Dwarkesh Patel and Noah Smith on AGI and the Economy

In this episode, Erik Torenberg is joined in the studio by @DwarkeshPatel and Noah Smith to explore one of the biggest questions in tech: what exactly is artificial general intelligence (AGI), and how close are we to achieving it? They break down: - Competing definitions of AGI - economic vs. cognitive vs. “godlike” - Why reasoning alone isn’t enough - and what capabilities models still lack - The debate over substitution vs. complementarity between AI and human labor - What an AI-saturated economy might look like - from growth projections to UBI, sovereign wealth funds, and galaxy-colonizing robots - How AGI could reshape global power, geopolitics, and the future of work Along the way, they tackle failed predictions, surprising AI limitations, and the philosophical and economic consequences of building machines that think—and perhaps one day, act—like us. Timecodes: 0:00 Intro 0:33 Defining AGI and General Intelligence 2:38 Human and AI Capabilities Compared 7:00 AI Replacing Jobs and Shifting Employment 15:00 Economic Growth Trajectories After AGI 17:17 Consumer Demand in an AI-Driven Economy 31:14 Redistribution, UBI, and the Future of Income 31:58 Human Roles and the Evolving Meaning of Work 41:21 Technology, Society, and the Human Future 45:43 AGI Timelines and Forecasting Horizons 54:04 The Challenge of Predicting AI's Path 57:37 Nationalization and the Global AI Race 1:07:10 Brand and Network Effects in AI Dominance 1:09:31 Final Thoughts and Preparation for What’s Next Resources: Find Dwarkesh on X: https://x.com/dwarkesh_sp Find Dwarkesh on YT: https://www.youtube.com/c/DwarkeshPatel Subscribe to Dwarkesh’s Substack: https://www.dwarkesh.com/ Find Noah on X: https://x.com/noahpinion Subscribe to Noah’s Substack: https://www.noahpinion.blog/ Stay Updated: Let us know what you think: https://ratethispodcast.com/a16z Find a16z on Twitter: https://twitter.com/a16z Find a16z on LinkedIn: https://www.linkedin.com/company/a16z Subscribe on your favorite podcast app: https://a16z.simplecast.com/ Follow our host: https://x.com/eriktorenberg Please note that the content here is for informational purposes only; should NOT be taken as legal, business, tax, or investment advice or be used to evaluate any investment or security; and is not directed at any investors or potential investors in any a16z fund. a16z and its affiliates may maintain investments in the companies discussed. For more details, please see a16z.com/disclosures.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  AI might be generating hundreds of dollars of value for me a month, but like humans are generating thousands of dollars or tens of thousands of dollars of value for me a month. Why is that the case? And I think it's just like AIs are lacking these capabilities, humans have these capabilities.

• 0:11Noah Smith?

  You are a natural general intelligence, but we cannot easily do each other's jobs, even though our jobs are fairly similar.

• 0:17Speaker 00

  The reason humans are so valuable is not just their raw intellect. It's their ability to build up context, it's to interrogate their own failures and pick up small efficiencies and improvements as they practice a task. Whereas within AI model, it's understanding of your problem or business will be expunged by the end of a session.

• 0:36Noah Smith?

  Every other technological tool is a compliment to humans. And yet, when people talk about AI and think about AI, they essentially never seem to think in these terms. They always seem to think in terms of perfect substitutability.

• 0:46Erik Torenberg?

  What happens when AI can do almost every white collar job, but still can't remember what you told it yesterday? What does that mean for AGI, the future of work and the shape of the global economy? I sat down with Noah Smith, author of No Opinion, and Dorkesh Patel, host of the Dorkesh Podcast, to unpack what's real and what's hype in the race against AGI. We talk about continual learning, economic substitution, galaxy scale growth, and whether humanity's biggest challenge is technological or political. Let's get into it.

• 1:17Erik Torenberg?

  As a reminder, the content here is for informational purposes only. Should not be taken as legal, business, tax, or investment advice, or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any A16Z fund. Please note that A16Z and its affiliates may also maintain investments in the companies discussed in this podcast. For more details, including a link to our investments, please see a16z.com/disclosures.

• 1:47Erik Torenberg?

  Dorkesh, Noah, welcome. Our first podcast ever as a trio.

• 1:51Speaker 00

  Yes, excited. I'm very excited.

• 1:53Erik Torenberg?

  So, Dorkesh, you came up with the scaling era. It's almost like you're a future historian. You're sort of telling the history as it's being written. And so it's only appropriate to ask you, what is your definition of AGI? And how has that evolved over time?

• 2:05Speaker 00

  I feel like I'm like five decades too young to be a historian. It's gotta be like in your 80s or something before I could...

• 2:09Erik Torenberg?

  But we're living in history right now.

