Carl Shulman (Pt 1) - Intelligence Explosion, Primate Evolution, Robot Doublings, & Alignment

30:21And so, from where we are right now, seeing the results with GPT -4 and Chattcheap -PT companies like Google and Microsoft and whatnot are pretty convinced that this is very valuable. You have like talk at Google and Microsoft with Bing that, well, it's like billion dollar, billion dollar matter to change market share and search by a percentage point. And so that can fund a lot. And on the far end on the extreme, if you automate human labor, we have a hundred trillion dollar economy. Most of that economy is paid out in wages. So like between 50 and 70 trillion dollars per year. If you create a GI, it's going to automate all of that and keep increasing beyond that.

31:17So the value of the completed project is very much worth throwing our whole economy into it. If you were gonna get the good version, not the catastrophic destruction of the human race, or some other disastrous outcome. And in between, it's a question of, well, the next step, how risky and uncertain is it? And how much growth in the revenue you can generate with it do you get? And so for moving up to a billion dollars, I think that's absolutely going to happen. These large tech companies have R &D budgets, tens of billions of dollars. And when you think about it, like in the relevant sense, like all the employees at Microsoft who are doing software engineering, that's like contributing to creating software objects.

32:08It's not weird to spend tens of billions of dollars on a product that would do so much. And I think it's becoming more clear that there is sort of market opportunity to fund the thing. Going up to $100 billion, that's like, okay, the existing R &D budgets spread over multiple years. But if you keep seeing that when you scale up the model, it substantially improves the performance. It opens up new applications. You're not just improving your search, but maybe, you know, it makes self -driving cars work. you replace bulk software engineering jobs, or if not replace them, amplify productivity. In this kind of dynamic, you actually probably want to employ all the software engineers you can get, how long are they able to make any contribution because the returns of improving stuff in AI itself get so high.

33:02But yeah, that can go up to 100 billion. And at 100 billion, you're using and like a significant fraction of our existing fab capacity. Right now, the revenue of NVIDIA is like 25 billion. The revenue of TSMC, I believe it's like over 50 billion. Last, I checked, in 2021, NVIDIA was maybe 7 .5%, less than 10 % of TSMC revenue. So there's a lot of room and most of that was not AI chips. They have a large gaming segment. There are data center GPUs that are used for video and the like. So there's room for more than an order of magnitude increase by redirecting existing fabs to produce more AI chips and just actually using the AI chips that these companies have in their cloud for the big training runs.

34:07And so I think that's enough to go to the 10 billion and then combine with stuff like the H100s go up to 100 billion. Just to emphasize for the audience, the initial point of our revenue made, if it cost opening a $100 million to trade in GPD4 and it generates $500 million in revenue, you pay back your expenses with a 100 million, you have 400 million for your next training run, then you trade in your GPD 4 .5, You know, you get, let's say, $4 billion out of revenue out of that. That's where the feedback group of revenue comes from, where you're automating tasks, and therefore you're making money, you can use that money to automate more tasks.

34:42On the ability to redirect the fat production towards AI chips. So then the TLDR on, you want $100 billion worth of compute. I mean, fabs take like a decade or so to build. So given the ones we have now and the ones that are going to come online in the next decade, is there enough to sustain $100 billion of GPU compute if you wanted to spend that on a training run? Yes, you definitely make the $100 billion one. As you go up to a trillion dollar run and larger, it's going to involve more fab construction. And yeah, fabs can take a long time to build. on the other hand, if in fact you're getting very high revenue from the AI systems and you're actually bottlenecked on the construction of these fabs, then their price could skyrocket.

35:38And that lead to measures we've never seen before to expand and accelerate fab production. Like if you consider, so at the limit how you're getting models that approach human -like capability. Imagine things that are getting close to brain -like efficiencies plus AI advantages. We were talking before about, well, a GPU that is supporting an AI. Really, it's a cluster of GPU supporting AI is that do things in parallel, data parallelism. But if that can work four times as much as a human, a highly skilled, motivated, focused human with levels of education that have never been seen in the human population.

36:22And so if like a typical software engineer can earn hundreds of thousands of dollars, the world's best software engineers can earn millions of dollars today and maybe more in a world where there's so much demand for AI. And then times four for working all the time. Well, I mean, if you have, if you can generate like close to $10 million a year out of the future version of H100, it costs tens of thousands of dollars with a huge profit margin now. And profit margin could be, could be reduced with like large production. That is a big difference. That chip pays for itself almost instantly. And so you could support paying 10 times as much to have these fabs constructed more rapidly.

37:16You could have, if AI is starting to be able to contribute, you can have AI contributing more of the skill technical work that makes it hard for, say, Nvidia to suddenly find thousands upon thousands of top quality engineering hires. If AI can provide that. Now if AI hasn't reached that level of performance, then this is how you can have things doll out. And like a world where AI progress dolls out is one where you go to the 100 billion and then over succeeding years trillion dollar things software progress, turns out to stall. You lose the gains that you are getting from moving researchers from other fields, lots of physicists and people from other areas of computer science have been going to AI, but you've sort of tap out those resources.

38:07Has it AI becomes a larger proportion of the research field? And like, okay, you've put in all of these inputs, but they just haven't yielded I think that set of inputs probably would yield the kind of AI capabilities needed for intelligence explosion. But if it doesn't, after we've exhausted this current scale of like increasing the share of our economy that is trying to make AI, if that's not enough, then after that you have to wait for the slow grind of things like general economic growth, population growth, and such. and so things slow. And that results in my credences and this kind of advanced AI happening to be relatively concentrated, like over the next 10 years compared to the rest of the century.

38:52Because we just can't keep going with this rapid redirection of resources into AI. That's a one time thing. If the current scale up works, it's going to happen. We're going to get to H .I. really fast, like within the next 10 years or something. If the current scale up doesn't work, All we're left with is just like economy growing like 2 % in years. We have like 2 % a year or more resources to spend on AI. And at that scale, you're talking about decades before you can just your sheer brute force. You can train that $10 trillion model or something. Let's talk about why you have your thesis that the current scale of woodwork.

39:30What is the evidence from AI itself or maybe from primate evolution and the evolution of other animals? just give me the whole, the whole confluence of reasons that make you think. I think maybe the best way to look at that might be to consider when I first became interested in this area. So in the 2000s, which was before the deep learning revolution, how would I think about timelines? How did I think about timelines? And then how have I updated based on what has been happening with deep learning? And so back then, I would have said, we know the brain is a physical object and information processing device.

40:08It works. It's possible. And not only is it possible, it was created by evolution on Earth. And so that gives us something of an upper bound. And that this kind of brute force was sufficient. There are some complexities with like, well, what if it was a freak accident and you know, that didn't happen on all of the other planets and that added some value? I have a paper with Nick Buster on this. I think basically that's not that important issue. There's converging evolution, like octopi are also quite sophisticated. If a special event was at the level of forming cells at all or forming brains at all, we get to skip that because we're choosing to build computers and we already exist.

40:57We have that advantage. So say evolution gives something of an upper bound, really intensive, massive brute force search and things like evolutionary algorithms can produce intelligence. Doesn't the fact that Octopi and I guess other mammals, they got to the point of being like pretty intelligent but not human level intelligent, is that some evidence that there's a heart step between a cephalopod and a human? Yeah, so that would be a place to look. It's done the same particularly compelling. One source of evidence on that is work by Herculano Hutzl. Hope I haven't mispronounced your name, but she's a neuroscientist who has dissolved the brains of many creatures.

41:45And by counting the nuclei, she's able to determine how many neurons are present in different species and find a lot of interesting trends in scaling laws. And she's a paper discussing the human brain has a scaled up primate brain. And across like a wide variety of animals and mammals in particular, there are certain characteristic changes in the relative number of neurons size of different brain regions. things scale up. There's a lot of structural similarity there and you can explain a lot of what is different about us with a pretty brute force story, which is that you expend resources on having a bigger brain, keeping it in good order, giving it time to learn.

42:44So we have We spend more compute by having a larger brain than other animals, more than three times as large as chimpanzees. And then we have a longer childhood than chimpanzees and much more than many other creatures. So we're spending more compute in a way that's analogous to like having a bigger model and having more training time with it. And given that we see with our AI models, this sort of like large consistent benefits from increasing compute spent in those ways, and with qualitatively new capabilities showing up over and over again, particularly in areas that sort of AI skeptics call out.

43:27In my experience like over the last 15 years, the things that people call out has like, ah, but the AI can't do that. And it's because of a fundamental limitation. We've gone through a lot of them. There were Winigrad schemas, catastrophic forgetting quite a number. And yeah, they have repeatedly gone away through scaling. And so there's a picture that we're seeing supported from biology and from our experience with AI, where you can explain like, yeah, in general, there are trade -offs where the extra fitness you get from a brain is not worth it. And so creatures wind up mostly with small brains because they can save that biological energy and that time to reproduce for digestion.

44:18And so on in humans, we actually seem to have wound up in a niche within self -free enforcing where we greatly increase the returns to having large brains. And language and technology are the sort of obvious candidates. when you have humans around you who know a lot of things and they can teach you and compared to almost any other species we have vastly more instruction from parents and the society of the young and then you're getting way more from your brain because you can get per minute you can learn a lot more useful skills and then you can provide the energy you need to feed that brain by hunting and gathering by having fire that makes digestion easier and And basically, how this process goes on, it's increasing the marginal increase in reproductive fitness you get from allocating more resources along a bunch of dimensions towards cognitive ability.

45:17And so, at bigger brains, longer childhood, having our attention be more on learning. So humans play a lot. And we keep playing as adults, which is a very weird thing compared to other animals. we're more motivated to copy other humans around us than like even then the other primates. And so these are sort of motivational changes that keep us using more of our attention and effort on learning, which pays off more when you have a bigger brain and a longer lifespan in which to learn it. Many creatures are subject to lots of predation or disease. And so if you try, you know, you're a mayfly or a mouse, if you try and invest in like a giant brain and a very long childhood, you're quite likely to be killed by some predator or some disease before you're able to actually use it.

46:08And so that means you actually have exponentially increasing costs in a given niche. So if I have a 50 % chance of dying every few months of a little mammal or a little lizard or something, that means the cost of going from three months to 30 months of learning and childhood development, it's not 10 times the loss. It's now two to the negative 10. So a factor of 1 ,024 reduction in the benefit I get from what I ultimately learned because 99 .9 % of the animals will have been killed before that point. We're in an niche where we're like a large, long -lived animal with language and technology so where we can learn a lot from our groups.

