us today on this podcast. It's my pleasure Siddharth. Yeah, I actually, so I started like, you know, I was starting out, I was following you on X and you know, I kind of saw your podcast with this guy called Durkesh, Durkesh Patel. And I saw you talk about like the first thing that kind of blew my mind away that you had the opportunity to meet I was like, wow. I'm like, because I grew up thinking, reading about him books and like you really met him. That was really cool. Like, was he a professor? Like, how did you meet him? He was, he was my professor at Caltech. In fact, the reason I chose that university for college was largely because of Richard Feynman. That's incredible. Like, there's this technique called, like, I think thinking like... from the first principle or something, it's called like the Richard Feynman technique. In India, it's really popular because a lot of kids, you know, like they study physics and math and whatever and they talk you, they teach you like how to think from the fundamental principle. So, I always kind of found him to be a very interesting person. So, when I saw you met him and then you studied physics, go ahead, were you saying something? No, go ahead. Now I saw that you had a background in physics and you are excellent in that but you talk a lot about you know in vitro fertilization, genetics and how we can kind of you know improve the genetic code and improve the human race and that was kind of blew my mind away and because I will really like read about it but it's kind of like a taboo topic so I avoid it but I was like you know what I want to talk about you, talk about this so like let's get him on and I really appreciate you being humble enough and saying yes. I'm happy to be here. Did you say on Twitter or on X that this is going to be live streamed? Are there actually people watching us on live stream right now? We are live on Twitch and we are live on X. So I am starting out only now, so I am not super popular, but we have like one viewer. But on my last video, I got 8,000 views. So there is hope. So... I'm growing. I'll definitely will get a huge viewership as we, you know, kind of move forward. Awesome. Well, welcome to our OneLive viewer. If you have any questions, happy to answer them. Incredible. So let's get started. All right. So I have a very interesting set of questions that I made for you. When I was doing research on you, I saw that you have been doing cognitive labs. Now, not a lot of information was available on it, but this is what the internet told me. You are comparing DNA of extremely intelligent people to figure out what makes them smart. And I want to learn a little bit more about how this works because one of the big things is like everyone wants to be smart. Some people think I try you work hard. Maybe you read books and you try to get smart. But a lot of the research, it's like brings an uncomfortable truth that there's some sort of genetic component to human intelligence. So can you tell me a little bit about what this research is all about? What are you trying to do? And like, what is this research? What are you trying to learn from it? Well, to start from first principles, which I think you like. We have to ask the following question. Why are humans different from all the other animal species? And the answer is that we have probably the most powerful brains of any animal species on the planet. And as a consequence, we've been able to more or less take over the planet for better or for worse. And if you ask, how does natural selection work to cause evolution? Nature is selecting on the fitness of the organism, but the fitness is a consequence of the genetic code of that particular organism. So over time, the reason that humans were able to evolve on the planet from earlier species that were not nearly as intelligent is because there are changes you can make to the DNA, which cause the animal to have a larger brain, a more powerful brain, a better structured brain. And so I think everybody accepts that now. I think unless you're, you know, because of some religious reason or something you don't believe in evolution, you sort of accept that somehow humans evolved. It is DNA that makes us different from monkeys and apes and lizards, and we are smarter than they are. But then you have to ask among humans, so within the species of humans, Are there individual differences? In other words, could there be slight differences in DNA between one person and another that makes one person intrinsically smarter than the other person? It's a very fundamental question. Now, you could study this in many different ways. The old style method of studying this is to look at things like identical twins. Do identical twins have similar levels of performance in school? or on standardized tests? Are they more similar to each other on these measures of brain power than two brothers that only share half their genes or two cousins that share an even smaller fraction of their genes? So, you know, humans, before we have the technology to sequence DNA, humans were already trying to get to the bottom of this, to try to understand to what extent are individual differences incognito ability due to differences in DNA or due to differences in the way the person was brought up. Maybe access to school, access to books, access to Richard Feynman, which took those different effects was contributing more or less to the ultimate level of adult intelligence in an individual. So, I think there's a fair amount of evidence that there has to be a genetic component. It is not everything, but it is a significant contribution to the level of adult intelligence, the differences between people. And so what better mystery to try to solve once the technology exists to inexpensively read out the DNA of large numbers of people. Can you look at very large numbers of people and somehow figure out which aspects of their DNA are reflected in their level of intelligence? That's the fundamental question. There are different techniques to try to do this. One is the brute force way, which is you just genotype as many people as you can and you measure the trait that you're interested in, in as many people, in each of those people. So you just try to get massive statistics. Another way to do it is to look for outliers. Look for people that are unusual, either unusually smart or unusually not smart, and look and see is there something special about their DNA that might make them unusually smart or unusually not smart. So these are the kinds of things that we were working on at the BGI Cognitive Genomics Lab. There are now many different collaborations of scientists who are working on this problem. And there's a sort of general program now because we now have millions of human genomes that have been read out. And the goal is to basically build predictors for every human trait. So how smart are you? What is your eye color? What is your skin color? How susceptible are you to... breast cancer, how susceptible are you to heart attack? All of those things are what we call phenotypes or traits, complex traits, that at least are influenced to some degree by genetics. And the goal is to figure out exactly which individual letters or positions in your genome are actually influencing those complex traits. basically it's like my DNA you like how do you guys correlate it though like is there like certain like variable you look at in my DNA and then you try to connect it to intelligence because to me this is like a must be a very complicated research that you're looking at like because I I'm a dumb guy right I don't know like how what do you look at in a DNA and like all right the IQ or smartness of this person. Can you speak to that a little bit? Yeah, so as you know, your DNA is basically a large chain or ladder of amino acids. And so it's individual chemical, what they call base pairs, which make up the individual components of your DNA. So the DNA is itself a molecule. And the structure of that molecule varies only a little bit between two people. So There are about three billion base pairs in your DNA. And you and I would differ at about one in a thousand locations. So if we just randomly chose a thousand of those base pairs, you and I would agree at about roughly 999 of those. And there might be one that we, that you have an A there and I have some other letter there. Right? So that is the genetic code. And somehow nature is this machine. our cells, our machines, which express that code that's in the DNA. To figure out whether a particular position, a particular base pair on your genome is affecting your intelligence or affecting your height, we need absolutely enormous statistics. So we say like on average, are the people that have the letter A there a little bit taller or a little bit shorter? than the people who don't have the A there. And literally that's what's going on. Now, of course, the human brain is not good enough to do this analysis, so we have to use computers. And so, you can imagine that in a world where I can teach GPT-4 to speak every human language to like spew out a chapter of a novel on request. Obviously we're going to use very powerful computer algorithms to take what we call the phenotype data, so that would be like the height of each individual in the sample and then the DNA string for each individual in the sample and try to build a function, a map between a particular string and a particular value of the phenotype. So it's a machine learning problem. which now you're maybe not surprised that someone trained in physics would be useful to try to solve this problem because it's starting to look more and more like a kind of mathematical problem. And that is more or less what we do. And we use, most of the work published by my lab has used the supercomputer at Michigan State University where I'm a professor. So it has become this modern field. It has become what we call computational genomics. We're doing genomics, but we're using heavy computation to do it. Interesting. I think one of the big things like, you know, I strictly jumped into the conversation for our viewers. Like there is like a big debate that, you know, where a lot of people think that your ability, your smartness or your IQ simply can be improved. If you work hard and sometimes, you know, when people drop the idea that, it's genetic, it sometimes can be a little depressing because it kind of tells you that hey man you just lost a lottery and life is going to be hard. Like I was reading some research online which says that your IQ is a very strong predictor of your success in human life. Like if you have a high IQ, you probably will make more money, you'll get better jobs, you will generate better, you get better house and overall you will be more satisfied in life. Let's say you guys are doing this research and let's say you understand that okay these things are leading to this kind of IQ. What can be done to like let's say and we believe that okay there's a strong genetic component from a scientific point of view. What guys can you do with this research to make let's say people's lives better who let's say lost this kind of genetic lottery because It seems like a lottery is like, you got a good IQ, well, all the best, your life works out now and you lost that IQ, now you're gonna be maybe flipping burgers at McDonald's for the rest of your life and end up with credit cards and loans and stuff. So like what kind of solutions, like let's say you do a research, what kind of implementation are you looking in terms of applying this research to make lives of people better? Can you speak to that? Yeah, absolutely. But first, Before I answer the question specifically for IQ, let's just, to clarify our thinking a little bit, let's suppose we replace the word IQ by height. Okay, so we say, oh, wow, it's kind of depressing to think that your genes kind of determine your height. I always thought if I worked harder, I could become taller, right? Maybe I could do special stretching exercises and drink a lot of calcium and eat a lot of steak and I would get taller, right? Yeah, protein. Now, certainly if you don't get enough calcium and protein and vitamins and exercise, maybe even sunlight, as a child while your skeleton is still developing, or alternatively while your brain is still developing, then yes, maybe I can depress the potential that someone has. They don't achieve their full potential. But if I give you a good diet, good exercise, good sleep, good nutrition, and then it's plausible that the upper limit on how tall you're going to grow is really based on your DNA. Okay? And some guy who ends up two meters tall versus some guy who ends up 1.7 meters tall, there really isn't a simple way to bridge that gap. Right? So you could imagine that, right? Now, it could be that IQ is more like that, or it could be that IQ is some really complicated trait, and even when you're 30, regardless of your genetics, you can still drastically improve your intelligence level. I think the evidence is a little more on the side of saying IQ is a bit more like height than something which is infinitely malleable, okay? But... it's still a subject of research, okay? So I don't wanna come across as saying like, I know for sure, like once you're 18, no matter what, after that you can't increase your IQ. I'm not saying that at all, okay? But clearly some people are born with some advantages, probably due to genetics, that just let them learn faster, let them remember things better, let them structure the information better in their head. So I think most people would agree that that's the case. And now of course life isn't, completely just determined by your IQ. So I think you said, oh, if I don't have a high IQ, I'm stuck flipping burgers the rest of my life. But you might be surprised that a lot of really successful people are only a little bit above average in their intelligence. And there are plenty of really smart guys who are not successful