• 2:10Speaker 00

  Right. So, the ultimate definition is can do almost any jobs, like 98% of jobs at least, as well, fast, cheaply as a human. I think the definition that's often useful for near-term debates is can automate 95% of white collar work. Because there's a clear path to get to that. Whereas robotics, there's like a long tail of things you had to do in the physical world and robotics is slower. So automated white collar work.

• 2:33Noah Smith?

  Hmm. That's interesting because it's an economic definition. It's not a definition about like how it thinks, how it reasons, et cetera. It's about what it can do.

• 2:40Speaker 00

  Yeah, I mean, we've been surprised what capabilities have come first in AIs. Like they can reason already and why they seem to lack the economic value we would have assumed would correspond to that level of capability. This thing can reason, but it's making OpenAI $10 billion a year. And McDonald's and Kohl's make more than $10 billion a year, right? So clearly there's more things relevant to automating entire jobs than we previously assumed. So then it's just useful to say like, who knows what all those things are, but once they can automate it, then it's AGI.

• 3:11Erik Torenberg?

  And so when Ilya or Meta is using the word super intelligence, what do they mean? Do they mean the same thing or something totally different?

• 3:17Speaker 00

  I'm not sure what they mean. There's a spectrum between God and just something that thinks like a human, but much faster. Do you have some sense of what you think they mean?

• 3:28Noah Smith?

  God. I think probably they mean something they would worship as a god.

• 3:31Speaker 00

  Yeah.

• 3:32Erik Torenberg?

  And so when Tyler says we've achieved AGI and you differ from him, where is the tangible difference there?

• 3:37Speaker 00

  I'm just noticing that if there was a human who was working for me, they could do things for me that these models cannot do, right? And I'm not talking about something super advanced. I'm just saying I have transcripts from my podcast. I want you to rewrite them the way a human would, and then I'll give you feedback about what you messed up. And I want you to integrate that feedback as you get better over time. You learn my preferences, you learn my content. And they can't learn over the course of six months, how to become a better editor for me or how to become a better transcriber for me. And since a human I hire would be able to do this, they can't. So therefore it's not AGI.

• 4:06Noah Smith?

  Now I have a question. I am a natural general intelligence. You are a natural general intelligence, but we cannot easily do each other's jobs, even though our jobs are fairly similar.

• 4:15Speaker 00

  Right.

• 4:15Noah Smith?

  Put me in the Dwarkesh podcast and I could not interview people nearly so well. If you had to write Substack articles like several times a week.

• 4:22Speaker 00

  Yes.

• 4:22Noah Smith?

  On economics, you might not do as well. But we are general intelligences and we're not exactly substitutable. So why should we use substitutability as the criterion for AGI?

• 4:31Speaker 00

  What else is it that we want them to do? I think with humans, we have more of a sense of there is some other human who theoretically could do what you do. An individual copy of a model might be, say, fine tuned to do a particular job. And it would be fair to say then, why expect this particular fine tune to be able to do any job in the economy? But then there's a question of, well, there's many different models in the world and each model might have many different fine tunes or many different instances. Any one of them should be able to do a particular white collar job for it to count as AGI. It's not that like any AGI should be able to do every single job, that like some artificial intelligence should be able to do this job for this model to count as AGI.

• 5:06Noah Smith?

  I see, okay, but so let's take another similar example, let's take Star Trek.

• 5:09Speaker 00

  Yeah.

• 5:10Noah Smith?

  Okay, you got Spock, he's very logical, he can do stuff that Kirk and whoever can't do. But then those guys can do stuff that Spock can't do.

• 5:16Speaker 00

  Right.

• 5:16Noah Smith?

  Get in touch with their emotions, intuition, stuff like that. They're both general intelligences, but they're alien to each other. So AI feels alien to me.

• 5:23Speaker 00

  Mm-hmm.

• 5:23Noah Smith?

  Sometimes it talks just like us, it was built off of our thoughts, obviously, but then sometimes it talks just like us and sometimes it's just like very alien. And so should we ever expect that to change such that it's no longer an alien intelligence?

• 5:36Speaker 00

  I think it'll continue to be alien, but I think eventually we will gain capabilities which are necessary to unlock the trillions of dollars of economic value that are implied by automating human labor, which these models are clearly not generating right now. So you could say like if we substituted jobs right now, immediately there'd be a huge productivity dip. But over time, we would learn to start doing them better. I mean, maybe a better example is just that like you hire people to do things for you. I don't know if you actually hire people, but I assume.

• 6:05Noah Smith?

  Okay.

• 6:05Speaker 00

  Yeah, yeah. Okay, why are you still having to do that rather than hiring an AI? And I have like many rules where it's like an AI might be generating hundreds of dollars of value for me a month, but like humans are generating thousands of dollars or tens of thousands of dollars of value for me a month. Why is that the case? And I think it's just like AIs are lacking these capabilities, humans have these capabilities.

• 6:21Noah Smith?