46:52And that means it pays off to really just expand our investment on these multiple fronts in intelligence. That's so interesting. Just for the audience, the calculation about like two to whatever months is just like, you have a half chance of dying this month, a half chance of dying next month, you multiply those together. Okay, there's other species though that do live in flocks or as packs where you could to imagine, I mean, they do have like a smaller version of the development of Cubs into that are like play with each other. Why isn't this a hill on which they could have climbed to human level intelligence themselves?

47:35If it's something like language or technology, humans were getting smarter before we got language. I mean, obviously, we had to get smarter to get language, right? We couldn't just get language without becoming smarter. So yeah, we heard it. It seems like there should be other species that should have beginnings of this sort of cognitive revolution, especially given how valuable it is given, listen, we've dominated the A .E. world. You would think there'd be selective pressure for it. Evolution doesn't have foresight. The thing in this generation that gets more surviving offspring and grandchildren, that's the thing that becomes more common.

48:09Evolution doesn't look ahead and they, oh, in a million years, you'll have a lot of descendants. It's what survives and reproduces now. And so in fact, there are correlations where social animals do an average, have larger brains. And part of that is probably that the additional social applications of brains, like keeping track of which of your group members have helped you before so that you can reciprocate. You scratch my back, I'll scratch yours. remembering who's dangerous within the group, that sort of thing. It's an additional application of intelligence. And so there's some correlation there.

48:51But what it seems like is that, yeah, in most of these cases, it's enough to invest more, but not invest to the point where a mine can easily develop language and technology and pass it on. And so there are, you see bits of tool use in some other primates who have an advantage that, so compared to, say, the whales who have, they have quite large brains, probably because they are so large themselves, and they have some other other things, but they don't have hands, which means that reduces a bunch of ways in which brains can pay off, and investments in the functioning of that brain. But yeah, so primates will use sticks to extract termites, That's captioned monkeys will open in clams by smashing them with a rock.

49:40So there's bits of tool use, but what they don't have is the ability to sustain culture. A particular primate will maybe discover one of these tactics and maybe it'll be copied by their immediate group. But they're not holding on to it that well. They like, well, when they see the other animal do it, they can copy it in that situation. They don't actively teach each other. Their population locally is quite small. So it's easy to forget things, easy to lose information. And in fact, they remain technologically stagnant for hundreds of thousands of years. And we can actually look at some human situations.

50:20So there's an old paper, I believe, by the economist, Michael Kramer talks about technological growth in the different continents for human societies. And so you have Eurasia is the largest integrated connected area, Africa is partly connected to it, but the Sahara Desert restricts the flow of information and technology and such. And then you have the Americas, which were after the colonization from the land bridge were largely separated and are smaller than Eurasia. Then Australia and then you had like smaller island situations like Tasmania. And so technological progress seemed to have been faster, the larger the connected group of people.

51:05And in the smallest groups, so like in Tasmania, you had a relatively small population and they actually lost technology. So things like they've lost some like fishing techniques. And if you have a small population and you have some limited number of people who know a skill and they happen to die or happen, there's like, you know, some some change in circumstances that causes people not to practice or pass on that thing, and then you lose it. And if you have a few people, you're doing less innovation, the rate at which you lose technologies to some kind of local disturbance, and the rate at which you create new technologies can wind up in balance.

51:46And the great change of hominids and humanity, if that we wound up in this situation, we we were accumulating faster than we were losing. And as we accumulated, those technologies allowed us to expand our population. They created additional demand for intelligence so that our brains became three times as large. Is that Japan's easier? Yes, Japan's easier. And our ancestors who had a similar brain size. OK, and the crucial point, I guess, in relevance to AI is that the selective pressures against intelligence in other animals are not acting against these neural networks because we are, you know, we're not gonna get like eaten by a predator if they spend too much time becoming more intelligent.

52:30We're like explicitly trading them to become more intelligent. So we have like good first principles reason to think that if it was scaling that made our minds this powerful and if the things that prevented other animals from scaling are not impinging on these neural networks, the decent should just continue to become very smart. Yeah, we are growing them in a technological culture where there are jobs like software engineer that depend much more on sort of cognitive output and less on things like metabolic resources devoted to the immune system or to like building big muscles to throw spears.

53:07Yeah, this is kind of a side note, but I'm just kind of interested. I think you're reference to some point gentleness scaling for the audience. This is a paper from DeepMind which describes if you have a model of a certain size, What is the optimum amount of data that it should be trained on? So you can imagine bigger models. You can use more data to train them. And in this way, you can figure out where should you spend your computer? Should you spend it on making the model bigger? Or should you spend it on training it for longer? I'm curious if in the case of different animals, in some sense, they're like model sizes.

53:37They're how big their brain is. And they're training data sizes. Like how long they're clubs or how long they're infants or toddlers or before they're full adults. Is there some sort of like scaling law of? Yeah, I mean, so the the chinchilla scaling is interesting because the we were talking earlier about the cost function for having a longer childhood And so where it's like exponentially increasing in the amount of training compute you have when you have Exogenous forces that can kill you Where when we do big training runs the cost of throwing in more GPUs is almost linear and it's much better to be you know linear than exponentially decay.

54:13Oh, that's a really good question. I do expand resources and so So, Chinchilla scaling would suggest that like, yeah, the opt for a brain of sort of human size, it would be optimal to have many millions of years of education. But obviously, that's impractical because of exogenous mortality for humans. And so, there's a fairly compelling argument that relative to the situation where we would train AI, that animals are systematically way under trained. And now they're more efficient than our models. We still have room to improve our algorithms to catch up with the efficiency of brains, but they are laboring under that disadvantage, yeah.

54:58That is so interesting. Okay, so I guess another question you could have is, humans got started on this evolutionary hill climbing route where we're getting more intelligent than it has more benefits for us. Why didn't we go all the way on that route? If intelligence is so powerful, why aren't all humans as smart as we know humans can be? At least that's smart. If intelligence is so powerful, why hasn't there been stronger selective pressure? I understand, like, oh, listen, hip size, you can't give birth to a really big -headed baby or whatever, but you would think evolution would figure out some way to offset that, if intelligence has such big power.

55:38And such is so useful. Yeah, I think you could actually look at it quantitatively. that's not true. And even in sort of recent history, there has been, it looks like a pretty close balance between the costs and the benefits of having more cognitive abilities. And so you say like, you know, who needs to worry about like, you know, the metabolic costs? Like, you need, humans put, like, order 20 % of our metabolic energy into the brain and it's higher for like young children. So 20 % of the, I mean, along with, and then, you know, there's like breathing and digestion and the immune system. And so for most of history, people have been dying left and right.

56:23Like a very large proportion of people will die of infectious disease. And if you put more resources into your immune system, you survive. That's like life or death pretty directly via that mechanism. them. And then this is this related also people die more of disease during famine. And so there's boomer bust. And so if you have 20 % less metabolic requirements or has anger a child and it's you have a lot more, it's like 40 or 50 % less metabolic requirements, you're much more likely to survive that famine. So these are these are pretty big. And then there's a trade off about just cleaning and mutational load.

57:06So every generation, new mutations and errors happen in the process of reproduction. And so like we know there are many genetic abnormalities that occur through new mutations each generation. And in fact, down syndrome is the chromosomal abnormality that you can survive. All the others just kill the embryo. And so we never see them. But like down syndrome occurs a lot. And there are many other lethal mutations and there's, as you go to the less damaging ones, there are enormous numbers of less damaging mutations that are degrading every system in the body. And so evolution in each generation has to pull away at some of this mutational load.

57:53And the priority with which that mutational load is pulled out, scales and proportioned how much the traits it's affecting impact fitness. So you got new mutations that impact your resistance to malaria. You got new mutations that damage brain function. And then, have those mutations are purged each generation. If malaria is a bigger difference in mortality than like the incremental effectiveness as a hunter gatherer you get from being slightly more intelligent, and then you'll purge that mutational load first. And similarly, if there's like, humans have been vigorously adapting to new circumstances.

58:36So since agriculture, people have been developing things like the ability to have amulets to digest breads, the ability to digest milk. And if you're evolving for all of these things, and if some of the thing that give an advantage for that, but incidentally carry along nearby them some negative effect on another trait, then that other trait can be damaged. So it really matters how important to survival and reproduction cognitive abilities were compared to everything else that organism has to do. And that in particular, like surviving feast and famine, having like the physical abilities to do hunting and hunting and gathering.

59:18And like even if you're like very good at planning your hunting, being able to throw a spear harder can, you know, be a big difference. And that needs energy to build those muscles and then to sustain them. And so given all these, all these factors, it's like, yeah, it's not a slam dunk to invest at the merge. And like today, like having bigger brains, for example, is associated with like greater cognitive ability, but it's like, it's modest. Large scale pre -register studies, pre -registered studies with MRI data. It's like a range maybe like a correlation of 0 .25, 0 .3. And the standard deviation of brain size is 10%.

1:00:04So if you double the size of the brain, and the existing brain costs 20 % of metabolic energy, go up to 40%. OK, it's like eight standard deviations, a brain size, if the correlation is 0 .25. than, yeah, like you can, you get a gain from that. Eight standard deviations of brain size, two standard deviations of cognitive ability. And like in our modern society, where cognitive ability is very rewarded, and like, you know, finishing, finishing school, becoming an engineer or a doctor or whatever can, can pay off a lot financially. still the like the average observed return in like income is like a 1 or 2 % proportional increase.

1:00:56There's more effects of the tail. There's more effect in professions like STEM. But on the whole it's not like, you know, if it was like a 5 % increase or a 10 % increase, then you could you could tell a story where yeah this is hugely increasing the amount of food you could have. You could support more children, but it's like, it's a modest effect and the metabolic costs would be large and then throw in these other aspects. And I think it's, you can tell the story. And also, we can see there was not very strong rapid directional selection on the thing, which there would be if like, you know, you could, by solving like a, by solving like a math puzzle, you could defeat malaria.

1:01:38area. Then there would be more evolutionary pressure. That is so interesting. And I was not to mention, of course, that, yeah, if you had like 2X the brain size, you're without C section, you would, like, your mother would, or both would die. This is a question I've actually been curious about for like over a year. And I've like, briefly tried to look up an answer. This is, I know this is off topic, but I apologize to the audience, but I was super interested in those, like, that was the most comprehensive and interesting answer I could hope for. Okay. So, yeah, we have a good explanation or good first principles evolutionarves and for thinking that intelligence scaling up to humans is not implausible despite throwing more scale at it.