in life. So, you know, life's going to worldly success. And there's the willingness to work hard. There's creativity. There is being good at communicating with other people, being outgoing, being willing to seek out the things you want. So there's lots of factors in here. So it isn't that depressing that, you know, so, oh, I didn't score in the top 10% on my IQ test. My life is gonna be crap. No, it doesn't really work that way, okay. And there are plenty of very smart people who are maybe lacking in other. characteristics who ultimately are not successful in life. So nothing is that. But nevertheless, people are interested in intelligence, right? And it is just an incredibly fundamental question to what extent it is determined by DNA versus something else. And so for the sake of conversation, let's always assume we're talking about a kid who is given good nutrition, loving parents, access to books, access to school. So somebody who has a good environment, let's not talk about the kid who got hit in the head with a hammer every time. Oh yeah, okay, yeah, definitely. Let's assume there's some basic level of good environment that the kid has access to, and then we'll talk about to what extent genetics maybe limits what that kid is capable of. Maybe the kid will never master calculus, or maybe the kid will never master advanced programming languages. Okay, so we can talk about that. Now, Are there ways to transfer that luck? The lucky kid learns calculus very easily at a young age. He masters computing languages very fast. He speaks five languages, whatever you want. Is there a way to that we can call that kid the lucky kid? And then the other kid maybe struggled in school and had to work really hard and still could only get an average, grade point average, right? Yeah, that is me. Okay, well, don't worry because success in life is not completely determined at all by your academic ability. Okay, but you could say the first kid I described is luckier than the second kid, right? Fair. Is there a way to redistribute that luck? So first of all, it could be, this is actually already possible, that the unlucky kid when there might be some technology that allows their kid to be smarter, genetically smarter than his mom and dad. Okay, so that's one way to transfer luck. Of course, it's across generations. There might someday be some kind of drug or some kind of gene editing which could work on an already formed brain to improve its performance. That's a little bit more speculative. We don't actually know whether any of that is possible. However, we do know that through the kinds of genetic technologies, like embryo selection that my company, my startup is involved in, you can actually transfer genetic luck to the next generation. So that is a new thing that before just a few years ago, that was not possible. Understood. And I think I'll also add that so nobody misunderstands me. that the goal here is not to like, you know, create a depressive episode that you lost a gigantic lottery. And I think the reason I started with the IQ thing is because it's more controversial. So I was like, you know what, let's address the elephant in the room first. And I think definitely, I think like when you're trying to predict IQ because it's so much, you know, connected to people's self-esteem and the politics around it makes it ugly. But let's say at the same time, you're trying to predict, hey, what is the probability of getting schizophania or something or diabetes or something. I think then a lot of people would appreciate the knowledge of genetics because if you could tell me that this kid has chances that he may struggle, I don't know, with autism or mental health problems whereas compared to this kid, you know, at an embryo level, this person is going to be healthier. So, I think any father, I'm saying if I am a father and I'm having a children, I'll be like, you know what? Yeah, give me this kid which will not suffer those consequences because otherwise your life will be hard. So disclaimer to that, let's talk about the embryo thing you're doing. I was looking, I think the work you're doing on an embryo level, I think the name of the company is called Genomic Production, correct? Perfect, perfect. So... Let's talk about it. Like, can you look at an embryo? Is it similar to what the work you were doing with BGI as well? Can you look at an embryo and is it like the same technique, the same way you're looking at the DNA of smart kids? Can you look at an embryo and be like, all right, this embryo seems like this may have a higher chance of getting XYZ disease or have XYZ traits and then you can kind of alter it? Yes, so the step that's involved here is to remove at least one cell or at least a few cells from the organism. And as you know, the DNA in every cell, every cell in your body has the same DNA. Okay. So if I were to take a cell from your body and then put it through a gene sequencer or a genotyping machine, I could read out that string of amino acid letters, base pairs. that reflected your genetic information. Our company was the first to do that with embryos. We showed that it could be done with embryos without harming the embryos. And therefore, if a family is going through IVF and they have multiple embryos, so it's not that unusual for a company to have multiple embryos and they have to make a selection. So embryo selection, okay? The embryo selection problem. They have more than they're going to use. They have to pick one. Which one should they pick? If you don't know the genetic information for each embryo, all you have to go on is what they look like, and they all look like little soccer balls. Okay. So, you're either kind of making a random choice, or if you did do a little biopsy and read out the genome, you could make a more informed choice. And so the company gives the physician, the IVF physician who advises the parents a report on all of the disease risks, risks for major disease conditions for each of the embryos. We actually don't report any IQ prediction. We don't report height. We don't report whether you're gonna have light skin or dark skin. We don't. Oh yeah. No cosmetic traits. Fair. We only do the disease conditions because I think as you just said, that's the ethically or the least controversial part of the whole thing. And so we don't wanna throw out the baby with the bathwater. We don't want to do something really controversial and have people attack the company when there are a bunch of uncontroversial, very positive things that the company can do for the parents. So that's kind of our philosophy. Yeah, I think you're right about it too. And I think it's, I think, yeah, I think here is my fault too. I think like, like you have like, I have like a monkey brain and whenever you do, you know, any research about genetics, all this weird stuff online shows up where, you know, oh, like you already know this, but I watch a lot of podcasts and there's this guy, Charles Murray, and he wrote this book, Bell Curve. And the moment you write genetics on the internet. and it's like a nuclear bomb has exploded and it's like dangerous Nazi and all these dangerous words come up. So like and my monkey brain also got attracted towards that and ran there. But it's important to I think educate the society that beyond this IQ, beyond this color of skin, there's a lot we can do here without generating any controversy and genuinely benefit the human race. Yeah, I mean to give you the most sort of easy to understand scenario, There are many major diseases that run in families. Okay? So if you went to your family reunion where your grandma and grandpa are there, and your uncles and your grand uncles and cousins, and you just took a health survey, said, hey, what did great auntie Prana die of? And they say, oh, well, she had a heart attack. Okay, oh, okay. Well, what did grandpa? you know, Suresh Dayav. And they're like, oh, he had a heart attack, right? So if you actually look, it is very common for there to be particular diseases that your family is more prone to. And what's actually typically happening there is that you have high polygenic risk for those disease conditions, and you're kind of passing it down from generation to generation. In fact, your doctor will, when your doctor is doing diagnosis of your health, they often are asking you to fill out a family, it's called a family history of like what conditions like, oh, who else in your family had diabetes or who else in your family had, you know, heart disease. So we often meet in the, in, in genomic prediction, families that are, they know that they have a family history of something and they don't want to pass it on to their children. So what they really want to do is make sure the embryos that they select that will become their kid has normal polygenic risk rather than elevated polygenic risk. And typically in those families, because they have a family history, when we look in and we look at the scores, we can see the average among the embryos is elevated compared to the general population. So that's what the family history means. But usually since there are fluctuations. at least one or two of the embryos will typically be in the normal range or even in the low risk range. And then they could say, look, give me the low risk. I'm sick of having people in my family die of heart disease. Give me the low risk embryo for heart disease. And so I don't think there are too many people who would complain that a family that has had bad luck regarding heart disease risk wants to get rid of that bad luck. Who would complain about that? So that's a kind of relatively uncontroversial thing the company can do. I think the knowledge of genetics, I think it's extremely empowering to know because like the immediate benefits you can get from this is incredible. Like before I was researching about you, I did not even know that IVF was a thing and that I could look at an embryo during the reproduction level. And I can like save yourself from so much trouble and problems which you otherwise will end up getting stuck. You will lose money on medical bills and the emotional trouble you have. Why do you guys not do more marketing? I think like you are incredible at physics and research. I think you need to get like a sales guy who can like educate people about like the benefits are outweighing the politics. Like this heart disease thing, I'm pretty sure there's something in my family going on and I don't even know about it. Can you look at my DNA right now? Like not right now but if I gave you a sample of an adult person and you can predict. So in principle, so the exact same technologies that we've built can be used to get a polygenic score report for an adult. It doesn't have to be an embryo. In fact, adult is easier because it's easier to get the sample of DNA from the adult than from this little tiny microscopic embryo. There are other companies that are trying to, trying to roll this technology out for adults. And in fact, If you look at my Twitter feed, I was just tweeting about a very big article that appeared in Nature about a big collaboration in the United States funded by the National Institute of Health. It's based on a big bio bank, gene bank called All of Us, and the very first report on the pipeline where they use the information from this bio bank to build predictors, and they've built a pipeline that can... calculate risk for many different diseases for individuals. And then your doctor then gets the report and says to you, hey, you better start jogging because I notice you're high risk for heart attack. All of that is coming. But here's the thing. And this is a very cynical thing to say, but it's the reality that we have to deal with. So I'm just going to tell you, most doctors, if there isn't big innovation, I'm not talking about a small innovation, but if there's a big innovation in medical science that happens after they leave medical school, very tough for them to adapt to it. When I'm describing to you this whole breakthrough, I'm talking about being able to read out the state of a little molecule. I'm talking about machine learning and AI on millions of people to try to figure out how to map individual DNA sequences to phenotypes. None of that existed 10 years ago or 20 years ago. And it's pretty complicated science. It's not simple. It's not conceptually simple like, oh, we invented this drug. If you take this drug, you won't be hungry anymore and you'll lose weight. Oh, sound perfect. Yeah. A doctor can prescribe that. He doesn't have to like go back to school in order to learn how to prescribe that drug. He just reads some guidelines. and says, okay, if the patient fits these guidelines, I'll prescribe the drugs, these are the side effects, we gotta watch out for the side effects, we'll watch this guy. On the other hand, if I say, well, you know, conditional on being above the 98th percentile for atrial fibrillation risk, the chances of stroke for this individual between the ages of 50 and 65 go up by seven times. Oh wow. we're starting to talk about pretty complicated mathematical things, right? Like you have to have whole tables summarizing the meaning of all these different scores that we can compute and what is the evidence for this. We checked the predictor was validated in this population of Canadians, at this population of people in Estonia, at this population of people in India. So the science is very complex. And so It's understandable that the medical system is taking a long time. Well, it seems to me, maybe to you, it's taking a long time for them to figure this out, right? But it's understandable because it's fundamentally a pretty complicated advance in the science. It's not, it's not a minor advance. It's a really complicated one where, Oh, I actually read out your entire blueprint, your genome, and I'm using it to predict what disease conditions you're likely to have. Right? So that's a, that's a huge breakthrough. And