  And is the main thing missing in your view, sort of continual learning? What is the bottleneck?

• 6:26Speaker 00

  The reason humans are so valuable is not just their raw intellect. It's not mainly their raw intellect, although that's important. It's their ability to build up context, it's to interrogate their own failures and pick up small efficiencies and improvements as they practice a task. Whereas with an AI model, it's understanding of your problem, your business, will be expunged by the end of a session. And then you're starting off at the baseline of the model. And with a human, you gotta train them over many months to make them useful employees.

• 6:53Noah Smith?

  Yeah. And what will need to change in order for AI to develop a capability?

• 6:57Speaker 00

  Hmm, I mean, I probably wouldn't be a podcaster if I knew the answer to that question.

• 7:03Speaker 00

  It just seems to me that like a lot of the modalities we have today to teach LLM stuff do not constitute this kind of continual learning. For example, making the system prompt better is not the kind of continual learning or on the job training that my human employees experience, or RL fine tuning is not this. But like what the solution to this looks like, it's precisely because I don't have an obvious solution that I think we're many years away.

• 7:25Noah Smith?

  Okay, so here's my question about replacing jobs. It seems to me that it's partly by demand. So for example, suppose that AI has already replaced my job or can replace my job. So that suppose that anyone who fires up ChatGPT or whatever model and says, "Search the web, find the most interesting topics that people are talking about economics and write me an insightful post telling me some cool new thing I should think about that." And they just do that every day and then they get a better blog than no opinion. I don't know if that's happened yet. I mean, I've tried that and I don't like it as much. But suppose that most people will like it as much and so my job's been automated and people just don't realize it. Or people have this sort of idea in their mind of like, "Well, is it really a human?" And blah, blah. And then as generational turnover happens, young people won't care about reading a human. They'll care about reading an AI. But in terms of functional capabilities, it's already there. But in terms of demand, it's not there.

• 8:10Noah Smith?

  How much of that could there be?

• 8:12Speaker 00

  I expect there will be much less of that than people assume. If you just look at the example of Waymo versus Uber. I think previously you could have had this thing about people, "We hesitate to take automated rides." And in fact, in the cities where it's been deployed, people love this product despite the fact that you had to wait 20 minutes because the demand is so high. And it's still like it got some glitches to iron out. But just the seamlessness of using machines to do things for you, the fact that it can be like personalized to you, it can happen immediately. One thing people will be like, "Okay, well, doctors and lawyers will set up guilds and so you won't be able to consult." I think there might be guilds and who can call themselves a doctor or a lawyer. But I just think if like, if genuinely ChatGPT can give me as good medical advice as a real doctor, the experience of just talking to a chatbot rather than spending three hours in a waiting

• 8:55Speaker 00

  room is so much better that I think a lot of sectors in the economy look like this where we're like, we're assuming people will care about having a human, but in fact, they will not. If you assume that they will genuinely have the capabilities that the human brings to bear.

• 9:05Noah Smith?

  Right. So it's interesting, AI is better for diagnosis on a lot of things than humans. Right. But then something about having humans to follow up with makes me also want to check with a human after I've gotten diagnoses from an AI on something. And so that might vary by job, like cars may be one thing, but maybe it is about capabilities. I can't say, I'm just saying like, everybody seems to think that AI is a perfect substitute for humans and that's what it should be and that's what it will be. And everyone seems to think of it in that case, however, every other tool that's ever been made, every other technological tool was a compliment to humans. It could do some things humans could do, maybe even it could do anything humans could do, but at different relative costs, different relative prices. So that you'd have humans do something and the tool do other things and you'd have this complementarity between the two.

• 9:50Noah Smith?

  And yet, when people talk about AI and think about AI, they essentially never seem to think in these terms. They always seem to think in terms of perfect substitutability. And so I'm trying to get to the bottom of like, why people insist on always thinking in terms of perfect substitutability when every other tool has been complimentary in the end.

• 10:04Speaker 00

  Well, human labor is also complimentary to other human labor, right? There's increasing returns to scale. But that doesn't mean that Microsoft has to hire some number of software engineers. And like, it will care about the cost of what the software engineers cost. Like, it will go to markets where they can get the highest performance for the relative value those software engineers are bringing in. I think it will be a similar story with AI labor and human labor. And AI labor just has the benefit of having extremely low subsistence wages. Like, the marginal cost of keeping an H100 running is much lower than the cost of keeping a human alive for a year.

• 10:36Erik Torenberg?

  Noah, would you say you're AGI-pilled in the sense that Dorcas described the term? We've talked a little bit about AI's effect on labor when you shared why you're perhaps a little bullish that there'll be a plenty for humans to do and that'll be more complimentary?

• 10:48Noah Smith?

  What is AGI-pilled?

• 10:49Erik Torenberg?

  We just believe in that it will automate a huge swath of the economy or labor?

• 10:54Noah Smith?