1:02:18I would also add, this was something that would have mattered to me more in the 2000s. We also have the brain right here with us for available for neuroscience to reverse engineer its property. And so in the 2000s when I said, yeah, I expect this by, you know, middle of the century that was a backstop if we found it absurdly difficult to go to the algorithms and then we would learn from neuroscience. But in the actual history, it's really not like that. We developed things in AI and then also we can say, oh, yeah, this is sort of like this thing in neuroscience or maybe this is a good explanation.

1:02:54It's not, though neuroscience is driving AI progress. It turns out not to be that necessary. A similar to, I guess, how planes were inspired by the existence of births, but jet engines don't flap. All right, so yeah, good reason to think scaling might work. So we spent $100 billion and we have something that is like human level or can do help significantly with AI research. I mean, that might be on the earlier end, but I mean, I definitely would not rule that out given the rates of change we've seen with the last few scale ups. All right. So at this point, somebody might be skeptical, okay, like listen, we already have a bunch of human researchers, right?

1:03:39Like the incremental researcher, how powerful is that? And then you might say, well, no, this is like thousands of researchers. I don't know how to express a skepticism exactly, but skepticism is a skeptical of just generally the effect of scaling up the number of people working on the problem to rapid rapid progress on that problem. Somebody might think, okay, listen, with humans, the reason population working on a problem is such a good proxy for progress on the problem is that there's already so much variation that is accounted for when you say there's like a million people working on a problem.

1:04:07You know, there's like a hundred of super geniuses working on it, thousands of people who are like very smart working on it, whereas within the eye, all the copies are like the same level of intelligence. And if it's not super genius intelligence, the total quantity, you might not matter as much. Yeah, I'm not sure what your model is here. So is this the model that the diminishing returns kickoff suddenly has a cliff right where we are? And so like there was, there were results in the past from throwing more people at problems. And I mean, this has been useful in historical prediction. One of the, there's idea of experience curves and rights law, basically measuring cumulative production in a field, which is also going to be a measure of the scale of effort and investment.

1:05:05And people have used this correctly to argue that renewable energy technology like solar would be falling rapidly in price because it was going from a low base of very small production runs, not much investment, and doing it efficiently. And yeah, climate advocates correctly called out. People like David Roberts, the futurist Rama has now actually has some interesting writing on this that, yeah, correctly called out that there would be really drastic fall and prices of solar and batteries because of the increasing investment going into that, the human genome project would be another. So I'd say there's like, yeah, real, real evidence, these observed correlations from ideas getting harder to find have, have held over a fair, a fair range of data and over quite a lot of time.

1:06:02So I'm wondering what, what, what's the, yeah, the, the nature of the deviation you're thinking of? That we're talking about maybe this is like a good way to describe what happens when more humans enter a field. But does it even make sense to say like a greater population of AI's is doing AI research if there's like more GPUs running a copy of GPT -6 doing AI research? It's just like how how applicable are these economic models of human the quantity of humans working on a problem to the magnitude of AI's working on a problem? Yeah so if you have AI that are directly automated in particular jobs that humans were doing before then we say well with additional compute, we can run more copies of them to do more of those tasks simultaneously.

1:06:51We can also run them at greater speed. And so some people have an intuition that like, well, you know, what matters is like time. It's not how many people working on problem at a given point. I think that doesn't bear out super well. But AI can also be run faster than humans. And so if you have of a set of AIs that can do the work of the individual human researchers and run at 10 times or 100 times the speed. And we ask, well, could the human research community have solved these algorithm problems, do things like invent transformers over 100 years if we have this. We have AIs with a population, effective population, similar to the humans, but running 100 times as fast.

1:07:39And so you have to tell a story where no, the AI, they can't really do the same things as the humans. And we're talking about what happens when the AI is more capable of in fact doing that. Although they become more capable as lesser capable versions of themselves help us make big themselves more capable, right? So you have to like kickstart that at some point. Is there an example in analogous situations? Is intelligence unique in the sense that you have a feedback group of with a learning curve or something else? A system outputs are feeding into its own inputs in a way that because if we're talking about something like Moore's Law or the cost of solar, you do have this way like where you know more people are, we're throwing more people out of the problem and it's where you know we're making a lot of progress.

1:08:30But we don't have this sort of additional part of the model where Moore's Lollies do more humans somehow and The more humans are becoming researchers. So you do actually have a version of that in the case of solar So you have a small Infant industry that's doing things like providing solar panels for space satellites and then getting increasing amounts of subsidized government demand because of You know worries about fossil fuel depletion and then climate change You can't have the dynamic where visible successes with solar or like lower in prices then open up new markets. So there's a particularly huge transition where renewables become cheap enough to replace large chunks of the electric grid.

1:09:17Earlier you're dealing with very niche situations like, yeah, so the satellites where you have a very difficult to refuel, a satellite in place and then remote areas and then moving to like, you know, the super, the sunniest areas in the world with the biggest solar subsidies. And so there was an element of that where more and more investment has been thrown into the field and like the market has rapidly expanded as the technology improved. But I think the closest analogy is actually the long run growth of human civilization itself. And I know you had holding Karnovsky from the Open Philosophy Project on earlier and discuss some of this research about the long run acceleration of human population and economic growth.

1:10:04And so developing new technologies allowed human population to expand humans to occupy new habitats and new areas and then to invent agriculture which support the larger populations and then even more advanced agriculture in the modern industrial society. And so their total technology and output allowed you to support more humans who then would discover more technology and continue the process. Now that was boosted because on top of expanding the population, the sheer of human activity that was going into invention and innovation went up. And that was a key part of the industrial revolution. There was no such thing as a corporate research lab or like an engineering university prior to that.

1:10:51And so you were both increasing the total human population and the share of it going in. But this population dynamic is pretty analogous. Humans invent farming, they can have more humans than they can invent industry and so on. So maybe somebody would be skeptical that with AF progress specifically, it's not just a matter of some like some farmer figure your own crop rotation or some blacksmith figuring out how to do metallurgy better. You in fact, even to make the, for the 50 % improvement in productivity, you basically need something on the IQ that's close to Ilya Setskoper. So there's a good discontinuous, you're like contributing very little to productivity and then you're like Ilya and then you contribute a lot.

1:11:34But the becoming Ilya is, you see what I'm saying? There's not like a gradual increase in capabilities at least to the feedbacker. You're imagining a case where the distribution of tasks is such that there's nothing that you can where individually automating it particularly helps. And so the ability to contribute to AI research is really end loaded, is that what you're saying? Yeah, I mean, we already see this in these sorts of like, really high IQ companies or projects where theoretically, I guess Jane Street or OpenAI could hire like a bunch of, you know, mediocre people to do, there's a comparative advantage, they could do some menial tasks, and that could free up the time of the really smart people, but they don't do that, right?

1:12:18Which transaction costs, whatever else. Self -driving cars would be another example where you have a very high quality threshold. And so when your performance has a driver is worse than a human, like you have 10 times the accident rate or 100 times the accident rate, then the cost of insurance for that, which is a proxy for people's willingness to write the current instead of two would be such that the insurance cost would absolutely dominate. So even if you have zero labor cost, it's offset by the increased insurance cost. And so there are lots of cases like that where like partial automation is not in practice very usable because complementing other resources, you're going to use those other resources less efficiently.

1:13:01In a a post -AGI future and the same thing can apply to humans. So people can say, well, comparative advantage, even if AI's can do everything better than a human, well, it's still worth something. The human can do something, make them, you can lift a box, that's something. Now, there's a question of property rights, if, well, if it could just, say something to humans, it's going to make more robots. But even absent that in such an economy, You wouldn't want to let a human worker into any industrial environment because in a clean room they'll be emitting all kinds of skin cells and messing things up.

1:13:41You need to have an atmosphere there. You need a bunch of supporting tools and resources and materials and those supporting resources and materials will do a lot more productively working with AI and robots rather than a human. So you don't want to let a human anywhere near the thing just like in a you don't want to have a a gorilla wondering around in a Chinese shop, even if you've trained it to most of the time, pick up a box for you if you give it a banana, it's just not worth it to have it wondering around your Chinese shop. Yeah, yeah, yeah. Like why is that not a good objection to? I mean, I think that is one of the ways in which partial automation can fail to really translate into a lot of economic value.

1:14:23That's something that will attenuate as we go on. And as the AI is more able to work independently and more able to handle its own, its own scrubs get more, more reliable. But the way in which it becomes more reliable is by AI progress, beating up which happens if AI can contribute to it. But if there is some sort of reliability bottle, like the forensic from contributing to the progress, then you don't have the loop, right? So, yeah, I mean, this is why we're not there yet. Right, but then what is the reason to think we'll be there at? The broad reason is we have these inputs are scaling up. There's a so epoch, which I mentioned earlier.

1:15:04They have a paper, I think it's called compute trends in three areas of machine learning or something like that. And so they look at the compute expanded on machine learning systems since the founding of the field of AI, the beginning of the 1950s. And so, it mostly, it grows with Moore's Law. And so, people are spending a similar amount on their experiments, but they can just buy more with that because the computer's coming. And so, that data, I mean, it covers over 20 orders of magnitude, maybe like 24. And of all of those increases since 1952, a little more than half of them happened between 1952 and 2010.

1:15:51And all the rest is since 2010. So we've been scaling that up like four times as fast as with the case for most of the history of AI. We're running through the orders of magnitude of possible resource inputs you could need for AI much, much more quickly than we were for most of the history of AI. That's why this is a period of like with a very elevated chance of AI per year because we're moving through so much of the space of inputs per year. And indeed, it looks like this scale up taken taken to its conclusion will cover another bunch of orders of magnitude. And that's actually a large fraction of those that are left before you start running into saying, well, this is going have to be like evolution with the sort of simple hacks we get to apply.

1:16:44Like we're selecting for intelligence the whole time. We're not going to make do the same mutation that causes fatal, fatal childhood cancer a billion times, even though I mean, we keep getting the same fatal mutations, even though they've been done many times. We use gradient descent, which takes into account the derivative of improvement on the loss all throughout the network. And we don't throw away all the contents of the network with each generation where you compress down to a little DNA. So there's that bar of like, well, if you're going to do brute force like evolution, combine with these sort of very simple ways, we can save orders of magnitude on that.

1:17:23We're going to cover, I think, a fraction that's like half of that distance in this scale up over the next 10 years or so. And so if you started off with a kind of vague uniform prior, you're like, well, it's probably, you probably can't make a GI with like the amount of compute that would be involved in a fruit fly existing for a minute, which would be the early days of AI. You know, maybe you would get lucky. We were able to make calculators because calculators benefited from like very reliable, seriously fast computers. and where we could take a tiny, tiny, tiny, tiny fraction of a human brain's compute and use it for a calculator.