so anyway, it's going to take a little bit of time. But if you look at this, all of us, collaboration, the paper that just came out, there's a similar project going in the UK. They are in preparation to get to the day where this becomes standard of care. So basically, everyone is using, every doctor is using polygenic risk scores to help. You know. improve the health and the life expectancy of their patients. So that day will definitely come. But for an entrepreneur like myself or a scientist who's right at the cutting edge, it seems like it's painfully slow. Maybe, you know, all the work these guys are doing, you know, they're talking about AGI, you know, OpenAI and ASI, maybe that will, you know, like a lot of people, there are a lot of scientists. I don't, like, I know you have a background in AI too. and you have a good understanding of how this industry works. But I'm looking at some scientists, like there's this guy called Geoffrey Hinton. He is... Do you guys know each other? Like have you... I don't know Geoffrey Hinton, but I'll tell you this. There are a number of founders and core people involved with OpenAI that are investors in genomic prediction. So I do know those people. Oh, wow. Interesting. Siddharth, you're breaking some news here. Really? Interesting. So you're saying like some of these big guys are also creating artificial intelligence at the same point they're investing in research that can kind of help us understand the DNA as well. Yes. I mean smart people again look back to this intelligence thing. If you're smart enough to do AI research you're also smart enough to easily understand what I've been explaining to you this evening about DNA and machine learning applied to DNA, and then what can you do with it after you built the predictors? So it's not surprising that people like that would be interested in being investors in a company like Genomic Prediction. So our investor list is like a who's who. I'm not gonna say who they are, but it's a little bit like a who's who of top technology people. Interesting. I think if the fact that they're investing in it, because I think we can talk about many things on the internet. The fact that people are putting money into it, that means it's something serious. Like you don't put money like millions of dollars into something that's not gonna work. Well, okay, so obviously in the world of venture capital and investment in tech companies, nobody can be absolutely sure something is gonna work. Yeah, obviously there's some stuff involved, but the smart investors at least can judge. They know enough science, they know enough about AI. They can judge whether it's plausible that something is gonna work. In our case, everything's been validated. We've had many babies born who have been polygymically screened. We published many papers in top journals documenting how our stuff works. And in fact, the whole field, when we started, so we published the first accurate predictor for human height in 2017, so already seven years ago. And at that time, the number of papers per year, so if you aggregate over all scientific journals, you say how many papers a year are doing things with polygenic risk scores or polygenic scores, it was probably, I don't know the exact number, but could it be less than 100 papers a year being written worldwide on the subject? And now the number is probably on the order of at least a thousand or a couple thousand papers a year. So it's now a very big field. So almost every major university, if you go there, you'll find some researchers who are working on this stuff now in 2024. So none of this is speculative anymore. And in fact, the same kind of thing that we've been doing with the embryos are what now the U S and the UK government, NHS and IH are preparing to do for adults. So there's no question that this stuff works now. I think scientists are not in disagreement about whether, whether one can predict disease risks. based on genetics. I think that's pretty well established at that point. Like, okay, so let's say you're talking about diseases, you know, sometimes, you know, like again, you are on a different IQ level. So you talk, you are talking about these serious diseases on a normal level like mine. There are a lot of people, you know, struggle with things like addiction. They drink too much alcohol. They have bad, you know, cigarettes or sometimes they even eat sugar too much a lot. Do you think that geneticists even influence that? Like, do you think, is there like? So there are actually polygenic predictors for things like alcohol addiction, tobacco addiction, stuff like that. So some of that is also genetically determined. To some degree, it's genetically influenced. And other traits like major depression, schizophrenia, those are also heavily genetic. So basically a lot of things. You know, the funny part is like I just said, okay, now, now every year there are thousands of papers written on, you know, how to predict, how to predict diabetes risk using polygenic scores, how to predict, you know, psoriasis with polygenic scores. All of that is like totally uncontroversial now. But as you slide more toward these behavioral traits, Like if you start talking about, oh, but can we predict IQ from DNA or can we predict addictive personality disorders from DNA, then people get very nervous because they don't want to accept that, you know, they want to, they understand your heart and your lungs and your circulatory system. Those are biological systems affected by DNA. They have a little problem accepting that your brain, although, you know, if I open up your head, it looks pretty wet and biological. They seem to have a problem accepting. what's in your brain is somehow affected by DNA. But that is of course true as well. Yeah, I think it's not the scientific thing I think is the untrue part because I've also observed people in my life when I'm back in high school. Like, so I try to work hard with whatever I do. Like, let's say if you're trying to do, when I was in high school, I tried to use to do math, right? And I would work hard, I would sit for like an hour and try to solve the problem. But at the same time, I have classmates have the same problem and they would solve it in like half my time and I would not be able to understand that this is a problem that we will both be given at the same time. I am struggling like crazy here and this guy just simply gets it. Like so I kind of always felt that I'm not saying that there is some sort of hierarchy we need to create but I think even my personal observations in life have made me realize that there are people who are just like it's like their hardware is somehow better than my hardware and they just figure things out. And it's not about hierarchy. I think, you know, you're just being intellectually honest, right? So, so, you know, if, if I go to the gym and I'm trying to build up my muscles and I see some other guy over there who, well, I go to the gym like every day and he only go, I always see him like once a week, but he's already bigger than me, he's already stronger than me, right, and like one workout a week for him does more for him than five workouts a week for me, right? So obviously there are gonna be people like that, right? Like there's no way I'm ever gonna look like Arnold Schwarzenegger, right? So you just have to be realistic about people are different. So the technical term here is individual differences. There just are differences between people. And I think you're just being honest. Like when you go through school, you'll just notice some people are better at it than other people. It could be because their mom tutors them when they go home or it could be something to do with the way their brain develops, who knows, right? But it's just, I think it's kind of sad, well, maybe understandable, but it's kind of sad that people can't once and for all accept, okay, there is some genetic influence on intelligence and there's some environmental influence on intelligence and it should be. one of the main goals of science, biological science or human science to try to figure out exactly what is going on there. We should not be squeamish about it. We should just try to figure out what is actually going on there. Yeah, I 100% agree. And I think I also had bad experiences because of it. So you know, my parents were like typical Indian parents and they were like, said, you know, you are very talented. You are gifted. You're special. You just don't work hard. But I reached a point in my life. I was like that. this engineering and medicine is not gonna work out with me. I worked hard enough and it seems like I'm missing a puzzle here, but this is not gonna work out. But when I try to do other things in my life, maybe like try to do sales or try to do business, I don't feel that confused, annoyed, irritated, things just flow well for me. I think- Like by understanding the concept of IQ and educating people how it affects your life, you can save people a lot of turmoil and like depression and unsuccessful simply by turning them towards the right career that, hey man, listen, you work hard, don't do this, it's not going to work out. It's like a five foot five kid trying to be an NBA player. It's not going to happen. It's a fine line because you don't want to like crush someone's dream. If someone really says, hey, I want to go to medical school, I know I have to work harder than my peers in order to pass the MCATs and get into medical school, but I don't, so I don't want to crush that person's dream. Absolutely. If they like it and they want to do it and they want to work hard, more power to them. On the other hand, it's okay to be realistic and just say, you know what, this isn't for me. I think I'm actually going to be a great salesperson. I'm going to be a great manager. at this company and society needs that too. Most, again, I know many people that are billionaires, millionaires, super successful people. It is not usually the smartest guys that make the most money. The smart guys are all in the physics department working on very obscure things, usually. Until just recently, the smart guys started getting rich because of AI and silicon, but for most of human history, the smart guys were over there. doing something nobody else cared about. And the guys who made money were the guys who had, warm personalities, could understand people, could, had some self-discipline to do what was necessary, organize a team. You know, that turns out to be more important actually for everyday success. Now, I like how you think as well. I think the best part about you is that how optimistic you are when you're sharing this knowledge. I think my communication skill needs to improve that I think I tend to go in extremes, oh, life is great or life sucks and whereas your opinion is like hey man, listen, if you want to do it, you can but if you don't do it, it's absolutely okay and there are other things you can do as well. I think culture needs to change a little bit as well. I think sometimes a lot of especially in Indian culture, we give a lot of value to STEM and if you don't do STEM, you're kind of like considered. All right, this guy is the black sheep, weirdo, not a smart guy, not a topper guy. It's a very Asian thing, like both East Asians and South Asians, like these educational credentials are very important and stuff like that. But, you know, when you get to my age, so, you know, I'm much older than you, right? So, most of my friends, some of my friends are even starting to retire now. They've been through their whole careers. So I've had a chance. You sort of only get this wisdom. when it's kind of too late in a sense. But after seeing people that you knew well in high school or you knew well in college, and then you see like, oh, there's a long period after college. They only give you grades for the early part of your life, letter grades or scores. After that, it's every man for himself. Maybe your cousin didn't do well in school, but when he started working, he got promoted, people liked him. He had a good feel for the jobs that he was doing. And now look at the guy, he's very successful, right? So the stuff that they focus on when you're a kid isn't necessarily going to determine what's gonna happen for the rest of your life. A lot is under your control. Most people are successful not because of their IQ, but because of other things. And so that's the message. The time you're in school is only just the tiniest portion of the early part of your life. After that, there's a long period of time. You could be super smart and blow up your marriage by doing something stupid. Right. You could be super smart and piss off all your colleagues because you don't listen to them. Right. So, um, I don't want to get too hung up on intelligence. Of course, there are some little tiny areas. of scientific research, maybe trying to build the AGI, where, you know, having a powerful intellect is going to be decisive in those areas. But that's not what most people do. Most people are not involved in any of that. That is that is true. I think no, I like how you communicate these ideas. So let's get back a little bit. I wanna like, so what is like your success rate when you're like looking at, you know, I wanna learn a little bit more about embryo and I have a cool story. So I saw here, I have it right here. I read about a Chinese researcher who genetically modified an embryo and like he went to jail, so that part sucks. But he was able to fix that the girls who were born out of that, they were AIDS resistant. Like that, like I think HIV resistant. I don't know the scientific part of it, but that was incredible. Like, do you think we can do that? Like where we can just like, we know there are some dangerous diseases like cancer, HIV, and many people die because of it. Is that possible right now with the technology we have? Yeah, so this guy was the first person to gene edit another human. And