  I mean, I am very unwilling to say like, here's something technology will never be able to do. I mean, that always seems like a bad bet. Here's two things people have been saying since the beginning of the industrial revolution, neither of which has ever remotely come close to being true, even in specific subdomains. The first one is, here's a thing technology will never be able to do. And the second one is, human labor will be made obsolete. Those, people have been saying those two things and you can just go, you can read it, you can even ask AI to go search and find, I have done this, and then find you examples of people saying those two things. People have been saying those two things over and over and over and over and over and it's never been true. That doesn't mean it could never be true. Sometimes something happens that never happened before, such as the Industrial Revolution itself.

• 11:37Noah Smith?

  You have this hockey stick where suddenly like, oh, we'll never get rich, we'll never get rich, oh, we're rich. And so sometimes that happens. The unprecedented can happen, however, I'm always wary because I've seen it said so many times. And so, within just the last 10 years or whatever, I've seen a couple predictions just spectacularly fail. So for example, in 2015, 10 years ago, I was sitting in the Bloomberg office in New York and my colleague, I won't name, he was physically yelling at me that truck drivers were in trouble and that truck drivers were all going to be put out of a job by self-driving trucks. And he said, this is gonna just devastate a sector of the economy, it's gonna devastate the working class, it's gonna devastate blue collar labor, blah, blah. And at the same time, I was reading like, I always read the sci-fi top stories of the year, whatever.

• 12:17Noah Smith?

  And so, there were two stories in the same year about truckers being mass unemployed by self-driving trucks. And then, 10 years later, there's a trucker shortage and the number of truckers we hire is higher than ever. I'm not saying truckers will never be automated, they may. However, I'm saying that was a spectacularly wrong prediction and you also got Jeffrey Hind's prediction that radiologists would be unemployed within a certain timeframe and by that time, radiologist wages were higher than ever and employment was higher than ever. I'm not saying this can't happen, I'm not smugly sitting here and saying there's a law of the universe that says you'll never see this kind of mass unemployment, blah, blah, blah. I mean, there were encyclopedia salespeople were mass unemployed by the internet, we've seen it happen in real life. But these predictions keep coming wrong and keep coming wrong and I'm trying to figure out, why is that true? Why do they keep coming wrong? Is it simply that people overestimate progress in technical capabilities?

• 13:00Noah Smith?

  Or are there complementarities that people can't imagine from sort of like the ONET division of tasks or the standard mental division of tasks?

• 13:08Speaker 00

  I think the problem has been that people underestimate how many things are truly needed to automate human labor. And so, they think like, we've got reasoning. And now that we've got reasoning, like, this is what it takes to take over a job. And I think, in fact, there's much more to a job than is assumed. That's why I wrote this blog post where I'm like, no, it's not a couple of years away, it might be longer than that. Then there's another question of like, by 2100... Will there be jobs that humans are doing? If you just like zoom out long enough, will we ever be able to make machines that can think and do physical labor at least as cheaply and as well as humans can? And fundamentally, the big advantage they have is like, we can keep building more of them. Right? So we make as many of those machines as the value they generate equals the cost of producing them.

• 13:53Erik Torenberg?

  And the cost will continue to go down.

• 13:54Speaker 00

  Right, yeah. And it will be lower than the cost of keeping a human alive. So even if a human could do the exact same labor, a human needs like a lot of stuff to stay alive, let alone to grow a human, everything. An H100 costs $40,000 today. The yearly cost of running it is like thousands of dollars. We can just buy more H100s. Right, like if currently we had the algorithm for AGI, we could run it on an H100 and yeah.

• 14:16Erik Torenberg?

  So however big the demand is, the latent demand

• 14:19Speaker 00

  Right.

• 14:19Erik Torenberg?

  that's unlocked by the more supply, we just increase the supply basically to meet that demand. So first, when AGI is here, what does the world look like? Because Sam Altman was reflecting on his podcast with Jack Altman the other week. He was saying, if you told me 10 years ago that we would have PhD level AI, I would think the world looks a lot different. But in fact, it doesn't look that different.

• 14:38Speaker 00

  Yeah.

• 14:38Erik Torenberg?

  And so is there a potential where we have much more increased capabilities, but actually the world does, it's like the Peter Thiel called the 1973 test or something. It's we have these phones, but the world just looks the same. We just have phones in our pockets.

• 14:50Speaker 00

  Yeah, I think if we have like chatbots that can answer hard math questions, I don't expect the world to look that different because the fraction of economic value that is generated by math is like extremely small. But there's like other jobs that are much more mundane than quote unquote PhD intelligence, which a chatbot just cannot do, right? A chatbot cannot edit videos for me. And once those are automated, I actually expect a pretty crazy world. Because the big bottleneck to growth has been that human population can only increase at this slow clip. And in fact, one of the reasons that growth has slowed since the seventies is that in developing countries, the population has plateaued. With AI, the capital and the labor are functionally equivalent, right? You can just build more data centers or build more robot factories.