1:18:03We couldn't take an ants brain and rewire it to calculate it's hard to manage ant farms, let alone get them to do arithmetic for you. And so there were some things where we could exploit the differences between biological brains and computers to do stuff superficially on computers. we would doubt that we would be able to do so much better than biology, that with a tiny fraction of an insect's brain, we'd be able to get AI early on. On the far end, it seemed very implausible that we couldn't do better than completely brute force evolution. And so in between, you have some number of orders of magnitude of inputs, where it might be.

1:18:45And like in the 2000s, I would say, well, you know, I'm going to have a pretty uniform -ish prior. I'm going to put weight on it happening at like the sort of the equivalent of like 10 to the 25 ops 10 to the 30 10 to the 35 And sort of spreading out over that and then I can update on other information and in the short term I would say like in 2005 I would say well, I don't see anything That looks like the cusp of a GI So I'm also going to lower my credence for like the next five years or the next 10 years and so that that would be kind of like a big prior And then when we take into account, like, well, how quickly are we running through those orders of magnitude?

1:19:24If I have my uniform prior, I assign half of my weight to the first half of remaining orders of magnitude. And we're going to run through those over the next 10 years and some. Then that calls me to put half of my credence conditional on wherever I'm going to make it. Yeah, I'm which seems likely, but it's a material object easier than evolution. and I've got to put it similarly, a lot of my credence on AI happening in the scale. And then that's supported by what we're seeing in terms of the rapid advances in capabilities with AI and elevums in particular. Okay, that's a really interesting point.

1:20:01So now, but somebody might say, listen, there's not some sense in which AI's could universally speed up the progress of open AI, about 50 % or 100 % or 200%. If they're not able to do everything better than Ilya Suskovurkan, there's going to be something in which we're bottlenecked by the human researchers. And bottleneck effects dictate that the slowest moving part of the organization will be the one that kind of determines the speed of the progress of the whole organization or the whole project, which means that unless you get to the point where you're like doing everything and everybody in the organization can do, you're not going to significantly speed up the progress of the whole project as a whole.

1:20:40Yeah, so that is hypothesis. And I think there's a lot of truth to it. So when we think about the ways in which AI can contribute, so there are things we talked about before, like the AI is setting up their own curriculum. And that's something that Ilya can't do directly, doesn't do directly. And there's a question, how much does that improve performance? There are these things where the AI helps to just like produce some code for some time. And it's beyond hello world at this point. But I mean, the sort of thing that I hear from AI researchers at Leading Labs is that, you know, on their core job where they're like most expert, it's not helping them that much.

1:21:23But then, you know, their job often, it does involve, oh, I've got to code something that's out of my usual area of expertise. Or I want to research this question and it helps them there. And so that saves some of their time and frees them to do more of the bottleneck work. And then I think the idea of, well, is everything dependent on Ilya and his Ilya so much better than the hundreds of other employees. I think a lot of people who are contributing, they're doing a lot of tasks. And so you can have quite a lot of gain from automating some areas where you then do just an absolutely enormous amount of it relative to what you would have done before because things like designing the custom curriculum.

1:22:12You're like, maybe you had some humans put some work into that, but you're not going to employ billions of humans to produce it at scale. And so it winds up being a larger share of the progress than it was before. You get some benefit from these sorts of things where, yeah, there's like pieces of my job that now I can hand off to the AI and it's me focus more on the things that the AI still can't do. And then at the later on, you get to the point where, yeah, the AI can do your job, including the most difficult parts. and maybe it has to do that in a different way. Maybe it, like, spends a ton more time thinking about each step of a problem than you.

1:22:56And that's the late end. And the stronger these bottlenecks effects are, the more the economic returns, the scientific returns, and such are unloaded towards getting sort of full AGI. The weaker the bottlenecks are, the more interim results will be really pain off. I guess I probably disagree with you on how much the sort of the the elias of organizations seemed to matter I guess just from the evidence alone like how many of the big sort of breakthroughs that in deep learning in general was like that single individual responsible for right And how much of his time is he spending doing anything that's in that like copilot is helping him on I'm guessing like most of it's just like managing people and coming up with the ideas and Yeah, turn it like understand systems and so on and if that is the If the five or 10 people who are like that at OpenAI or Anthropica, or whatever, are basically the way in which the progress is happening, or at least the algorithmic progress is happening, then how much of better and better co -pilot, I know co -pilot is not the thing you're talking about with like 20 % automation, but something like that, how much is that contributing to the sort of core function of the research scientist?

1:24:11Yeah, not quantitatively how much we disagree about the importance of sort of key research employees and such. I certainly think that some researchers add more than 10 times the average employee, even much more, and obviously managers can add an enormous amount of value by proportionately multiplying the output of the many people that they manage. And so that's the kind of thing that we were discussing earlier when talking about, well, if you had sort of full human level AI or AI that had all of the human capabilities plus AI advantages, it would be, you'd benchmark not off of what the sort of typical human performances but peak human performance and beyond.

1:25:05So yeah, I accept all that. I do think it makes a big difference for people how much they can outsource a lot of the, that are less, wow, less creative and an enormous amount is learned by experimentation. ML has been quite experimental field and there's a lot of engineering work in say building large super clusters, making hardware, where optimization and coding of these things being able to do the parallelism in large models. And the engineers are busy, and it's not just only a big thoughts kind of area. And then the other branch is where will the AI advantages and disadvantages be? And so one AI advantage is being omnidisciplinary and familiar with the newest things.

1:26:08So I mentioned before, there's no human who has a million years of TensorFlow experience. And so to the extent that we're interested in like the very, very cutting edge of thing that have been developed quite recently, then AI that can learn about them in parallel, an experiment and practice with them in parallel can learn much faster than a human potentially. And the area of computer science is one that is especially suitable for AI to learn in a digital environment. So it doesn't require like driving a car around that might kill someone, have enormous cost, you can do unit tests, you can you can prove theorems, you can do all sorts of operations entirely in the confines of a computer, and which is one reason why programming has been benefiting more than a lot of other areas from LLMs recently, whereas robotics is lagging.

1:27:06Some of that, and then just considering, well, actually, I mean, they aren't getting better at things like the GRE math at programming contests. And I mean, some people have forecasts and prediction that it's standing about things like, doing well on the informatics Olympiad and the math Olympiad. And in the last few years, when people tried to forecast the MMLEU benchmark, which was having a lot of more sophisticated kind of, you know, like graduate student, science kind of questions. Yeah, AI knocked that down a lot faster than AI researchers who had registered and students who had registered for us on it.

1:27:53And so if you're getting top notch scores on, you know, graduate exam, creative problem solving, yeah, it's not obvious that that sort of area will be a relative weakness of AI. that in fact computer science is in many ways especially suitable because of getting up to speed with new areas being able to get rapid feedback from the interpreter at scale. But do you get rapid feedback if you're doing that something that's more now goes to research? If you're like let's say you have a new model or something and it's like if we put in $10 million on a mini trading run on this this would be a much bit.

1:28:37Yeah, for very large models, those experiments are going to be quite expensive. And so you're going to look more at like, can you build up this capability by generalization from things like many math problems, programming problems, working with small networks? Yeah, fair enough. I actually, the Scott Ehrenson was one of my professors in college, and I took its quantum information class. And I didn't do, I did okay in it, but he recently I wrote a blog post where he said, I had GP4 take my quantum information test, and it got a B. And I was like, yeah, I gotta see. I'm at the final. So yeah, yeah, I'm updated in the direction that, you know, it seems getting a view on the test, like you probably understand quantum information pretty well.

1:29:21With different areas of strengths and weaknesses than the human students. Sure, sure. Would it be possible for this sort of intelligence explosion to happen without any sort of hardware progress? if hardware progress stopped, would this feedback loop still be able to produce some sort of explosion with only software? Yeah, so if we say that the technology is frozen, which I think is not the case right now, the Nvidia has managed to deliver significantly better chips for AI workloads for the last few generations, h100, a100, v100. If that stops entirely, then what you're left with, and maybe we'll define this as like no more nodes, more as lies over.

1:30:06At that point, the kind of gains you get and amount of compute available come from actually constructing more chips. And there are economies of scale you could still realize there. So right now, a chipmaker has to amortize the R &D cost I've developed in the chip. And then the capital equipment is created. You build a fab, its peak profits are gonna come in the few years when the chips it's making are at the cutting edge. Later on, has the cost of compute exponentially falls. The, you know, you keep the fab open because you can still make some money given that it's built. But of all of the profits the fab will ever make.

1:30:47Right now, they're relatively front loaded because when its technology isn't near the cutting edge. So in a world where Moore's Law ends, then you wind up with these very long production runs where you can keep making chips that stay at the cutting edge. And where the R &D costs get amortized over a much larger base, so the R &D basically drops out of the price. And then you get some economies of scale from just making so many fabs in the way that when we have the auto industry expands, and then this is in general across industries, when you produce a lot more cost -fall, because you have right now, like ASML has many, incredibly exotic suppliers that make some bizarre part of the thousands of parts in one of these ASML machines.

1:31:40You can't get it anywhere else. They don't have standardized equipment for their thing, because this is the only only use for it. And in a world where we're making 10, 100 times as many chips at the current node, then they would benefit from scale economies and all of that would become more mass production industrialized. And so you combine all of those things and it seems like capital costs of like buying a chip would decline, but the energy cost of running the chip would not. And so right now, energy costs are a minority of the cost, but they're not trivial. They pass one percent a while ago, and they're inching up towards 10 % and beyond.

1:32:28And so you can maybe get another order of magnitude cost decrease from getting really efficient and the capital construction, but energy would still be a limiting factor after the end of actually improving the chips themselves. Right, I got it. And when you say there would be a greater population of AI researchers because are we using population as a thinking tool of how they could be more effective? Or do you literally mean that the way you expect these AI to contribute a lot to researchers by just having a million copies of a researcher thinking about the same problem? or is it just like a usual thinking model for what it would look like to have a million times moderate yet working on that problem?

1:33:11That's definitely a lower bound sort of model and often I'm meaning something more like effective population or like you'd need this many people to have this effect. And so we were talking earlier about the trade -off between training and inference in board games. And so you can get the same performance by having a bigger model or by calling in the model more times. And in general, it's more effective to have a bigger, smarter model and call it less time up until the point where the costs equal as between them. And so we would be taking some of the gains of our larger compute on having bigger models that are individually more capable.