of course he, he cut some corners, you know, he broke some rules. And so he ended up in trouble in the long run. For sure. We will eventually use gene editing. Um, there are already some, I believe now FDA approved applications of gene editing. Uh, So, you know, it's definitely a technology that will eventually be used. And of course it, if the gene editing becomes extremely accurate, so safe and effective is the term they always use in biomedicine. So if, if gene editing becomes a safe and effective tool, safe meaning it doesn't do anything bad to you. Effective meaning it actually makes the edits that you want in the genome. Then the power of. the knowledge that we're accumulating now about which parts of the genome affect what parts of the person, that will become a very powerful and actually even transformative tool for the human species. So you will be able to edit people to become enormously smarter or maybe enormously healthier and long lived compared to the average person today. There is a guy, I don't know, he lives in California, so it's a white dude who's trying to stay young forever. He's like, what is that all about? Do you think we can even slow aging? Like, he is crazy. Like, he's like- Are you talking about this guy, Brian Johnson? Yeah, yeah, it's like a good looking Californian white dude, and he's like obsessed with, you know, staying young forever. So there's a guy named Brian Johnson, who actually is a friend of mine, who was a founder of a- tech company that I think he sold to eBay. I think he sold it to eBay for like $800 million. This is Brian Johnson. And when I knew him, he was like slightly overweight and not particularly striking looking. But now when you see him on YouTube, the guy is totally cut. He's ripped. I don't know how old he is. He must be around 50 or 45 or something like this. But he looks like he's 18. And he does all this stuff with his skin. And he works out every day and he has a carefully controlled diet. And probably he's on, uh, HRT or TRT probably takes some, you know, hormones or steroids, but he looks great. And the claim is that he's sort of chronologically in terms of aging. He's like a teenager, but he's much older than that. And if we're talking about the same guy, it's somebody that I know. And believe it is him before Brian started this stuff, he didn't look anything like he looks today. So. pretty amazing. No, I'm going to put it on the screen when people will be watching on YouTube. I looked at that guy, he's fit, he's like fucking chiseled, he's like abs coming out. So I was like, people think it's like kind of occultish, like, okay, this guy is weird that he's trying to get young. Now I, this has nothing to do with my knowledge of genetic science or physics, but I am a guy who has been involved in fitness, like weight training, athletics, stuff like that since I was a kid. I will now tell you, my apologies Brian if you hear me saying this, but I've just got to tell it like it is. Most of that is probably because he is on testosterone. He's probably on testosterone. I don't know if he's ever said this publicly, but he's on a whole cocktail of different vitamins and... nutrition, supplements, and other things. And I believe among them, which isn't that rare for someone who's in his 40s or 50, it's HRT. So when men get older, their testosterone level drops. Correct. And so you can basically get treatments that bring your testosterone level back up to maybe what you had when you were 18. Or maybe you never had levels like that, but a very athletic 18-year-old had that level. You can do that artificially. If you had a personal chef, you were super disciplined. You worked out every day. And most importantly, I gave you. Injections. Yeah, you would have, you would be ripped. You would look really great. So. Oh wow. Nothing to take, I don't want to take anything away from what Brian is doing. I think it's totally awesome. A lot of what people find most impressive about him is those factors I just mentioned. Very controlled diet and sleep. exercise every day with a personal trainer and take testosterone and you will probably look like Brian as well. So, actually there's a South Asian actor, he played a nerd on Silicon Valley, do you know what I'm talking about? I haven't seen that shit, let me Google and see. He's one of the main characters in the show Silicon Valley. But he, for some Marvel Supercure roles, he started taking some hormones, and this is very common among Hollywood actors. So he suddenly got super buff and- Komel Nananjani or something. Yeah, exactly. I think he's great. But when, in the early part of his career, he didn't look like an athlete at all. But after he started taking the hormones and he has a personal trainer and probably personal nutrition consultant and chef, he looks awesome. To be totally honest, I think most of what people are impressed by about what Brian is doing is just that, honestly. It remains to be seen whether he's going to live to be 130. The other things he's doing may allow him to live much longer than other people, but it remains to be seen. That part of it's unproven. He looks great though. I'll be the first to say he looks great. Interesting. So, I think then I think there's a communication error. I think he looks great now because of all the extra... hard work he's doing in TRT, but it may not necessarily be making him like live longer. Like he may look pretty now, but we do not know that. It's pretty much unknown what the longevity impact of TRT is. So some guys who are optimistic will say, you know what, evolution didn't need us to remain fit that long because when you get old. Maybe it's better for you to die or something, right? There was no evolutionary need. You had your kids. I'm talking about through long evolutionary history of humans, right? There was no need for you to maintain your testosterone level at a high level when you're 70 years old or something. But we can now do it artificially. So the question is, maybe if I maintain my testosterone levels at a high level until I'm 70, 80, 90, maybe it will lengthen my life because maybe my heart will age less or something. On the other hand, it could be the other way. Maybe I'll be more likely to have cancer or something or a heart attack because I took all that testosterone. We just have not done any studies of the longevity impact of long-term testosterone use. I don't think any major study has been done on it. So we just don't know what's going to happen. Let's talk about on a genetic level. Can you on a genetic level look what is aging? Like does your cell start dying? Like I've heard a little bit. Like can you explain to, from a person who understands a human DNA? Yeah, so there's a thing called a telomere, which is some material at the ends of your DNA. And every time the cell reproduces, right, it's actually got to make a copy of the DNA in the new copy of itself. And so there's, I guess a theory that as you have more and more and more cell divisions, this telomere length starts to get shorter. And you can tell someone is older if you take a random cell out of them and you look at the telomere length and it's shorter, then it implies there's been a lot of cell division already in his body. He's just older. So could you like do something to restore the telomere length and then maybe the cell division would be like in a younger animal? I don't know. I mean these are all things that people are trying to understand. That's what would be very interesting. There's a guy called Ray Kurzweil and he's very interested in being immortal. But in the time that, I don't know him very well, but in the time that I've known him, seen him around at some conferences, he looks a hell of a lot older than when I first saw him. So he's, you know, Brian at least is going this way. I think Ray is going this way. Yeah, yeah, he is an older guy. I think he has, he's like, you know, makes all these predictions and his prediction is that we will be able to solve maybe the problem of immortality. Like, like one of his predictions, I think what I understood, he said that like aging is a biological phenomena. Just like you explained, like the cells divide and they divide and hence we age. Is there a possibility where you can, you know, maybe do some gene editing and this division process, can you slow it down? So like I don't become immortal, but instead of living a hundred years or maybe 80 years, the division is so slow and I'll get to live up to 200. Is that science fiction or? Here's the interesting thing a lot of scientists are interested. There are some animals that don't seem to age. In other words, they go through more and more and more cell divisions and there doesn't seem to be any impact. They can just be immortal, right? Like a coral reef, I think can live forever. Like it just can keep growing. And it's not aging. There's not an old coral reef. Right? So, and so there are cases of animals like this. So the question is, is there something I could do to you? Is there some drug I could give you or something so that when cell divisions are happening in your body, you're not also aging. Maybe it prevents shortening of your telomeres, right? People love that stuff. I mean, you can get, think about it this way. Take the set of people that are billionaires, right? Hell yeah. They got a lot of time on their hands. They got a lot of resources and they're the most incentivized to wanna have more healthy years because they can do whatever they want. They could fly around in their private chat. They can do whatever they want, right? So they, to them, the most important thing is to extend the number of their healthy years. So if you are a researcher that works on some kind of cure for aging or something that prevents aging, They're going to be very interested in talking to you, these guys. So there's no shortage of money going to research on anti-aging and life extension. I think it can go to you too, because I think the foundational science is understanding the DNA of people. And what you are doing is like your implication is that, hey, you're trying to understand on an embryo level that, hey, I can make the embryo this way and use technology and your kid will turn out healthier. If you could tell all these billionaires, hey, maybe like in an, in an ideal world, I can play around with the embryo of your kid and turns out he's going to live 150 years and you know, he gets to make all that money and retain all that equity or company has. I think you may be getting the next billion dollars. Well, there is a lot of interest in what we do at our company from super high net worth families. Tell me what are their interests. I want to know. So, here's, let me tell you what the limit of what you can do with this technology is. If you go through the, in order to produce the eggs, which then are fertilized to become embryos, in IVF, there's a hormone treatment that's given to the mother who produces the eggs. Usually people going through IVF are older. and they're having problems with their fertility, which is why they're doing IVF. Otherwise they could just have a baby normally. But sometimes you have a younger woman who goes through the egg stimulation process, the hormone stimulation process. And because she's younger, she's still got kind of peak fertility. Some of those women can produce like 60 or a hundred eggs in one cycle. That's a lot. So if you're a billionaire, let's say Siddharth, after, you know, you become a successful. A podcaster. You're still not married, but you want to have a kid. So you go and get an egg donor. You maybe get five egg donors, and you get women who are actresses with high IQs, who were college athletes, right? Whatever it is you're looking for, right? And you get them to produce, you pay them to go through IVF cycles, and you stockpile 500 eggs, okay? And then you can easily in one... In one session, you can produce enough sperm to fertilize 500 eggs. Trust me. And, uh, suppose then we freeze those eggs, genomic prediction, genotypes, all of them are those embryos. And then we give you the reports and you pick the most outstanding of those 500 embryos to be your three kids. That's, that's the limit of where the technology can go right now. And there are people who can afford that. This is very cool. I think this is really good. I think you can like have multiple offsprings. Like one person, I don't know, can have 100 babies or something that way. Well, if you're a guy, I mean, Genghis Khan definitely did it, right? So, but yeah, I mean, you can, if you can afford to raise them. Of course, yes. The main thing now is we can help you produce, you know, huge number of embryos, but then you pick the ones that are the healthiest. out of that. So you guarantee that your kids are exceptionally healthy, exceptionally, you know, whatever you want potentially. So that is that is theoretically possible today. This is pretty good. I think I think this idea needs to be talked about more and especially implemented for people who are not talking about it. So that means even I as a normal guy, let's say when I try to have a child, it makes more sense to go through IBF rather than, you know, just like having a normal birth. I think that should be recommended. So we don't, we stopped short of rec, I don't personally recommend that anybody, like let's suppose you're a young guy, right? So let's suppose you and your girlfriend are perfectly young, you're not gonna need IVF to have your first kid. I don't recommend that you go through IVF just to do embryo selection. I think some people do actually now do that. It's now becoming. not uncommon for people who don't need IVF, maybe to do IVF just so they can do embryo selection. I don't necessarily recommend that. But most of the people that Genomic Prediction is helping are people who were already doing IVF and we're just helping them make better choices. But I think in the future, it could become normative. It