• 15:35Speaker 00

  And they can do real work or they can build more robot factories. And so you can have this explosive dynamic. And once we get like that loop closed, I think it would just be like 20% growth plus. Do you say feasible, possible 20% growth? Tyler, I believe, said 5%. 0.5% more than the steady state. 0.5%.

• 15:51Noah Smith?

  0.5%.

• 15:52Speaker 00

  What is the argument for that? For Tyler's argument? Bottlenecks. I think the problem with that argument is that there's always bottlenecks, right? So you could have said before the industrial revolution, well, we will never 10x the rate of growth because there will be bottlenecks. And that doesn't tell you what, like, you empirically have to just look at the fraction of the economy that will be bottlenecked, and what is the fraction that's not, and then like actually derive the rate of growth. The fact that there's bottlenecks doesn't tell you, yeah, okay. Is he basically referring to the regulation or? Yeah, and just that, like, we live in a fallen world and people will have to use the AIs and yeah, things like that.

• 16:22Noah Smith?

  Who'll be buying all the stuff? So, background, in economics, GDP is what people are willing to pay for.

• 16:28Speaker 00

  Right.

• 16:28Noah Smith?

  Who will be buying the stuff in a world where we get 20% growth?

• 16:31Speaker 00

  First of all, I don't know. So you could have said in 10,000 B.C., like, the economy is going to be a billion times bigger in 10,000 years. What does it mean to produce a billion times more stuff than we're producing right now? Who is buying all this stuff? You can't like predict that in advance.

• 16:44Noah Smith?

  In 1700s, I could tell you exactly who's buying stuff. It was everybody, peasants. In fact, people wrote these things around 1900 about what the world would look like in 100 years. You know, what we'll have. They didn't get exactly the right things right that we'll have, but they correctly identified that it would be regular consumers who would be buying all these things, regular people. And so that would, that came true. It was obvious. But here's my point. Suppose that 99% of people do not have a job and are not getting paid an income, and all the money is going to sort of Sam Altman, Elon Musk, and five other guys. Okay? And they're captive AIs that they own because for some reason our property rights system still exists.

• 17:19Speaker 00

  Right.

• 17:19Noah Smith?

  But okay, suppose that that's the future we're contemplating, right? And so 99% of people or more don't have any job, they don't have any income, they're out on the street. And yet, you're saying 20% growth a year. That growth is defined by people, consumers, paying for things and saying, here is my money.

• 17:36Speaker 00

  Justice people.

• 17:37Noah Smith?

  Okay, so then.

• 17:38Speaker 00

  I would just define it as like, the, I mean, I.

• 17:40Noah Smith?

  We will have AI purchase on agents.

• 17:42Speaker 00

  Yeah.

• 17:42Noah Smith?

  No, no, that doesn't count GDP. Only final good, only final.

• 17:45Speaker 00

  Okay, so we're like launching the Dyson spheres, we're not allowed to count that because the AIs are doing it. I mean, like, I want to know what the solar system will look like. I don't care like what like the semantics of that are. And I think the better way to capture what is physically happening is just.

• 17:56Noah Smith?

  Why will they do that? Why will they do that? Why will they do any of that?

• 17:59Speaker 00

  One argument is simply that if there's any agent, AI or human, who cares about colonizing the galaxy? Even if 99% of agents don't care about that, if one agent cares, they can go do it. Colonizing the galaxy is a lot of growth because the galaxy is really big, right? So it's very easy for me to imagine if like Sam Altman decides to launch the probes, how like, you know, breaking down Mars and sending out the virus probes like generates 20% growth.

• 18:21Noah Smith?

  I think what you're getting at here is that AI will have to have property rights. AI agents will have to be able to have autonomous control of resources.

• 18:29Speaker 00

  I guess it depends what you mean by autonomous. Today, we already have computer programs that have autonomous use of resources.

• 18:35Noah Smith?

  Okay, but the program goes off and colonizes the solar system. It's not like a dude telling it, colonize the solar system now and doing all this stuff. It's like the AI has made the decision to do it and Sam Altman is sitting back there saying, oh, well.

• 18:46Speaker 00

  I'm just saying this is not a crux. Sam Altman could say it or the AI could say it. If some agent cares about this and they're not stopped from doing it, like this is just like, physically, you can easily see where the 20% growth is coming from.

• 18:56Noah Smith?

  Let me make this a little more concrete. Suppose that AI is going to produce a bounty of the things that humans desire, and that's going to be what growth is. How will it get to the humans if the humans don't have a job? And if the humans don't have a job, why will AI be? So in other words, if there's no consumers to buy my cars, why am I building cars?

• 19:15Speaker 00

  You might be assuming there's some UBI or some sort of. No, no, I don't need to assume that, although, I mean.