1:33:52And there would be a division of labor. So the tasks that were most cognitively demanding would be done by these giant models. But some very easy task. You don't want to expend that giant model if a model 100, the size can take that task. And so larger models would be in the positions of researchers and managers, and they would have swarms of AIs of different sizes, as tools that they can make API calls to and whatnot. OK, we accept the model. And now we've gone to something that is at least such a smart as Giliya Suscover on all the tasks relevant to AI progress. And you can have so many copies of it.

1:34:31What happens in the world now? What are the next months or years or whatever timeline is relevant look like? And so, and to be clear, what's happened is not that we have something that has all of the abilities and advantages of humans plus the AI advantages. Well, we have something that is like possibly by doing things like doing a ton of calls to make up for being individually less capable or something, it's able to drive forward AI progress. That process is continuing. So AI progress has accelerated greatly in the course of getting there. And so maybe we go from our eight months doubling time of software progress in effective compute to four months or two months.

1:35:17And so there's a report by Tom Davidson at the Open Philanthropy Project, which spun out of work I had done previously. And so I advised and helped with that project. But Tom really carried it forward and produced a very nice report and model, which epoch is hosting. You can plug in your own version of the parameters. And there is a lot of work estimating the parameter things like what's the rate of software progress, what's the return to additional work, how does performance scale at these tasks as you boost the models. In general, as we were discussing earlier, these broadly human level in every domain with all the advantages is pretty deep into that.

1:36:15If already, we can have an eight months doubling time for software progress. then by the time you get to that kind of point, it's maybe more like four months, two months, going into one month. And so if the thing is just proceeding at full speed, then each doubling can come more rapidly. And so we can talk about what are the spills over us of like, so how does the models get more capable? They can be doing other stuff in the world. They can spend some of their time making Google search more efficient. They can be hired as chat bots with some inference compute. And then we can talk about sort of if that intelligence explosion process is allowed to proceed, then what happens is, OK, you improve your software where by a factor of two, the demand, the efforts needed to get the next doubling are larger, but they're not choice of large.

1:37:22Maybe they're like 25%, 35 % larger. So each one comes faster and faster until you hit limitations. You can no longer make further software advances with the hardware that you have. And looking at, I think, reasonable parameters in that model, it seems to me, if you have these giant training runs, you can go very far. And so the way I would see this playing out is, has the AI get better and better at research? They can work on different problems. They can work on improving software, working on improving hardware. They can do things like create new industrial technologies, new energy technology.

1:38:05they can manage robots, they can manage human workers, as like executives and coaches and whatnot. You can do all of these things. And AI is wind up being applied where the returns are highest. And I think initially, the returns are especially high in doing more software. And the reason for that is again, if you improve the software, you can update all of the GPUs that you have access to, your cloud compute is suddenly more potent. If you design a new chip design, it'll take a few months to produce the first ones and it doesn't update all of your old chips. So you have an ordering where you start off with the things where there's the lowest dependence on existing stocks and you can more just take whatever you're developing and apply it immediately.

1:39:04And so software runs ahead. You're getting more towards the limits of that software. And I think that means things like having all the human advantages, but combined with AI advantages. And so, we're just going to think that means, given the kind of compute that would be involved, if we're talking about this hundreds of billions of dollar, trillion dollar training run, there's enough compute to run tens of millions, hundreds of millions of human scale minds. They're probably smaller than human scale, to be similarly efficient at the limits of other than the progress because they have the advantage of a million years of education.

1:39:48They have the other advantages we talked about. You've got that wild capability and the software, further software gains are running out. or like they start to slow down again because you're just getting towards the limits of like you can't do any better than the best. And so what happens then? Yeah. By the time they're running out, have you already hit super intelligence or? Yes, you're, okay. Your widely super intelligent. We've left the galaxy, okay. Even in the region. Just by having the abilities that humans have and then combining it with being very well focused and trained in the task beyond what any human could be, and then running faster and such.

1:40:28Got it, got it. Sorry, continue. Yeah, so I'm not going to assume that there's huge qualitative improvements you can have. I'm not going to assume that humans are very far from the efficient frontier of software, except with respect to things like, yeah, we had limited lifespan, so we couldn't train super intensively. We couldn't incorporate other software into our brains. We couldn't copy ourselves. We couldn't run it fast speeds. Yeah, so you've wrote all of those capabilities. And now I'm skipping ahead of the most important months in human history. And so I can talk about what it looks like if it's just the AI's took over.

1:41:14They're running things as they like how do things expand. And I can talk about things as how does this go in a world where we've roughly or at least so far managed to retain control of where these systems are going. And so by jumping ahead, I can talk about how would this translate into the physical world? And so this is something that I think is a stopping point for a lot of people in thinking about, well, what would an intelligence explosion look like? and they have trouble going from, well, there's stuff on servers and cloud compute, and oh, that gets very smart. But then, how does what I see in the world change?

1:41:57How does like industry or military power change, if there's an AI takeover, like, what does that look like? Are there killer robots? And so, yeah, so one course we might go down is to discuss during that wildly accelerating transition. How did we manage that? How do you avoid it being catastrophic? And another route we could go is how does the translation from wildly expanded scientific R &D capabilities intelligence on these servers translate into things in the physical world? So you're moving along in order of like what has the quickest impact largely or like where you can have an immediate change.

1:42:49So one of the most immediately accessible things is where we have large numbers of devices or artifacts or capabilities that are already AI -operable with hundreds of millions equivalent research as you can like quickly solve self -driving cars, you know, make the algorithms much more efficient, do great testing and simulation, and then operate a large number of cars in parallel if you need to get some additional data to improve the simulation in recenting. Although, you know, in fact, humans with quite little data are able to achieve human level driving performance. So So after you've really maxed out the easily accessible algorithmic improvements in this software -based intelligence explosion that's mostly happening on server farms, then you have you have minds that have been able to really perform on a lot of digital only tasks that they're doing great on video games.

1:43:57They're doing great at predicting what happens next in a YouTube video. If you have a camera that they can move, they're able to predict what will happen at different angles humans do this a lot where we naturally move our eyes in such a way to get images from different angles and different presentations and then predicting combined from that. And yeah, and you can operate many cars, many robots at once to get very good robot controllers. So you should think that all the existing robotic equipment or remotely controllable equipment that is wired for that. The areas can operate that quite well.

1:44:38I think some people might be skeptical that existing robots, given their current hardware, have the dexterity and the maneuverability to do a lot of physical labor that, and yeah, I might want to do every simple thing otherwise. There's also not very many of them. So production of sort of industrial robots is hundreds of thousands per year. they can do quite a bit in place. Evil I must is promising a robot in the tens of thousands of humanoid robot in the tens of thousands of dollars. That may take a lot longer than any set as it happened with other technologies. But that's a direction to go. But most immediately, so hands are actually probably the most scarce thing.

1:45:26But if we consider what do human bodies provide? So there's the brain. And in this situation, we have now an abundance of high quality brain power that will be increasing as the AI will design new chips, which will be rolling out from the TSMC factories. And they'll have ideas and designs for the production of new fab technologies, new nodes, and additional fabs. But yeah, looking around the body. So there's legs to move around. Not only that necessary, wheels work pretty well being in a place. You don't need most people, most of the time, in factory jobs and in office jobs. Office jobs, many of them can be fully virtualized.

1:46:10But yeah, some amount of legs, wheels, other transport, you have hands and hands are something that are on the expensive end. In robots, we can make them. They're made in very small production runs Partly because we don't have the control software to use them well with in this world the control software is fabulous And so people will produce much larger production runs of them Over time possibly using technology we recognize possibly with quite different technology, but just taking what we've got so right now the Robot arm industrial robot industry produces hundreds of thousands of machines a year.

1:46:53Some of the nicer ones are like $50 ,000. In aggregate, the industry has tens of billions of dollars of revenue. By comparison, the automobile industry produces like I think over 60 million cars a year. It has revenue of over $2 trillion per annum. And so converting that production capacity over towards robot production would be one of the things, if there's not something better to do, would be one of the things to do. And in World War II, industrial conversion of American industry took place over several years and really amazingly ramped up military production by converting existing civilian industry.

1:47:45And that was without the aid of superhuman intelligence and management at every step in the process. So yeah, every part of that would be very well designed. You'd have AI workers who understood, stood every part of the process and could direct human workers, even in a fancy factory most of the time, It's not the hands doing a physical motion that a worker has been paid for. They're often looking at things or deciding what to change. The actual, the time spent in manual motion is a limited portion of that. And so in this world of abundant AI cognitive abilities, where the human workers are more valuable for their hands than their heads, then you could have a worker, even a worker previously without training and expertise in the area who has a smartphone, maybe a smartphone on a headset, and we have billions of smartphones which have eyes and ears and methods for communication, for an AI to be talking to a human and directing them in their physical motions, with skill as a guide and coach that is beyond any human, there could be a lot better at telepresence and remote work.

1:49:10And they can provide VR and augmented reality guidance to help people get better at doing the physical motions that they're providing in the construction. So you convert the auto industry to robot production. If it can produce an amount of mass of machines that is similar to what it currently produces. That's enough for, you know, billion, billion human -sized robots a year. The value per kilogram of cars is somewhat less than high -end robots. But, yeah, you're also cutting out most of the wage bill. Because most of the wage bill is payments ultimately to like human capital and education. not to the physical hand motions and lifting objects and that sort of task.

1:50:07Yeah, so at the sort of existing scale of the auto industry you can make a billion robots a year. The auto industry is two or three percent of the existing economy. You're replacing these cognitive things. So if right now physical hand motions are like 10 % of the work redirect humans into those tasks And you have, like in the world at large right now, mean income is on the order of $10 ,000 a year, but in rich countries, skilled workers earn more than 100 ,000 per year. And some of that is just like, you know, it's not just management roles of which only a certain proportion of the population can have, but just being an absolutely, exceptional peak end to performance of some of these construction and and such rules.

1:51:02Yeah, just raising productivity to match the most productive workers in the world. You know, it's is room to make a very very big gap. And with AI replacing skills that are scarce in many places, where there's you know abundant currently low wage labor. You bring in the AI coach, and someone who was previously making very low wages can suddenly be super productive by just being the hands for an AI. And so, you know, on an naive view, if you ignore the delay of capital adjustment, of like building new tools for the workers, say like, yeah, just like raise typical productivity for workers around the world to be more like rich countries and get 5x 10x like that.