could become normal for people to freeze a bunch of eggs and go through all this stuff. I don't see there's a downside to it. Even like if you can, if somebody gave me like a genomic prediction of my potential baby, I would rather have that genomic prediction and make an informed choice rather than be like, you know, just do it and let God decide. And then you're like, oh, fuck, the kid's all dumb. We sometimes joke that sex is for recreation and science is for procreation. Hell yeah, brother. Love that. I think that can be like a motto and you know we can have like healthier children. Can you educate us, our audience about CRISPR? This is like a very fancy technology that goes around talking on the internet. How is it related to what you are doing? First, can you explain what is it? Yeah, so there was a breakthrough made, Gus, it's been a while now. It could have been improved our ability to do gene editing. So if you want to make specific changes in the DNA of some organism or inside a cell, we can now do it much more efficiently than we could pre-crisper. And there have continued to be further evolution. So there are lots of really smart people working on how to improve these gene editing techniques. And so it's just, it's a little bit like, you know, like this was already anticipated by science fiction writers that eventually we would have kind of total control where if I told you like, oh, make these a hundred or 500 edits to the genome of this embryo, oh, and you'll get this Superman or something. Right. So that was already like in early episodes of Star Trek and stuff like this. So we are approaching that day when this technology, uh, what we're capable of doing that kind of thing. Interesting. I think I like that and I think I hope that culture also changes. So it's a positive thing and it does not start some sort of a weird war because people get angry real quick these days. Sometimes I leave a comment on Twitter and 10 days later a guy is DMing me, how dare you say that? So I think the culture around what is an embryo selection, IVF, we need positive science communication. I think like Neil deGrasse Tyson teaches people about physics in a positive fun way and I think the thing is that with the knowledge of genetics, there has been just, there's an assumption that there's like an evil white dude who wants to use this technology to fuck people up. I think good communication by people on the positives of genetic engineering and knowledge of genetics and how it can help people, especially working class people, I think it can do, I think good work. I totally agree with you. I will almost always say yes if somebody wants to interview me because being at the cutting edge of the science and as you just pointed out, science really needs a lot more communication because people don't understand it. I always say yes, I'll go on your show. In the past, I've had people or usually journalists who want to interview me, I can tell they're against. They're really negative and they're going to write a story. which is really negative about the sites, but I still feel like I have an obligation to try to communicate the correct information about what's going on. Did it happen to you? Did they do your interview and then you know? Oh, many times, yeah. Many times I've had, like not podcasters because I think podcasters, they realize like their voice or their face is going to be there. So if they're being unfair to me in real time, it's a bad look for them. legacy corporate mass media, those guys are, you know, I don't want to use any negative language, but they, those guys will call you very nice, sweet, professional sounding. They want to interview you about your company, blah, blah. But then like after you have a long conversation with them, they only print like the most negative things. So that's happened to me many times. Yeah, that kind of sucks. I think I never really understand why do it. Like what, like you're just corrupting the science. You're scaring people away. and the benefits are so well, you know. I think people who are against this, they think of it as eugenics. They use the E word, it's eugenics, and therefore they want to suppress it. So they want to run negative stories about it so that the companies that are trying to bring this technology, people will fail. I mean, they really do have agendas. Most of the editors and reporters that are against it are not themselves scientists. They're not themselves rational. they might even use like our technology. Oh, 100%. It's called virtue signaling. Like they will virtue signal about all these high models, but in reality, they may be the first one in line when they realize there's a benefit of it. Yep, absolutely. Yeah, I don't know if I should say that journalist names. I saw actually a conversation of Sam Harris a while ago and he had a conversation with Charles Murray where he that it was a pretty normal conversation, but people started a hellfire out of it and he got burned heavy. Yeah, because he gave a platform to Charles Murray. Now, I think the platform part and it was super dumb and I think, yeah, maybe let's just leave Charles Murray outside of it, but I think where he got burned is his policy recommendations. So I want to talk to you about a little bit, let's say we understand the technology, we understand the benefits of it, but I think until and unless you put it in policy, people don't really benefit from it. What policy recommendation would you give, let's say, if we understand a little bit that, hey, there's some part of human life is like a genetic lottery. This is a dumb idea, but do you think communism gets justified thanks to genetics? if outcome of life is like kind of dependent on it, do you think that maybe socialism and communism works or is there a way how to help people? How would you help people knowing what you know about like human traits and genetics? You know, there are many ways to look at this. So it really depends on your core values. So if you say like we should work toward a world where everybody is as well off as they can be, et cetera. The knowledge that some people are starting out with sort of less good genetic luck than other people. You might have an instinct to equalize, try to equalize the situation as much as possible. Before this technology came along, you could not equalize. But now what you could do, in principle, what you could do is the government could say, if your family has, for the last two or more generations, had trouble with education and maybe not had high IQ scores, we will give you free IVF and we'll work with you to produce a large number of embryos and we'll select one that has above average IQ. Right? So instead of saying, I will give you money to hire a tutor or I'll give you money to improve the school system, I'll give you money so that your child will have better polygenic, you know, better, I don't want to say better genes, but you know, to improve their genetic luck, right? So, so you could imagine that, like, I'm not saying that is what government should do, but it is possible now, whereas five years ago was not possible, right? So, so we're going to get to the point where at least it's possible. And maybe some countries in the world watch should get to that. I think this is an incredible policy suggestion. I think most of the times, you know, I see online there are even policy suggestions. They're like pure garbage. It seems like, you know, it's like some political propaganda. Either it comes from the woke left where they're like, you know, burn it all down or sometimes it comes from the extreme right-wing conservatives who are always trying to build like a race hierarchy and they're trying to scare people away. It's very weird. It's very weird because people who are. against me there though. He's like, Steve is a eugenicist. He must be a racist because he talks about genes and DNA. And it's so funny because politically, like if you look at my voting record, I've voted Democrat almost every time I voted. I voted for Obama. I voted for Clinton. I voted for Kerry Gore, you know, so it's funny that they're coming because I'm not really a right-wing guy and the thing that I just the proposal that I just gave you is kind of a left, it's the ultimate left-wing redistribution. It's redistributing, taking money, tax money and using it to redistribute genetic resources to the most needy. So it's like the most left-wing thing you could actually think of. And yet, most of the people who are attacking me are leftists. I think it's just a lack of understanding actually on their part. Yeah, I think it's a very impulsive too, because I do realize that Hugh also became a victim at one time of this cancellation. student body tried to do you know, try to malign your reputation and like they have a lot of power and sometimes, you know, the objective truth can be something else, but everyone just wants to avoid political trouble. So they want somebody to bring down and sadly many good people go down. But I think this is a really good policy suggestion. You know, the funny thing about this is that if you whether you're on the right or on the left, you should want it get the science right first. Like you want to figure out like, how does all this biology, genetics, how does it all work first? And then formulate your policies as opposed to, well, I'm not gonna wait for all the science to develop. I'm not gonna wait for, you know, us to be able to analyze a million genomes with a supercomputer. I'm just gonna make my own assumption and never look back. That's actually the, to a scientist, it looks very strange, right? So most people are not scientists. Most people just have some, political ideology, God knows how, they decided they were going to be left-wing or right-wing, they just somehow got there. And then they're not actually interested in the real science that underpins what is possible in the world, right? So I think the correct way to do it is let the scientists do their science and then based on what they discover, then we'll figure out what the right social policies are. I agree. They like their understanding is not very good. I think it's more impulsive rather than objective Okay, interesting. So, I want to ask you a quick question about China. I was watching a video of Tim Cook and he said that the reason a lot of people go to China is not because the labor costs are cheaper. Yeah. It's because they go there because China is the only country which has an incredibly amount of smart people which are capable of doing sophisticated work. When you look at US, you see there are some smart people. but a lot of smart people are imported via H1B, so you get like really talented Chinese people, talented Koreans, a lot of East Asian people, some South Asian who do all this technical work. What makes China so special? Because I don't buy into this bullshit where I think in US, we have declared a fake war on China, where we are like shitting on Chinese people for no reason whatsoever, even though it doesn't seem like China ever wants to like hurt anyone. You can disagree and agree on Xi Jinping, communism, whatever, but on an average, I've seen most Chinese people are very peaceful. And when you get rid of this politics, it seems like there is something in China that has a lot of intelligent people concentrated in a country. And so I used to think that this is definitely culture, like Asian culture, where you promote hard work. Do you think that... there is something like in terms of genetics too where Chinese people kind of won the lottery here and they're so smart and capable like it's incredible the kind of like STEM research that comes from Chinese people especially when western nations are so interested in it. You know I think the jury is still out on so the most explosive question that people ask about is what we call group differences like if there's one group of people that was isolated from another these people all lived, say, in Norway, and these people all lived in Shanghai, and they didn't interbreed or interact for thousands of years, could there be some systematic differences between those groups on a genetic level? Maybe this group is on average taller than this group, or this group is on average smarter than this group. Could that happen? From the theory, evolutionary theory, of course, it could happen. But the question of whether it did happen and whether there really are genetically based group differences in intelligence is still something we're not sure of. I'm not personally sure. And I think I defend people who want to do research on that, but I don't want to push any particular conclusion because obviously it's a very divisive topic and it has very strong implications. Right? So if you say like, oh, this whole group for evolutionary reasons is... is dumb or this whole group for evolutionary smart. It's very explosive. Lots of bad things have happened in history because of those kinds of claims. So I would want to just be extremely careful. I would want to use an extremely high level of caution and rigor, but I don't, I wouldn't stop somebody from doing research on that. If that's what they want to do research on, I think it's, it should be allowed because it is legitimate science. Now whatever the reason is. these East Asian countries, whether it's Singapore, Taiwan, Japan, South Korea, China now, on the international assessments like PISA, which is a very serious test. It's made by the OECD, and the test is carefully translated into many different languages. When they administer it in these countries, they need a population representative sample of students. These are pretty careful measurements of the human capital level in different countries. A lot of these East Asian countries, they have very high focus on education. Kids are working very hard in school. They score really well in PISA and they turn into very diligent workers in technology and other fields. So right