• 19:19Noah Smith?

  Let's assume there's not that.

• 19:20Speaker 00

  Yes, I don't need to assume that. It seems like you're saying, look, if 99% of consumers are no longer consumers, where's this economic activity coming from?

• 19:27Noah Smith?

  Yeah.

• 19:28Speaker 00

  And I'm just saying, okay, if one person cares about colonizing the galaxy, that's generating a lot of demand. It takes a lot of stuff to colonize a galaxy. So this world where like, even if there's not a new galaxy in the world where everybody is like roughly contributing equivalent amounts of demand, the potential for one person alone to generate this demand is so high enough that like.

• 19:45Noah Smith?

  So Sam Altman tells his infinite army of robots to go out and colonize the galaxy. We count that as consumption. We put a value on it and that's GDP.

• 19:51Speaker 00

  Yeah, or it might be investment. Maybe he's going to defer his consumption to once he's like after he's colonized the galaxy. Yeah, yeah. And I'm not saying this is the world I want. I'm just saying, think about it physically. If you're colonizing the galaxy, which you can do potentially after AGI, I'm not saying like it'll happen tomorrow after AGI, right? But this is a thing that's physically possible. Is that growth? Like something's happening that's like explosive.

• 20:09Noah Smith?

  Right, maybe. The thing is that. It's a very weird world. It doesn't look like the kind of economy you've ever had.

• 20:14Speaker 00

  Right.

• 20:15Noah Smith?

  And we created the notion of GDP to represent people exchanging money for goods and services, people like basically exchanging their labor for goods, exchanging the value of their labor for goods and services. That's at a fundamental level, that's what GDP is. We're envisioning a radical shift of what GDP means to a sort of internal pricing that a few overlords set for the things that their AI agents want to do. And that's incredibly different than what we've called GDP in the past.

• 20:39Speaker 00

  I think the economy will be incredibly different from what was in the past. I'm not saying this is like the modal world. There's a couple of reasons why this might not end up happening. One is even if your labor is not worth that much, the property you own is potentially worth a lot, right? If you own S and P 500 and there's been explosive growth, you're like a multi multimillionaire or the land you have is like worth a lot. If the AI can make such good use of that land to build the space probes, assuming that our system of property rights continues into this regime. And second, so I mean, in many cases, it's hard to ascribe how much economic growth there has been over very long periods of time. For example, if you're comparing the basket of goods that we can produce as an economy today versus like 500 years ago, it's not clear how you compare. We have antibiotics today.

• 21:24Speaker 00

  I wouldn't want to go back 500 years for any amount of money because they don't have antibiotics and I might die and it'll just suck. So there's actually like no amount of money to live in 1500 that we would rather have than live today. And so if we have those quality of goods for normal people, just like you can live forever, you have like euphoria, drugs, whatever, these are things we can imagine now. Hopefully it'll be even more compelling than that. Then it's easy to imagine like, okay, it makes sense why this stuff is worth way more than the stuff that the world economy can pursue for even for normal people today.

• 21:52Noah Smith?

  Right. And so I guess I'm just thinking about, this is a thing that economists really struggled with in the early 20th century. It's this idea that we had this capacity to expand production, expand production, expand production. And then the thing is that companies competed their profits to zero and the profits crashed and nobody wanted to expand production anymore because they weren't making any profit. We're seeing this happen again in China right now with overproduction. We're seeing BYD having to take loans from its suppliers just to stay financially afloat, even though it's the best car company in the world because the Chinese government has paid. A million other car companies to compete with BID. And so you overproduce, so you have this overproduction. So the solution was to expand consumption. This is the solution people are recommending for China now.

• 22:29Speaker 00

  Yeah.

• 22:29Noah Smith?

  To expand consumption so that you can reflow the profit margins of all these companies and have them continuous companies. And so the idea is if AI is producing all this stuff, but it's overproducing services, it's overproducing whatever AI can produce. And the profits from this go negative, that makes the GDP contribution go to zero. And basically, OpenAI and Anthropic and XAI and whatever will just be sitting there saying, why am I doing this again? Why am I, no one's buying this shit. And so at that point, it seems like there will be corporate pressure on the government to do something to redistribute purchasing power so that they don't compete their profits to negative. And so they have some reason to create more economic activity so they can take a slice of it, which is essentially what happened in the early 20th century.

• 23:09Speaker 00

  Yeah, I disagree with this. I think.

• 23:10Noah Smith?

  Oh, I'm not saying this will happen. I'm saying like, that would be the analogous thing.

• 23:13Speaker 00

  I disagree. I would prefer it to be the case that even as a libertarian, I would prefer for significant amounts of redistribution in this world because the libertarian argument doesn't make sense if there's no way you could physically pick yourself up by the bootstraps. Like, your labor is not worth anything, or your labor is worth less than subsistence calories or whatever, which is the more relevant thing. But I don't think this is analogous to the situation in China. I think what's happening in China is more due to the fact that you have the system of financial repression, which redistributes money and also currency manipulation, which basically redistributes ordinary people's money to basically producing one EV maker in every single province. So it is the market distortion that the government is creating that causes this overproduction. We can go into what like the analogous thing in the AI case looks like.