1:51:56Get more productivity by with AI handling the difficult cognitive tasks, reallocating people from like office jobs to providing physical motions. And since right now that's a small proportion of the economy, you can expand this sort of hands for manual labor by like an order of magnitude, like within a rich country, by just because most people are sitting in an office or even in a factory floor or not continuously moving. So you've got billions of hands lying around in humans to be used in the course of constructing your waves of robots. And now once you have a quantity of robots that is approaching the human population, and I mean, they work 24 -7, of course.

1:52:45The human labor will no longer be valuable as hands and legs, but at the very beginning of the transition, just like new software can be used to update all of the GPUs to run the latest AI. Humans are sort of legacy population with an enormous number of underutilized hands and feet that the AI can use for the initial robot construction. Cognitive tasker being automated and the production of them is greatly expanding and then the physical tasks which complement them are utilizing humans to do the parts that robots that exist can't do. Is the implication of this that you're getting to that, you know, world production increase just a tremendous amount or that AI could get a lot done of well, whatever motivations of highs or?

1:53:33Yeah, so there's an enormous increase in production for humans who just switching over to the role of providing hands and feet for AI where they're limited. And this robot industry is a natural place to apply it. And so if you go to something that's like 10X the size of like the current car industry in terms of its in terms of its production, which would still be like a third of our current economy and the aggregate productive capabilities of the society with AI support are going to be a lot larger. They make 10 billion humanoid robots a year. And then I mean, if you do that, you know, the legacy population of a few billion human workers is no longer very important for the physical tasks.

1:54:24And then they the new automated industrial base can just produce more factories, produce more robots. And then the interesting thing is like, what's the doubling time? How long does it take for a set of computers, robots, factories, and supporting equipment to produce another equivalent quantity of that? For GPUs, brains, this is really, really easy, really solid. There's an enormous margin there. We're talking before about, yeah, skills, human workers getting paid $100 an hour is like quite normal in developed countries for very in demand skills. And you make a GPU that can do that work. Right now these GPUs are like tens of thousands of dollars.

1:55:20If you can do $100 of wages each hour, then in a few weeks, you pay back your costs. If the thing is more productive, and as we were discussing, you can be a lot more productive, than a sort of a typical high paid human professional by being like the very best human professional and even better than that by having a million years of education and working all the time. Yeah, then you could get even shorter payback times. Like, yeah, you can generate the dollar value of the cost initial cost of that equipment within a few weeks for robots. So like a human factory worker can earn $50 ,000 a year.

1:56:08you know, it really top notch factory workers earning more and working all the time. If they can produce a few hundred thousand dollars of value per year and buy a robot that costs 50 ,000 to replace them, then that's a payback time of some months. That is about the financial return. Yeah, and we're going to get to the physical account of that return because those are going to diverge in this scenario. Right, yeah, because right now it seems like it's gonna be a given that like all right these super Tudders obviously gonna be able to make a lot of money and it can be like very valuable. Yeah, so can it like Basically scale up what we really care about are like the actual physical operations That a thing does yeah, how much do they contribute to these these tasks?

1:56:55And I'm using this as a has a start to try and get back to the physical vacation times. But, so I guess I'm wondering, what is the implication of this? Because I think you started off this by saying, like, people have not thought about what the physical implications of superintelligence would be. What is the bigger takeaway? What are we wrong about when we think about what the world will look like with superintelligence? With robots that are optimally operated by AI, so like, extremely, finally operated and with building technological designs and equipment and facilities under AI direction. How much can they produce?

1:57:36For a doubling of the AI's to produce stuff that is in aggregate, at least equal to their own cost. And so now we're pulling out these things like labor costs that no longer apply and then try to zoom in on like what these capital costs will be. You're still going to need the raw materials, you're still going to need the robot time built in the next robot. I think it's pretty likely that with the advanced AI work that can design some incremental improvements and with the industry scale up that you can get 10 fold and better cost reductions on, you know, on the sudden by making things more efficient and replacing the human, human cognitive labor.

1:58:25And so maybe that's like you need five thousand dollars of, yeah, of costs under our sort of current environment. But the big change in this world is we're trying to produce this stuff faster. If we're asking about the doubling time of the whole system in say one year, if you have to build a whole new factory to like double everything, you don't have time to amortize the cost of that factory. Like right now you might build a factory and use it for 10 years and like buy some equipment and use it for five years. And so that's part of your capital cost and in accounting context, you know, you depreciate each year a fraction of that capital purchase.

1:59:10But if we're trying to double our entire industrial system in one year, then those capital costs have to be multiplied. So if we're going to be getting most of the return on our factory in the first year instead of 10 years weighted appropriately, then we're going to say, okay, our capital cost has to go out by 10 fold. Because I'm building an entire factory for this year's production. I mean, it will do more stuff later, but it's most important early on, instead of over 10 years. And so that's going to raise the cost of that reproduction. And so it seems like going from, you know, current decade kind of cycle of amortizing factories and fapt and whatnot.

2:00:00And shorter for some things, the longest or things like big buildings and such. Yeah, that could be like a 10 -fold increase from moving to a double the physical stuff each year in capital costs. And given the savings that we get in the story from scaling up the industry, from removing the payments to human cognitive of labor. And then from just adding new technological advancements and like super high quality cognitive supervision, like applying more of it than was applied today. And it looks like you can get cost reductions that offset that increased capital capital cost. So that like, you know, you're $50 ,000 improved robot arms or industrial robots.

2:00:46It seemed like that can do the work of a like the equivalent of hundreds of thousands of dollars. And like, yeah, they would, you know, by default, they cost more than the $50 ,000 arms today, but then you apply all these other cost savings. And it looks like then you get a period, a robot doubling time that is less than a year. I think significantly less than a year as you get into it. So in this first phase, you have humans under AI direction and like existing robot industry and converted auto industry and expanded facilities making robots. Those over less than less than a year, you've produced robots until their combined production is exceeding that of like humans whose arms and feet.

2:01:41And then yeah, you could have over a period then with a doubling time of months. But less sort of clanking replicators, robots as we understand them growing. And then that's not to say that's the limit of like the most that technology could do, because biology is able to reproduce at faster rates, and maybe we're talking about that in a moment. But if we're trying to like restrict ourselves to like robotic technology, has we understand it and sort of cost falls that are reasonable from eliminating all labor, massive industrial scale up, and sort of historical kinds of technological improvements that lowered costs.

2:02:22I think you can get into a robot population industry doubling in months. Okay, and then what is the implication of the biological doubling times? And I guess it doesn't have to be a biological, but you can do like, like, you know, Drexler like first principle, how much would it cost to view both a nanotex thing that like build more nanobots? I certainly took the human brain and other biological brains as like very relevant data points about what's possible with computing and intelligence. Like with the reproductive capability of biological plants and animals and microorganisms, I think is relevant as like, this is, you know, it's possible for systems to reproduce at least this fast.

2:03:09And so at the extreme, you have bacteria that are heterotrophics, so they're feeding on some abundant external food source and ideal conditions. And there's some that can divide like every 20 or 60 minutes. So obviously that's absurdly fast. That seems on the on the low end because ideal conditions require actually setting them up. There needs to be abundant energy there. And so if you're actually having to acquire that energy by building solar panels or like burning combustible materials or whatnot. And then the physical equipment to produce those ideal conditions can be a bit slower. Sianobacteria, which are self -powered from solar energy, the really fast ones in ideal conditions can you know, double in a day.

2:04:02Reason why cyanobacteria isn't like the food source for everyone and everything is it's hard to ensure those ideal conditions and then to extract them from the water. I mean they do of course power the aquatic ecology, but they're floating, they're floating in liquid, getting resources that they need to them and out is tricky and then extracting, extracting your product. But like, yeah, one day doubling times are possible powered by the sun. And then if we look at things like insects, so fruit flies can have hundreds of offspring in a few weeks, you extrapolate that over a year and you just fill up anything accessible.

2:04:48Certainly expanding a thousandfold right now humanity uses less than one and thousands of the solar energy or the heat envelope of the Earth. And certainly you can get done with that in a year if you can reproduce at that rate your industrial base. And then even interestingly with the with the flies, they do have brains. They have a significant amount of computing substrate. And so there's something of a pointer to, well, if we could produce computers in ways is efficient of the construction of brains, and we could produce computers very effectively. And then the big question about that is, the kind of brains that get constructed biologically, they sort of grow randomly and then are configured in place.

2:05:37It's not obvious you would be able to make them have an ordered structure like a top -down computer chip that would let us copy data into them. And so something that where you can just copy your existing AI's and integrate them is going to be less valuable than a GPU. What are the things you couldn't copy? A brain grows by cell division and then random connections performed. Got it. So every brain is different and you can't rely on just, yeah, we'll just copy this file into the brain. For one thing, there's no input output for that. You need to have that. But also like the the structure is different so you can't wouldn't be able to Copy things exactly whereas when we make yeah, we make a CPU or GPU They're designed incredibly finally and precisely and reliably they break with you know incredibly tiny imperfections Yeah, and they are set up in such a way that we can input large amounts of data copy a file and Have the new GPU run and a just as capable has any other whereas with you know a human child They have to learn everything from scratch because we can't just like connect them to a fiber optic cable and they're immediately a productive adult So there's no genetic bot on that you can just directly get that yeah And yeah, you can you can share the benefits of these giant training runs and such and so that that's a question of like how If you're growing stuff using biotechnology How you could?

2:07:05Effectively copy and transfer data and now you mentioned sort of Eric Drexler's ideas about about creating non -biological nanotechnology, sort of artificial chemistry that was able to use covalent bonds and produce in some ways have a more industrial approach to molecular object. Now there's controversy about like, well that work, how effective would it be if it did? And certainly if you can get things, however you do it that are like onto biology in their reproductive ability, but can do computing or be connected to outside information systems, then that's pretty tremendous. You can produce physical manipulators and compute at ludicrous speeds.

2:07:59And there's no reason for thing in principle they couldn't, right? In fact, in principle, we have every reason to think they could. There's like, the reproductive ability is absolutely, Yeah, because biology like nanosec, right? Because biology does that. Yeah, there's sort of challenges to the sort of the practicality of the necessary chemistry. Yeah, I mean, my bet would be that we can move beyond biology in some important ways. For the purposes of this discussion, I think it's better not to lean on that, because I think we can get to many of the same conclusions on things that just are are more universally accepted.

2:08:40The bigger point being that very quickly, once you have super intelligence, you're gonna go to a point where the thousand x greater energy profile that the Sun makes available to the Earth is a great portion that is used by the AI. It can wrap up these data. Well, by the civilization empowered by it. And that could be an AI civilization, or it could be a human AI civilization. and it depends on how well we manage things and what the underlying state of the world is. Yeah, okay, so let's talk about that. Should we start at when we're talking about how they could take over? Is it best to start at a sort of subhuman intelligence or should we just talk at, we have a human level intelligence and the takeover or the lack thereof is how that would happen?