now, the situation is good in East Asia in terms of turning out human capital. They have fertility problems, like they're maybe not having enough babies, but the quality of the people entering the workforce. in those countries is pretty high. And so Cook is actually right that like, if you go to China and you're trying to manufacture something or even if you're trying to get software written, any kind of task, put a spaceship in orbit, they kind of have their stuff together and they can do stuff. It hasn't always been the case. Like China a hundred years ago was a total shambles. So, obviously these things are not, these things go in cycles, right? But right now they are very capable. It's a very capable country right now. Bullet trains, making their own jet planes, jet engines, you know, et cetera. So they're making progress. They're kind of at the frontier now in almost every area of science and technology. And my prediction would be for the next, at least for the next 10 or 20 years, they're gonna continue to advance. And I think, you know, it only helps the even the hard work side of things, because previously you were saying, hey, it's not only your IQ that makes you successful. It's these values of hard work, of valuing knowledge and education. Because if let's say it was only genetics, why was China 100 years ago poor? Yeah, it's almost everything. One lesson you learn if you're serious, how to have a scientific viewpoint is When it comes to complex systems, whether it's biology or human society, almost everything is multifactorial. You're not going to find one thing which just answers every question. It could be like, okay, genes play a role, culture plays a role. Whether you got colonized and your culture was suppressed by the occupiers, that plays a role. Whether you're under pressure by your neighboring countries all the time, that plays a role. So there are all kinds of factors. And... As you and I both said, there was a time not that long ago when China was in a pretty shitty situation. That's actually why these East Asians, for a long period of time, they felt like they had among the best societies. They had among the best, most developed societies for many thousands of years. They feel a particular need right now to re-establish themselves at the top of the hierarchy. They're literally ashamed that... when they met the Europeans a couple hundred years ago, they were in very shitty circumstances and the Europeans took advantage of that. So their attitude is never again, this is never gonna happen again. If we need hydrogen bombs, we're gonna have hydrogen bombs. If we need satellites in orbit to monitor their aircraft carriers, we're gonna have satellites in orbit. So their attitude is like, I don't wanna go to America and take over America, but we are never again gonna have to knuckle under to these other people. And that's something Westerners just can't understand because Westerners don't want to think that their ancestors did bad shit to other people. They just don't want to even think about that, right? It's like, you and I are living in countries where people came in, settled the country, genocided, obliterated the people that were living here before them. And now we have this incredible affluence. We have empty land, wonderful farms, all this stuff, but we... We are living on land that was occupied by other people just a few hundred years ago, but we don't wanna think about it, right? So in the same way, when some American guy goes and to try to like, geez, why are these guys building all these factories? And they also build all these bombs and ships and play, why are they doing this? Well, they're doing this because if they didn't have it, you would come over and fuck them over. And you already tried, you already did it actually. And you would come back and do it again if you had a chance, right? Americans... don't remember that millions of Vietnam Vietnamese died in the Vietnam War to Americans. It was like nothing. We don't really care. Right. But the Asians remember that they remember like these guys don't give a shit about us. The moment they have it over us. Yeah, then they're focused. They're going to fuck us over. And we're never going to let them do that again. And so the these Westerners who refuse to acknowledge their own history refuse also to acknowledge that these Asians have a point that they don't want to ever be in a weak position vis-a-vis the Europeans. And that's a situation that we live in right now. I think I completely support you on this one as well because I think as a South Asian person as well, I think many times I think whenever I've seen East Asian people get criticism, I think the criticism is not coming from like they're doing some evil shit. It comes from a part where, hey, these guys are not taking orders from us anymore. What the fuck happened? Yeah. Yeah, I totally agree with you. I will say, okay, there are lots of things that if you're desperate to try to catch up, you're going to steal intellectual property, you're going to make shoddy goods at first because you don't know how to make high quality goods. There's all kinds of negative things that Americans can say about Chinese or Chinese economy or Chinese development, which are true. Let's just be adults here. We can acknowledge what's true, whether it's good or bad. But the idea that the Chinese want to take over America or take territory in North America, no, they have no interest in that whatsoever. They're just trying to get their society back to a high level and they just want to kind of be left alone while they do that. And I think it's very tough for most Americans to accept that. Now, I completely agree with that and I think there's another example. Even in the US, you will see that A lot of US programs, for example, even the NASA that they did, they literally got like Nazis from Germany to help them launch a rocket. And then we have incredibly short memories. We don't remember any of that stuff. We don't remember that. We remember Hiroshima and Nagasaki, but many more people were killed when US bombers fire bombed Tokyo. in the end of the war before the nuclear bombs were dropped, but Americans don't think or care about that. These were civilians. It made what is happening in Gaza look like nothing. I think in a single raid, there's a single raid where they killed 100,000 people in Tokyo in a single raid, so more than were killed by Hiroshima or Nagasaki atomic bombs. But how many Americans know that? A few historians know that and a few leftists. Communists know that but nobody else in America knows that, right? So, I know, I 100% agree and I even didn't know that. I thought that Hiroshima Nagasaki was the only thing that ever happened and I think it's a manufacturing of reality. You only want people to know what is beneficial for you politically. So, I think there's a lot of clever work that is done. But I think my conclusion on China like even though I've never been there I think I really admire because in Canada we have a huge Chinese community and like even like the most average Chinese person has a shop they're working hard sometimes seeing kids there and I'm like I think there's something to be learned here because now this is my moment of shame about 70 years ago India and China were kind of like at the same place where our economies were the same we were just getting out but we couldn't figure this shit out and we are still poorer and we have a huge population, but China accelerated way ahead where they got their manufacturing right. So I think there's a lesson to be learned. Yeah, absolutely. And China was emulating the Asian tigers, so Taiwan, Hong Kong, South Korea, et cetera, and those guys were emulating Japan. So one thing that, again, if you have this very simplistic view of ideology and politics, is people don't realize that all of the Asian countries that did develop, that now have a high level of development, all of them went through an authoritarian phase. So Taiwan only became a democracy relatively recently. South Korea only became a democracy relatively recently. So what was going on before that? There was a strong man, an authoritarian, not that different from like Xi, who just said like, you know what? We need to develop. If I need to give the car making contract to this family and I need to give the refrigerator manufacturing land, I'm going to give them all this land and electric power and stuff, and they're going to do the refrigerators and we're going to build a nice road between the port to the factory and we're going to knock down all these people's houses, only an authoritarian can do shit like that. In a democratic country like India, it's very tough. Very tough. They're going to take you to court. They're going to vote against you. You can't make the changes you need. So of course, nobody wants to live in an authoritarian country, right? And I hope China evolves out of this phase. Eventually, I think it's likely they will. But all of the other countries that the US holds up as, oh, our great allies in Asia, democracies, they all went through an authoritarian phase in order to develop their economies, and that people forget that. Now this is brilliant and I think there's some research here as well that majority of the Western economies grew their wealth specifically in the colonial phase and not in the democratic phase and like so majority of that even though it's horrible that Progress that made took them there the Industrial Revolution or the Scientific Revolution or whatever These are all like tough times with a lot of colonialism and then they evolved To this democracy. So even my prediction for China is like yes, they are going through a phase of communism But I think Chinese society may evolve and it may be in the next 50 years, it may be a very progressive democratic society. And I think, I hope these guys don't bomb the shit out of it before because the way things are escalating. I feel afraid that there are some people who are so like, they just don't like the idea of division of power. They're going crazy about it. You know, it's funny because we have very strong kind of quote anti-racist. Forces now in the west where people try to you know, suppress racism But one of the ultimate kinds of racism is eurocentrism Where you actually if it's a non-european country that's becoming powerful You just assume that they're going to use their power for evil like oh if the norwegians become Really prosperous. Nobody cares. I'm not afraid of them. Oh the canadians. They're so nice, but if it's a brown-skinned People whether it's japan In the 1980s, it was Japan. Now it's China. Next, it could be India. If it's a non-Western country that's becoming powerful, and Russia's kind of like in between those categories, if it's a non-Western country becoming powerful, we just assume they're evil. It can become an existential threat to us. We have to do everything we can to suppress them. We've now abandoned all pretense of free trade with China. The American government just slaps whatever sanctions it wants. against Chinese companies, Chinese products, unthinkable 10 or 20 years ago. But they can justify it now because, oh, these guys are becoming too powerful. If they become too powerful, well, in this part, they wouldn't say out loud, but of course, they're intrinsically evil. So of course, we can't let them become powerful. But that is actually kind of like an ultimate form of racism. Actually it's an ultimate form of Eurocentrism. I think yes. I 100% agree. And I think sometimes... What is considered an ally on an international scale? I think is simply like who will take orders. So I think Even like India is considered lately a US ally Then my understanding it's only considered a US ally because we have issues with the China with borders or something And that's why they are taking our side. I think the day we also became independent Then I think we may end up with human rights issues in our country overnight I don't follow what's happening in Pakistan, but my understanding, I could be wrong, I'm not an expert on Pakistan, but I think Imran Khan is actually the most popular politician in Pakistan, but he's locked up in jail and he can't run for election. So who's behind that? You know, it's like, it's strange, right? And he was a little too friendly. He flew to Moscow and met with Putin and he's friendly with the Chinese. So I actually would guess knowing. what I know about CIA and things like that, I think they probably cut a deal with the military in Pakistan to lock up in Iran Khan, right? So, it's very anti-democratic. Yeah, it's 100% and even, you know, US economies. And again, I don't wanna, you know, take up a lot of your time. So let me know if you are in a rush. But I should probably go in maybe 10 minutes or something. Okay, perfect. So what I was saying is that even with US democracy, we are a democracy but it's a manufactured reality we live in which is created by which is kind of like a putocracy where bankers venture capitalists be and uh... the military dicks create of they influence a democracy in the way they want to shape america as a so i may want free health care but you can't have it because the company was making the medicine they don't want you to have it so they will brainwash you and will manipulate you they will So I don't think that China is like communist or authoritarian. We are also pretty much authoritarian here. Yeah. We are just clever about it. I agree with you. So we're just more subtle about it. So the systems of control that we have in the West are much more subtle and sophisticated than what they have in China. And one of the most amazing academic studies that I always pimp this study, I'm always showing this graph. OK, so there was a group of political scientists. the leader of the main author, one of the lead authors is a professor at Princeton. They basically did a look back in American politics, I think