• 23:54Speaker 00

  But I think if there isn't some market distortion, I just think like, people will use AI where it has the highest rate of return. If it's not space colonization, it'll be like longevity drugs or whatever.

• 24:03Noah Smith?

  I'm just asking, like, why would I invest all this money into AI producing stuff? Why would I just invest the massive hundreds and billions of trillions and whatever of dollars into producing stuff for people who are all going to be out of a job and won't be able to buy this?

• 24:15Speaker 00

  But again, I don't think you'll be producing it for them. I think you'd be producing it for whoever does that. Like, there's stuff in the world, just somebody will have stuff. Maybe it's the AIs, maybe it's Sam Altman.

• 24:23Noah Smith?

  Right.

• 24:23Speaker 00

  You're producing it for whoever has the capability to buy your stuff. And will they want AI? And I'm just saying AI can do so many things, least of which is like colonizing the galaxy. People are willing to pay a lot of stuff.

• 24:33Noah Smith?

  Right, I'm just trying to-

• 24:34Speaker 00

  To colonize the galaxy.

• 24:35Noah Smith?

  Right. I'm just trying to get this straight in my head of what this economy looks like. And I'm seeing a picture of the trillions of dollars needed to build out all these data centers will be done not for profit, not to make money from a consumer economy for the creators of the AI, but to satisfy the whims of a few robot lords to colonize the galaxy.

• 24:50Speaker 00

  I think you're making two different points and they're getting-

• 24:53Noah Smith?

  Complet-

• 24:53Speaker 00

  They're getting wrapped into one.

• 24:55Noah Smith?

  Complet-

• 24:55Speaker 00

  I'm saying, yes. Important word. I, so there's one about do you expect it to be the case that the robot overlord world happens? And I'm saying, no, actually even without redistrib- First of all, I expect redistribution to happen. I hope it happens, but even if it doesn't, and I don't think it'll happen because people like corporations want it to redistribute happen. I think it'll be good to happen for independent reasons, but I don't buy this argument that the corporations will be like, we need somebody to buy our AI, therefore we need to give the money to the organ to your consumer.

• 25:20Noah Smith?

  You believe broad-based asset ownership will create a whole lot of broad-based consumer demand, even in the absence of labor income?

• 25:26Speaker 00

  Honestly, I don't have like a super strong opinion, but I think that's like plausible. But independent of that, I'm like, okay, even if that demand doesn't exist, just like the things you can do with a new frontier of technology, as long as one person wants it, there's so much room to do things. Space colonization is an obvious example.

• 25:42Noah Smith?

  It costs a lot of money.

• 25:44Speaker 00

  Right, there's like obvious demand for the things that AI will be able to produce, right? Like one of the things that AI can produce is colonize the galaxy.

• 25:49Noah Smith?

  Right, exactly. So, but the question is like, I can see a paperclip maximizing a autonomous intelligence as colonizing the galaxy, but in terms of-

• 25:56Speaker 00

  That's a lot of growth.

• 25:57Noah Smith?

  That is. But in terms of, and so by the way, I would like to say that I am a paperclip maximizer. I am the real paperclip maximizer. I want to maximize rabbits-

• 26:06Speaker 00

  Uh huh.

• 26:07Noah Smith?

  -in the galaxy. I want to turn the entire galaxy into fluffy rabbit.

• 26:09Speaker 00

  Right.

• 26:09Noah Smith?

  That's my goal.

• 26:10Speaker 00

  Yes.

• 26:11Noah Smith?

  And so my goal with AGI is to enlist the AGIs to help me in this goal, but then to align them toward rabbit.

• 26:16Speaker 00

  Yes.

• 26:16Noah Smith?

  But anyway.

• 26:17Speaker 00

  Get this guy in front of the OpenAI Board of Directors.

• 26:19Noah Smith?

  I know. I mean, like, the social welfare function is floofiness. But I guess my point here is as long as AI still doesn't have property rights and it's humans making all the economic decisions, be it Sam Altman and Elon Musk, or you know, you and me, then at that point, like- That really matters for what gets done because if we're talking about the money needed to build all these massive data centers, which currently it's a lot of money. It's a ton of money required to build these data centers. And that money need will not go away. We can't just say, oh, cost goes to zero because we can say unit cost goes to zero, but total cost doesn't go to zero, nor has it. It has increased. The total spend on data centers has increased, and I think everyone expects it to increase for the foreseeable future.

• 27:01Noah Smith?