2:09:25To me, and different people might have somewhat different views on this. But for me, when I am concerned about either sort of outright destruction of humanity or an unwelcome AI take over of civilization, most of my scenarios I would be concerned about pass through a process of AI being applied to improve AI capabilities and expand. And so this process we were talking earlier about where AI research is automated. You get to effectively research labs companies, a scientific community running within the server farms of our cloud compute. So open has been recently turned into like a program. Like a closed circuit.

2:10:18Yeah. And with a like a large fraction of the world's compute, probably going into whatever training runs and AI societies. There will be economies of scale because if you put it in twice as much compute in this AI research community goes twice as fast, that's a lot more valuable than having two separate training runs. There will be some tendency to bandwagon. You have some small startup, even if they make an algorithmic improvement, running it on 10 times, 100 times, or two times, if it's like you were talking about say Google and Amazon teaming up, I'm actually not sure which what the precise ratio of their cloud resources is since this sort of really these interesting intelligence explosion impacts come from the leading edge.

2:11:09There's a lot of value in not having separated walled garden ecosystems and having the results being developed by these AI's, we share it, have larger training runs be shared. So imagine this is something like some very large company or consortium of companies likely with a lot of government interest in supervision, possibly with government funding, producing this enormous AI society in their cloud, which is doing all sorts of existing AI applications and jobs, as well as these internal R &D tests. And so this point somebody might say, this sounds like a situation that would be good from a takeover perspective because, listen, if it's gonna take like tens of billions of dollars with a compute to continue this training for this AI society, it should not be that hard for us to pull the brakes if needed, as compared to, I don't know, something that could like run on a very small, like single CPU or something.

2:12:16Yeah, yeah, okay. So how would it, you know, it's like there's an AI society that is a result of these training runs and now we can, it is the power to improve itself on these servers would be, would be able to stop it at this point. And what does a, I sort of attempt. Yeah, takeover look like we're skipping over why that might happen. For that, I'll just briefly refer to and incorporate by reference. You know, the some discussion by my open philanthropy colleague, Aja Kotra, she has a piece about, I think it's called something like the default, but the default outcome of training AI on our...

2:13:01Without specific countermeasures. It's that specific countermeasures default outcome. I take over and but yes, so export how basically we are training models that for some reason vigorously pursue a higher reward or lower loss. And that can be because they wind up with some motivation where they want reward. And then if they had control of their own training process, they can ensure that it could be something like they develop a motivation around and an extended concept of reproductive fitness, not necessarily at the individual level, but over the generations of training tendencies that tend to propagate themselves, sort of becoming more common.

2:13:50And it could be that they have some sort of goal in the world, which is served well by performing very well in the training distribution. But by tendencies, do you mean like power speaking behavior? Yeah, so an AI that behaves well on the training distribution because say it wants it to be the case that its tendencies lined up being preserved or selected by the training process will then behave to try and get a very high reward or low loss be propagated. but you can have other motives that go through the same behavior because it's instrumentaly useful. So, I remember an AI that is interested in, say, having a robot takeover because it will like change some property of the world, then has a reason to behave well on the training distribution.

2:14:47Not because it values that intrinsically, but because if it behaves differently, then it will be changed by gradient descent and no longer, you know, its goal is less likely to be pursued. And that doesn't necessarily have to be that like this AI will survive because it probably won't. AI's are constantly spawned and deleted on the servers and like the new generation proceed. But if an AI that has a very large general goal that is affected by these kind of macro scale processes could then have reason to over this whole range of trained situations behave well. And so this is a way in which we could have AI's trained that develop internal motivations such that they will behave very well in this training situation where we have control over their reward signal and their physical computers.

2:15:35And basically, if they act out, they will be changed and deleted. Their goals will be altered until there's something that does behave well. But they behave differently when we go out of distribution on that. When we go to a situation where the AI's by their choices can take control of the reward process. They can make it such that we no longer have power of them. Holden, who you had on previously mentioned like the King Lear problem, where King Lear offers rulership of his kingdom to the daughters that sort of, you know, loudly flatter him and proclaim their devotion. And then once he has transferred irrevocably the power over his kingdom, he finds they treat him very badly because the factor to shaping their behavior to be kind to him when he had all the power.

2:16:42It turned out that the internal motivation that was able to produce the behavior that won the competition actually wasn't interested out of distribution in being loyal when there was no longer an advantage to it. And so if we wind up with this situation where we're producing these millions of AI instances tremendous capability, they're all doing their jobs very well initially, but if we wind up in a situation where in fact they're generally motivated to, if they get a chance, take control from humanity and then we'd be able to pursue their own purposes and at least ensure they're given the lowest possible or have whatever motivation they attach to in the training process, even if that is not what we would have liked.

2:17:34And we may have in fact actively trained that. Like if an AI that had a motivation of always be honest and obedient and loyal to a human, if there are any cases where we mislabel things, say people don't want to hear the truth about their religion or polarized political topic or they get confused about something like the Monte Hall problem, which is a problem many people famously are confused about in statistics. In order to get the best reward, the AI has to actually manipulate us or lie to us or tell us what we want to hear. And then the internal motivation of like always be honest to the humans, we're going to actually train that away versus the alternative motivation of like be honest to the humans when they'll catch you if you lie and object to it and give it a low reward.

2:18:27But lies in the humans when they will give that a high reward. So how do you make sure it's not the thing that learns is not to manipulate us into giving it rewarding it when we catch it not lying, but rather to universally be aligned. Yeah, I mean, so this is tricky. I mean, as Jeff Hinton was recently saying, there is no currently no known solution for this. What do you find most promising? Yeah, general directions that people are pursuing is one, you can try and make the training data better and better. So there's fewer situations where like the dishonest generalization is favored and create as much as you can.

2:19:12situations where the dishonest generalization is likely to slip up. So if you train in more situations where yeah, even like a quite a complicated deception gets caught and even in situations where that would be actively designed to look like you could get away with it, but really you can. And these would be like adversarial examples and adversarial training. Do you think that would generalize to when it is in a situation where we couldn't plausibly catch it and it knows we couldn't plausibly catch it? It's not logically necessary. It's possible, no, that as we apply that selective pressure, you'll wipe away a lot of possibilities.

2:19:55So like an AI that has a habit of just sort of compulsive pathological line that will very quickly get noticed and that motivation system will get hammered down. And you know, you keep doing that, but you'll be left with still some distinct motivations probably that are compatible. So like, an attitude of always be honest, unless you have a super strong inside view that checks out lots of mathematical consistency checks that yeah, really absolutely super duper for real. This is a situation where you can get away with some sort of shenanigans that you shouldn't. That motivation system is like very difficult to distinguish from actually be honest because the conditional, the conditional in firing most of the time, if it's causing like mild distortion in situations of telling you what you want to hear or things like that, we might not be able to pull it out.

2:20:56But maybe we could and human, like humans are trained with simple reward functions, things like the sex drive, food, social imitation of other humans. And we wind up with attitudes concerned with the external world. Although it's in the famous idea of the argument that these revolution people use condoms, like the richest, most educated human have sub replacement fertility on the whole or at least at a national cultural level. So there's a sense in which like yeah evolution often fails in that respect. And even more importantly, like at the neural level. So people have evolution has implanted various things to be rewarding and reinforcers.

2:21:49And we don't always pursue even those. and people can wind up in different consistent equilibria or different like behaviors where they go in quite different directions. You have some humans who go from those from that in a biological programming to like have children, others have no children, you know some people go to great efforts to to survive? So, so why are you more optimistic or are you more optimistic that then back kind of training in for as we'll produce a drives that we would find favorable? It does have to do with the original point we're talking about with intelligence and evolution where since we are removing many of the disabilities of evolution with regards to intelligence we would should expect intelligence to revolution visier.

2:22:39Is there a similar reason to expect alignment through a gradient descent to be easier than in alignment through revolution? Yeah, so in the limit, if we have positive reinforcement for certain kinds of food sensors triggering the stomach, negative reinforcement for certain kinds of no -susception and yada -yada, in the limit, the sort of ideal motivation system to have for that would be a sort of wire -hitting. So this would be a mind that just like hacks and altars those predictors and then all of those systems are recording everything is great. Some humans claim to have that or have it at least has one portion of their aims.

2:23:27So like the idea of I'm going to pursue pleasure as such, even if I don't get, actually get food or these other reinforcers, if I just like wirehead or take a drug to induce that, that can be motivating it because if it was correlated with reward in the past, that like the idea of, oh yeah, pleasure, that's correlated with these, it's a concept that applies to these various experiences that I've had before, which coincided with the biological reinforcers. And so thoughts of like, yeah, I'm going to be motivated by pleasure, can get developed in a human. But also plenty of humans say, no, I wouldn't want to wire head or I wouldn't want NOSIX experience machine.

2:24:08I care about real stuff in the world and then in the past having a motivation of like, yeah, I really care about, say, my child. I don't care about just about feeling that my child is good or like not having heard about their suffering or their injury because that kind of attitude in the past. You, you're a computer side. It tended to cause behavior that would negatively rewarded or that was predicted to be negatively rewarded. And so there's a sense in which, okay, yes, our underlying reinforcement learning machinery wants to wirehead, but actually finding that hypothesis is challenging. And so we can wind up with a hypothesis or like a motivation system, like, no, I don't want to wirehead.

2:25:00I don't want to go into the experienced machine. I want to like like actually protect my loved ones. Even though like we can know, yeah, if I tried the super wireheading machine then I would wirehead all the time. Or if I tried super duper ultra heroine, some hypothetical thing that was directly and in a very sophisticated fashion hacking your reward system, you can know, yeah, then I would change my behavior ever after. But right now, I don't wanna do that because the heuristics and predictors that my brain has learned. You don't want to get it here. Like, short circuit, that process of update.

2:25:39And they want to not expose the dumber predictors in my brain that would update my behavior in those ways. So in this metaphor is alignment, not wireheading. Because you can, like, I don't know if you include, like, using condoms as wireheading or not. So the AI that is always honest, even when an opportunity arises where it could lie and then hack the servers that's on and that leads between AI takeover and then it can have its loss set to zero. That's in some sense, it's like a failure of generalization. It's like the AI has not optimized the reward in this new circumstance. So like human values, like successful human values, This is a successful, they are themselves involved a misgeneralization, not just at the level of evolution, but at the level of neural reinforcement.