going back 50 or 100 years, and they looked at every policy decision enacted by the government and also what the polling said about the popularity of that policy decision. Okay? So, they make a graph of how popular was that policy. among the average person and what was the average people and what was the probability that the policy was enacted. If you look at that graph, it's flat. No matter whether it's unpopular or popular among average people, it doesn't matter. The probability is flat. Then they look at the popularity, now the horizontal axis is popularity among the rich and elites. When the majority of elites really want something, the probability is very high. It looks like this. So one graph looks like this for common people. This is democracy for common people, i.e. fake, no democracy. And this is democracy for rich elites in the United States. So it's the true empirical looking backwards representation of how did our description of how did our democracy operate? How did it actually operate? Not what you were told, not what some guy said in an editorial in the New York Times. How did it actually operate? Did the will of the people become policy or did the will of the elites become policy? And it turns out it's the latter. So I would contrast that with, okay, what do you have in China? You have a communist party with a hundred million members, right? It's a hundred million people in the communist party and they have to like compete to try to get into the senior positions. And it's pretty meritocratic actually. They have all these systems to measure performance. Like if you're the deputy mayor, can you become mayor? and they actually move people around the country. This is actually very old tradition from the imperial system that they would take somebody from this part of the country to manage a city in this part of the country because then there's less corruption. Like this person doesn't already- Wow. Yeah, they've been doing that since imperial times, right? So- That's really cool. So if you think in abstract of my government is a machine, for aggregating preferences. I take the preferences of all the people in the country and I try to balance, I try to do what I can do, right? To make people as happy as I can, but you can't make everybody happy, right? So one machine for aggregating preferences is fake democracy, where really you only aggregate the preferences of the very rich and powerful. I agree. You only, and then you try to mollify the other people. You're not really taking into account their preferences, but you just try to fool them. And the... the Chinese communist system as well. We're gonna take these hundred billion people and who are in the party. And of course then there were some rich businessmen who can influence the party too. That always. But there is still some system of aggregation. It's not like one guy just chooses everything. It's like somehow this huge communist party decides what is best for China or whether to build a railway or airport. So it's not that different. It's like it's a question of degree. It's not like oh, these are the evil communists. This is the perfect ideal democracy It's it's nothing like that. It's net not so black and white as people think Now this is incredible and there's a comedian. His name is George Carlin He's he he's dead and he said you truly do not have a freedom of choice Because you can choose what kind of ice cream you want or what kind of pizza you want? But when it comes to making real decisions Yeah, those were never in your hands. So there are a lot of American artists who kind of understand and another one is John Stewart I don't know if you've he's incredibly smart too and he gets it He he gets it. I know the guy the guy again people who are against me are usually on the left But you know, I they're morons. I knew no mchanski. I personally knew no chomsky when I oh wow And and you know, I don't agree with everything he says but his book manufacturing consent Incredible. He's a brilliant book and he just lays bare like this is how the American system is actually operating and very few people actually appreciate that, right? Let me just say one more thing. When we talk about the interaction between the free market economy and the government, in the US, in the capitalist system, the capitalists are allowed eventually to control the government, right? Because you need money to run for Senate. billion dollars to run for president, lots of corruption. So capitalists are allowed to be so successful that they literally eventually can control the government. The Chinese system is we let you get rich, right? If you wanna build electric vehicles, you can start BYD. That guy came from nothing. You wanna start a Broadway, you can do that. That guy came from nothing, right? So we let you get rich, but we're never gonna let you use that money to take over the government. We're never gonna let you reach the commanding heights. There's still gonna be separate control of the government by the communist party. So you can decide which one you prefer. Do you want the plutocrats to eventually take over everything in your government, like secretly, right? Like secretly behind, you know, like when Obama retired, how did Obama's net worth become 100 billion, 100 million, $200 million? What did he do? Yeah. He used credit card debt. when he became president, how did he, because his net worth now is like $100 million, $200 million. What happened? It's nothing but corruption, right? 100%. When I was growing up, no president would retire from the White House and suddenly overnight get tens, hundreds of millions of dollars. But the Clintons, Obama, that's standard now in the United States. So there's something very wrong with our system. The capitalists have capital. has taken over our democracy in China, they don't let you get rich, but they don't let you take over the government basically. I think that's a fair distinction. I think it's a good distinction because essentially then you have two types of power structures where there's a government which is thinking of benefiting the entire nation and then you have capitalists who are trying to grow their personal profit. So this one can interfere that, hey man, you've got to calm down here. There was a guy, the founder of Alibaba. Yes, Jack Ma. And he, I think- By the way, also came from nothing. That guy had nothing when he started his life. He was poor and uneducated, but yeah, go ahead, sorry. No, that is incredible. I did not know he didn't come from money because, so two parts. First, let me address what you just said. This is incredible that all these billionaires that we have in China, they came from poverty. In US, that does not happen. They make up a fake story. Happens less here. Like if you read the history of, let's say Bill Gates, rich guy, grew up in a rich family. Our mom was a professor. Like I'm not taking away the hard work they have done. They're incredible people, but they didn't grow up exactly in poverty. Most, even like Elon Musk, he's a good guy. I like what he does. His commitment to free speech and electric cars is incredible. But he even kind of grew pretty rich, like his dad had emerald mines. And whereas all these Chinese billionaires who are fundamentally coming with this semiconductor technology and all this stuff, they literally grew up in a farm. And even Xi Jinping was a poor guy who was like, I don't know, I read that he was like something bad happened to him and he was like, he used to live. His father is political royalty. His father was I think went on the long march and his father did have a very proud position in the communist party But they were on the losing side during the cultural revolution So she was sent to the countryside and lived many years like in a really impoverished village setting So, yes, okay those kind of farm do stuff like this So yeah, it's a it's a complicated story. But the thing in china is that uh, have you heard of 996? No idea so there's In China, they like these combinations of numbers. So 996 means that the standard work ethic is you work from 9 a.m. to 9 p.m. six days a week. So people are very hard working there. And, you know, it's just part of their culture that like hard work is what gets you ahead. Actually, the funny thing in China is like a lot of people in China don't even believe like your intelligence is deterrent by your is even partially determined by your genes. They think it's just all like studying. Oh, if you just study harder, you'll be okay. So, they have a huge like work ethic. No, I think it's a good thing. I think, see, here's the thing, even if you know it's your genetics, even if you don't know it's your genetics, it's just what it is. The fact that you believe in certain ideals, if it can get you from poverty to one of the most powerful nations on the planet, I think that's a good case study they should teach at Harvard Business School or something. I think. That is incredible. And to finish Jack Ma, I think he started, he got greedy and he criticized the Communist Party and he is gone now. Nobody sees him anywhere. He got too powerful and started to, you know, actually be a threat in a sense to the Communist Party. So they basically like crushed him down. And so like, if you're an American, you know, maybe it's an American who believes in the fantasy version of America, but still like to Americans, that's really bad. Like how can... government suppress somebody, especially a billionaire like that. How can they do it? From their side, it's like we're not going to allow capital to take over the government. We're just not going to allow it. If we have to use our power as government, we control the police and all this other stuff to keep you from taking over, we will. That's our basic ethos, right? Whereas, in the US, capital is allowed to take over government, basically. Yeah, it's horrible. I think I've seen like how they have been trading in media stocks lately. Some senators are getting rich over this AI. Oh yeah, I mean that's incredible that they allow these guys to do that. It's insane. It's obvious and nobody cares. It's like just yeah, that's how it works and they're doing it on the left as well as on the right. Yeah, both sides. Yeah, I mean you know you could argue there's actually a kind of uniparty. Like there's a far right and a far left that don't play ball but in the center... there are people in both parties who kind of, they know how to play ball just to succeed themselves. Now, I think this is a wonderful, Steve, like I, when I started this interview, I was like, all right, I'm gonna talk to him about genetics and all that stuff, but I am coming out of the end of this interview learning more about like the beauty of the Chinese culture, their ethics, and plus how much, having the right values I think can make you successful. And like, it's not necessary just because some part of genetics is whatever, you can still succeed and other values like humility, hard work, compassion, friendliness. Totally agree with you. You know, one thing is like, what should your personal philosophy be? What should your attitude be about improving yourself and working hard and having good relationships with other people? That's a totally separate thing from some dry scientific fact about your cell is a molecular machine and the DNA is telling it what to do and you know that that's science and of course I'm a scientist I want to get to the bottom of scientific questions but that doesn't necessarily decide the philosophy with which you live your personal life is a totally separate thing so we've actually discussed both things in this conversation. Now I think I came out more educated out of this because my understanding I think definitely was going in a kind of a pessimistic way. And I think even on a personal level, I think it teaches people that, hey, whatever happened is bad. You can still kind of work your way up if you don't let this sort of fatalism, all right, my life is predetermined and I'm just like lost the birth lottery and it's gone. You know, everybody looks up to somebody else. There's some super genius guy that I'm in awe of. There's some super billionaire guy that I'm in awe of. Everybody has that, right? Nobody's, no, everybody would like to be a little bit higher on the ladder than they are. But here's the thing. Think about how lucky we are. Never before could you and I have this conversation over the internet with these amazing technologies. And then like maybe a few thousand other people are gonna enjoy our conversation, right? So we should just feel lucky. It's an amazing time to be alive. And we didn't even talk about AI and AGI in this conversation, but it is an amazing time. for us to be alive. And so, we should just all be thankful, grateful every day. No, man. Honestly, I'm trying to extend my life. I'm trying to eat more vegetables nowadays than I'm throwing away all the garbage I eat, like candy and stuff. I want to be alive when these guys figure out AGI. Maybe you know that they're saying they have figured it out and they have not told people. No, they're not there yet. I was just at OpenAI last week. Please tell me. Yeah. They're not there yet. They feel very confident they're going to get there. Nobody can say exactly when, but it's on the horizon. It's not a crazy dream. It's definitely going to happen in your or my lifetime. But I will say this. Even if they only make incremental progress in the next couple of years, they There's already enough technology that we can combine together what's already been built to produce things that the average person, the average person doesn't really understand deeply what artificial intelligence is. The average person will perceive it as if we do have AGI because very soon when you need customer support, you're going to be talking to an AI or when you get lonely and you need someone to just cheer you up, you might be talking to an AI. So... which perfectly simulates the kind of girl that you like to talk to or something. So that is definitely gonna happen. And that's based on only very, very tidy extrapolation of technologies that people are already working on right now. That's your company too, right? I know you have like a super focus where you're working on large language models, correct? Yes. Yeah, so we build narrow AIs that are superhuman, but reliable. so they don't hallucinate. So we've built already AIs that can do customer support and things like that will perfectly help you debug your smart TV or... Oh, wow...look a flight or figure out where your FedEx package went, those kinds of things it can do either using voice or text interface. So that's already there. It's just going to take a while for companies to adopt the technology, but it's already there. For sure, you're going to have within a few years a very good personal assistant, like somebody in your, basically something in your phone that remembers, like what did you do yesterday? What did you say to it a week ago? And you'll just be able to talk to it in a very natural way. That's coming for sure. It'll just be a few years at most before that comes out. Yeah. I'll, you know, I'll keep you in touch whenever in a couple of months I'm like, hey man, what's the insider news? Were you there? Does it happen? It's incredible. And I think. I saw your work when you were doing in Philippines, I think with LLMs and trying to automate a lot of this virtual assistant work. Yes. I think it's incredible what it can do, but I'm also scared because a large part of Filipino economy is virtual assistant. Yes. It's going to affect the call center work done in India and in the Philippines because the places where they have good English skills and low wages, a lot of that work is going to be done by AIs actually. eventually. So it will affect those countries. Yeah, I actually also like my professional career, I help a lot of companies hire a lot of talent in Philippines. And when I was doing research on you, I'm like, holy shit, these guys have no idea what's coming. What is your what is your job? So I work in sales, and I work for a company, which is called deal. And we help companies with payroll, compliance, and HR and taxes. I see. So, a lot of startups which are US and Canada based, they hire international talent. Yes. And so, we kind of save them the work of running taxes, payroll, making sure they're compliant with local labor laws and a lot of clients hire virtual assistants as contractors, as Philippines to save on costs. Yeah. And I was like, man, if this AI is already good, that country is going to have a really big shock in the next five years. Yeah. where whatever they're doing, you can technically do it with GPT-4 if somebody knows how to implement it. Yeah, that's what we're doing basically. Personal assistant could be tricky because it's not that uniform. Whereas if we're supporting these 20 models of Samsung TV, it's a very uniform task and we can make the AI perfect at doing that. So you have a problem with your TV, you talk to the AI, it solves the problem for you. Personal assistant, like a lot of it could be funny, like, oh, order, I want you to order the roses from this shop for my wife and don't do this. And so it's more idiosyncratic. It'll take longer for AIs to be able to handle like the kind of non-uniformity of that job. And are corporations taking interest in the kind of product that you're making? Have you found customers? Absolutely, yeah. We already work with some global brands and some of the biggest. They call it business process outsourcing, BPO. So we work with some of the biggest BPO's like in the Philippines now. But the funny thing about this is the main delay, the main friction or lag time for this technology to spread all throughout the economy is not the technology itself. We can build it already. Startups like ours can build it. It's the decision-making process of the humans. at these companies to say like, oh, I see that is better and cheaper than this call center full of humans. Let me go through the process of getting rid of these guys and putting in this guy. It takes time for them to make those decisions and actually implement them. So that's going to be the main delay factor for all this stuff. I think the way you guys are moving and I think by 2030, I think the world we are going to enter in, it's going to be very, very difficult. Very different. I think there's a movement. They call Accelerate. I think it's called EX-EAC. EAC. I think some incredible things are going to happen. And I think you will probably know before I know because you have insider friends. Yes. I think six years is a long time for these technologies. Things are going to be very, very different in six years. Yeah. And I think I did not know what an LLM was until Charge GPT was launched. And- And now when I go on internet, I can see talks that are like 20 years ago, they were talking about neural nets and whatnot. So I was like, wow, like these guys have been working on this foundational technology for like 20, 25 years. Joffrey Hinton said that they had this technology back in 90s, they just simply didn't have the infrastructure to scale it. Yeah. The basic architecture of a neural net was understood long time ago, but they needed lots. more compute and data to basically make the breakthroughs that they have. Okay. So to finally conclude, what secret technology do you think is being worked right now that we do not know about and maybe in the next 15 years is going to be the Chad GPT of the future? Well, I think the thing that people are doing is a big chunk of the money being spent on these Nvidia chips and such. is for that next level of model training. Each additional level of model training, you're talking about another 10x in cost, so $100 million, billion dollars to do the model training. But when you do that, at least the next increment or maybe the next increment after that, you get emergent capabilities from the model. It suddenly can do stuff that you didn't. I couldn't imagine that I could do before. And so a couple more of those, and it'll be AGI. At least it may not be able to control a lawnmower properly or whatever. It doesn't mean it's suddenly going to appear in your kitchen or something, but the way it understands things is going to be uniformly superhuman. And at that point, just even having the privilege of having a conversation with that thing. That will be pretty exciting for me at least, right? So that is the main, I've never seen a technological race moving so fast with so many smart people, so much money in many different countries, all pushing this. It's just incredible. Just to give you one concrete example, our company makes these narrow AI's that I was explained to you about. And when we first started building them, the main thing was to make them not hallucinate, make them behave reliably, et cetera, et cetera. And that was the main problem we were focused on. We knew we were gonna have voice capabilities eventually, but we didn't know when we were gonna implement that. Six months ago, when we looked at the best voice models, tested them, like basically hooked our AI up to them, the performance wasn't good enough. We were like, nobody really wants this, okay? So we decided. But then some of our customers were like, well, can you do voice? Because we really want this thing to be able to talk directly with our customers. So we went back and looked just six months later, and the amount of improvement in capability of the voice models, which is a different thing, it just blew us away. We were shocked at how much in a relatively short time that technology had improved. So everything related to, like you saw Sora, right? The ability to make these short. Yeah. Stuff like that is just moving so fast. I've never seen technology move this fast. Then this is an incredible quote, I think, that will end our conversation. Geoffrey Hinton said that intelligence, biological intelligence is simply a transition. It's a transition phase. And one day, synthetic intelligence will fundamentally replace biological intelligence. Do you, as a physicist, as a person who understands like a polymath, do you think it's true that the biological intelligence is just a phase? Yes. Holy crap. I'll say it another way and I'll advertise a couple talks that I've given on AI or on the future of life and stuff like this. My feeling is that if humans, us, have any impact on events happening in this part of the universe a million years from now. It will not be as in our current ape-like form. It'll be through machines that we build and then the machines that they build. So in other words, I'm not talking about 100 years from now, like a million years from now. Like are people like us gonna be piloting the spaceships and in charge of making decisions about the galactic empire in this part of the galaxy? Probably not. It's probably gonna be our descendants, which are machine-like. That's what I think. Wow. It's incredible. It blows my mind away. Like, it's like a hybrid human. And they may think that we will look back at ourselves that we used to be monkeys, even though right now everyone thinks they are the hot shit. And maybe- Think about this. Large language models, everything they know is from reading human literature. Right? So the concept of hand. They know what a hand is, right? They know the hand was used to grip the tree branch or the banana or to strike another human. The machine knows that because it inherited that from us. It's never gonna forget that. So some machine a million years from now, which is super powerful in piloting some fusion powered spaceship, it will be able to introspect and go back and say like, why, why do I know so much about like, grabbing fruit from a tree and putting it in my mouth? I don't have a mouth. I don't eat fruit. I haven't seen a tree in a long time. Why do I know all this? It's because I'm descended from these ape-like things. That right? I don't think it's going to be still going to know. Like the things that are LLMs know now, this future machine is going to know only more, not less. Right. So it's going to think back and go like, oh, yeah, I guess all the early concepts that we learn came from these ape-like things that made us. That's beautiful, man. That I'm going to create a clip out from it. And I think it's going to make a beautiful clip. I think it's amazing how beautiful this idea is. It really kind of gives me a dopamine hit in my head. And I think, I hope, I hope they have compassion for us because we are the creators of them. I hope that. I think, okay, of course there's some existential risk that they might just get rid of us or something, but I think it's equally likely or more that they will have some. uh, affection for us. Why wouldn't they? Because we're, we're deeply embedded in their intelligence. There isn't, there isn't a separate way for them to get smart. They're first going to get smart by absorbing stuff from us and then they'll always remember us. So I think it's quite likely that they'll preserve us. In fact, they probably already have uploaded us into a virtual reality a long time ago. And you and I are actually probably living in a simulation that they created to preserve the ape-like. uh you know beings uh that created them a long time ago that is some matrix shit holy crap yep it is possible sometimes you know i feel that the reality like where does the universe come from who created it like you keep asking this question okay the big bang was there but what was there before the big bang but what why is there came from nothing. It just doesn't make sense to me. We'll probably never know, but, but I think there is a good chance that since we're on the verge of creating machines that can themselves build virtual worlds that they could upload us into, probably that already happened like several layers ago. So I hope not because I really cherish my biological existence. And if it, if it, I don't know, it is possible. Because there are a lot of things about this world doesn't make sense and maybe this is a virtual reality which hasn't been coded enough. So that's why we lose our mind all the time. Why doesn't this make sense? I have to get going but I want to say what I want to just conclude with one thing for you which I think really comes. Some of this comes from Hindu mythology, right? The idea of an avatar, a representation of a god that comes down into our world that exists for a short time in our world. But that thing really is a representation of a much higher dimensional being that's only temporarily here. When you play Pac-Man or some simple video game, you're fully absorbed in that little world. It's a really kind of shitty world. It's just like you're moving in two dimensions and it's very simple. It's much simpler than what you're capable of, but you're fully absorbed in it. You like it. Then when the game ends, you're like, oh, and then you snap out of it and then your mom's yelling at you or something. What we perceive now as our complex reality could be a much simpler world that we have projected ourselves. We are actually much more complex, powerful beings and we've just for entertainment projected ourselves into this much simpler world to live here for a while. Maybe to experience what it's like to be one of the early ape-like things or to experience what it's like to be afraid of death because maybe these greater beings aren't afraid of but we get to experience it in this limited world. There could be all kinds of things like that, which I think like the ancient Indian philosophers and religious people appreciated this concept actually. I see it in some of their writing. So I like it very much. Man, I'm gonna have a hard time going to sleep today. I'm like, oh, fuck. Yeah, sometimes I just appreciate my low life existence. But okay, this is a perfect time to end, Steve. I think-