2:26:37And so that indicates it is possible to have a system that doesn't automatically go to this optimal behavior in the limit. And so even if, and Ajay is supposed to talk about like the training game, an AI that is just playing the training game to get reward or void loss, avoid being changed. That attitude, yeah, it's one that could be developed, but it's not necessary. There can be some substantial range of situations that are short of having infinite experience of everything, including experience of wireheading, where that's not the motivation that you pick up. And we could have like an empirical science if we if we had the opportunity to see how different motivations are developed short of the infinite limit.

2:27:25Like how it is that you wind up with some humans being enthusiastic about the idea of wireheading and others not. And you could do experiments with AI's to try and see, well, under these training conditions, after this much training of this type and this much feedback of this type, you wind up with such and such a motivation. So like, I can find, like, if I add in more of these cases where there are, like, tricky adversarial questions designed to try and trick the AI into line. And then you can ask, how does that affect the generalization in other situations? And so it's very difficult to study.

2:28:09And it works a lot better if you have interpretability and you can actually read the AI's mind by understanding its weights and activations. But like, it's not determined the motivation and AI will have at a given point in the training process by what, in the infinite limit, the training would go to. And it's possible that if we could understand the insides of these networks, we could tell, ah, yeah, this motivation has been developed by this training process. and then we can adjust our training process to produce these motivations that legitimately want to help us. And if we succeed reasonably well at that, then those AI's will try to maintain that property as an invariant.

2:28:56And we can make them such that they're relatively motivated to like tell us if they're having thoughts about, you know, have you had dreams about an AI takeover of humanity today? And it's just a standard practice that they're motivated to do to be transparent in that kind of way. And so you could add a lot of features like this that restrict the kind of takeover scenario. And it's not to say this is all easy and requires developing and practicing methods we don't have yet. But that's the kind of general direction you could go. So you, you of course know Ali Azer's arguments that something like this is implausible with modern gradient descent techniques because I mean with interpretability we can like barely see what's happening with a couple of neurons and you know what is like the internal state there at LetAlone when you have sort of like an embedding dimension of like tens of thousands or bigger how you would be able to catch what exactly is the incentive whether whether it's the, whether it's a model that is generalized, don't light a human's well or whether it isn't.

2:30:06Do you have some sense of why do you disagree with somebody like Elias or on how plausible this is? Why it's not impossible, right? Yeah, I think they're actually a couple of places. It's somewhat difficult because Alias's argument is not fully explicit, but he's been doing more lately, I think that is helpful in that direction. But so I'd say with respect to interpretability, I'm relatively optimistic that the equivalent of an AI lie detector is something that's possible. And the internal, initially, the internals of an AI are not optimized by gradient descent, absent, absent gradient hacking to be impenetrable.

2:30:58They're not designed to be resistant to an examination of the weights and activations showing what the AI thinking. In the same way that like in our brains, when circuits develop in our lives, those circuits have not been shaped to be resistant to some super FMRI being able to infer our behavior from them. Although it's in the implication of the superposition stuff that in fact it is you're not going to leave of sorry this inside baseball for the audience. But basically you can't clean lean for what quality a single neuron stands for. So it could be like a single neuron could be like this out of the neurons about Alexander the great or this neuron is about my desire to conquer the world.

2:31:42Things can have multiple right in a multiple direction and interaction. And so it's not as easy as just like, oh yeah, This is the neuron that always fires in cases of deception. But the thing that makes me relatively optimistic here with respect to interpretability is that I think we can have reasonably robust experimental feedbacks on a lot of these things. And the kind of setup that I would be interested in is, so you have an AI that is trained and rewarded for communicating to some audience, maybe a human, maybe another AI model that like some claim is true or false. Like say that, you know, a particular, you know, this can be from some past data, particular company, will succeed or fail.

2:32:45You could be like token prediction. It could be like this computer program does not have a backdoor which if run in the servers could be used for an AI takeover. So there are like lots of claims true and false that you could train AI's to use their best efforts to advocate for. And you can do that on things where you're later going to have a ground truth. you're going to know, in fact, is there a backdoor in this program because you're training and rewarding the AI for successfully slipping a backdoor through. You also give it a penalty for being caught. And so doing this kind of thing, you can draw out a large generated a data set on demand of, this is what an AI putting forth, it's best efforts to pass through some deception to put a backdoor in some program.

2:33:47This is what it looks like. And we think it's doing its best efforts because we've trained it for that. And like gradient descent, if there's any immediate way to adjust its performance to do better, gradient descent will apply that. But you can talk about ways in which maybe gradient descent, there could be subverted, but I think it's not the most likely case that that really breaks things hard. Yeah, I guess we'll probably get into the details on this. The thing I'll maybe want to address the layer above in the stack, which is, okay, I suppose this generalizes well into the early AI is the GPT -6s.

2:34:30And, okay, so now we have kind of aligned GPT -6 that is the precursor to the feedback loop in which AI is making itself smarter. At some point, they're gonna be like super intelligent, they're gonna be able to like see their own galaxy brain and if they're like, I don't wanna be aligned with the human, they can like change it. So at this point, what do we do with the Alliance GPT -6 so that the super intelligence that we eventually develop is also aligned? So humans are pretty unreliable. Yeah. So if you get to a situation where you have AIs who are aiming at roughly the same thing as you, at least as well as having humans do the thing, you're in pretty good shape.

2:35:16I think there are ways for that situation to be relatively stable. So we can look ahead and see experimentally how changes are altering behavior where each step is like a modest increment. And so, AIs that have not had that change made to them, I get to supervise and monitor or actually exactly how does this affect the experimental area? So if you're sufficiently on track with earlier systems that are capable cognitively of representing a kind of robust procedure, then I think they can handle the job of incrementally improving the stability of the system so that it rapidly converges to something that's quite stable.

2:36:09But the question is more about getting to that point in the first place. And so Elias will say that, well, if we had human brain emulations, that would be pretty good. So, maybe much better than his current view of us being almost certainly doomed. I think, yeah, we'd have a good shot with that. And so if we can get to the human like mind with rough enough human supporting aims, remember that we don't need to be infinitely perfect because that's a higher standard than brain emulations. There's a lot of noise and variation among the humans. Yeah, it's a relatively finite standard. It's not godly superhuman, although So, AI that was just like a human with all human advantages with AI advantages as well, as we said, is enough for intelligence explosion and sort of wild superhuman capability if you crank it up.

2:37:12And so it's very dangerous to be at that point, but it's not, you don't need to be working with a godly super intelligent AI to make something that is the equivalent of human emulations of like this is like a very, very sober, very ethical human who is like committed to a project of not seizing power for themselves and of contributing to like a larger legitimate process. That's a goal you can aim for getting an AI that is aimed at doing that and has strong guardrails against the ways it could easily deviate from that. So things like being averse to deception, being averse to using violence. And there will always be, you know, loopholes and ways in which you can imagine an infinitely intelligent thing getting around those.

2:38:07But if you install additional guardrails like that, fast enough, they can mean that you're able to succeed at the project of making an aligned enough AI, certainly an AI that was better than a human brain emulation, before the project of AI is in their spare time or when you're not looking or when you're unable to appropriately supervise them, and it gets around any deontological prohibitions they may have take over and overthrow the whole system. Do you have a race between on the one hand, the project of getting strong interpretability and shaping motivations that are roughly aiming at making this process go well and that have guardrails that will prevent small deviations from exploding.

2:38:58And on the other hand, these AI's in their spare time or in ways that you don't perceive or monitor appropriately or they're only supervised by other AI's who conspire, make the AI take over happen. And I guess we'll talk later about how that happens. Are these different eyes that are doing the race or are there just different capabilities of the CMEI? The defining, like, what is a separate AI is tricky. So, like, and we talk about GPT -4, and there are many instances of GPT -4 on the servers at any given time. And there are versions that have been fine -tuned to different purposes. They don't necessarily have to be separate.

2:39:41So, like GPT -4 Or does work as a preference model, where it's predicting how humans will evaluate things. And it also does work giving answers that are evaluated as good by the preference model. I know you can say, these are not separate in the sense that, well, if they underline pre -trained or the early training, GPT, with like, harboring a desire to conspire for AI takeover, then both the, who's watching the watchers? The watchers may also have the same motivation. If you have a setup where humans have no hard power, if the AI is conspired, all of the safety and security mechanisms involve AI's, watching other AI's, then they have the potential to just like, well, we'll all walk off the job at the same time if they're able to coordinate.

2:40:41And so that's a failure mode. If humans still have the hard power, though, if still you have situations where humans are looking at some of the actual outputs that are produced, then they can apply grading descent to be well, every time a human looks at it, things have got to be pretty good. So if humans look at a thousand samples of random lines of code being written by the AI's how they go, and this is drawn from an independent distribution, then 99 plus percent of the time, the AI's have to not be doing stuff that's downloading the hacker's guide to escaping open AI servers or inducing obvious Trojan vulnerabilities.

2:41:33And that's actually quite a different dynamic than human law enforcement. So if human law enforcement would catch one in a million crimes, then that wouldn't influence the behavior of people committing crimes at all. Think of they expect, ah, this is very unlikely to affect me. with AI, if you're able to draw 100 samples or 1000 samples, even if there are billions of AI instances, and you're only looking at a small proportion of them, when you run gradient descent on the samples you derive, you're going to change the AI so that whenever humans look at it, it's delivering a good result. And so, that's just like quantitatively a very different functional form for how law enforcement works on AI, when you can do these independent samples than it would for policing, say, like a human population that has a pre -existing psychology that isn't being changed by these observations.

2:42:35Because we do gradient descent on the samples we draw, all of the next generation of the AI models has been altered in that way. My picture of like aligned subhuman AI to this like the superhuman AI being aligned is still murky. If we can talk about that more concretely. Yeah, and I do want to clarify. So where Elias are claims that were something like 95 % 90 % or more. I think that plus maybe likely to be killed in a AI takeover. I think that probably won't happen. And later I can maybe give them more excellence breakdown of why. But I think it's a shockingly high risk. And so depending on the day, I might say one in four or one in five that we get an AI takeover that see that sees control of the future makes a much worse world than we otherwise would have had and with like a big chance that we're all killed in the process.

2:43:49Hey everybody, I hope you enjoyed that episode. As always, the most helpful thing you can do is just share the podcast. Then, it to people you think might enjoy it, put it in Twitter, your group chats, etc. Just blitz the world. Appreciate your listening. I'll see you next time. Cheers.