  The question is, is that money being spent because AI companies expect to reap benefits from consumers like you and me? Or to what extent is it that? And to what extent is it Sam Altman feels like doing some crazy stuff and Sam Altman's just godlike richer than everybody else? And so Sam Altman is actually consuming when he builds those data centers. He is building those data centers so that he can indulge his godlike whims.

• 27:27Speaker 00

  Yeah, I think the more plausible than either a single godlike person is able to direct the whole economy, or like there's this broad-based consumer, every person's buying-

• 27:36Noah Smith?

  Thought example, these are extremes.

• 27:37Speaker 00

  Yes. I think more plausible is like AIs will be integrated through all the firms in the economy. A firm can have property. Firms will be largely run by AIs, even though there's nominally a human board of directors. And it might not even be nominal, right? Like, maybe the AIs are aligned and genuinely give the board of directors an accurate summary of what's happening. But like, day to day, they're being run by AIs. And firms can have property rights, firms can demand things.

• 28:01Noah Smith?

  So say all you have is a board of directors and AI.

• 28:03Speaker 00

  Yeah.

• 28:04Noah Smith?

  Okay.

• 28:04Speaker 00

  I mean, in the ideal world.

• 28:06Noah Smith?

  Okay, so then what we're basically looking at is the labor share of income goes to zero or something approaching that.

• 28:12Speaker 00

  Depends on how you define the AI labor.

• 28:14Noah Smith?

  And capital income is distributed highly unevenly. It's more distributed much more unevenly than labor income, but it's still distributed reasonably broadly. Like, I have capital income, you have capital income. So at that point, we have just an extremely unequal society where owners get everything and then workers get nothing. And then so we have to figure out what to do about that.

• 28:30Speaker 00

  Yeah, 100%.

• 28:30Erik Torenberg?

  Piketty is killing himself somewhere.

• 28:32Noah Smith?

  Piketty's been wrong about it.

• 28:35Noah Smith?

  So let's hope he's wrong again. I mean, he'd be happy. He'd be like, see, I was right. Because for an economist, being right is the most important.

• 28:42Erik Torenberg?

  Yeah, exactly.

• 28:43Speaker 00

  I mean, the hopeful case here is the way our society currently treats retirees and old people, who are not generating any economic value anymore. And if you just look at like the percent of your paycheck that's basically being transferred to old people, it's like, I don't know, 25% or something. And you're willing to do this because they have a lot of political power. They've used that political power in order to lock in these advantages. They're not like so overwhelming, you're like, I'm gonna go to like Costa Rica instead. You're like, okay, I had to pay this money, I had to pay this concession, I'll do it. And hopefully, humans could be in a similar position to this massive AI economy that old people today have in today's economy.

• 29:20Noah Smith?

  All right.

• 29:20Erik Torenberg?

  What do humans do? Let's say they get some money, they have enough to live. How do they spend their time? Is it art, religion, poetry?

• 29:27Speaker 00

  Podcasting.

• 29:27Erik Torenberg?

  Drugs?

• 29:28Speaker 00

  Podcasting, it's the final job.

• 29:30Erik Torenberg?

  Yeah, we're out of the, out of the curve here.

• 29:33Speaker 00

  Or is it the last man of history?

• 29:37Noah Smith?

  Wait, so here's an idea. How about sovereign wealth fund?

• 29:39Speaker 00

  Uh huh.

• 29:39Noah Smith?

  Okay, so sovereign wealth fund, we tax Sam Altman and Elon Musk.

• 29:43Erik Torenberg?

  We're using Sam as a metaphor here. He's a friend of the firm.

• 29:46Noah Smith?

  Yeah, yeah, yeah. We tax him, we tax Mark, and so then we use their money to-

• 29:49Speaker 00

  Only the friends of the show will be taxed, Eric.

• 29:51Noah Smith?

  Right. We use that money to buy like shares in the things that those people have. So they get their money back because we're buying the shares back from them. Okay, so it's okay. And then we hire them.

• 30:01Erik Torenberg?

  Yeah.

• 30:02Noah Smith?

  Because then what we do is we hire a number of firms, including A16Z, and pay them 2 and 20 or whatever, to manage the investment of AI stuff on behalf of the humans. But then the humans become broad-based sort of index fund shareholders, or shareholders in whatever you guys choose to invest in, then you take a cut. And this could be the future economy. This is what my PhD advisor, Miles Kimball, has suggested. This is what the socialist Matt Bruning has suggested. And this is what Alaska actually does with oil. Capitalists like it, socialists like it, Alaska likes it?

• 30:31Speaker 00

  I think sovereign wealth funds generally have a bad track record. There's some exceptions that have managed to use their wealth well, like Norway or Alaska. But there's just like these political economy problems that come up when there's this tight connection between the investment, which should theoretically be just highest rate of return and politicians. So I don't have like have a strong alternative. Ideally, you just let the market decide how the investment should happen and then you can just take a tax. But then exactly where does that tax happen? I haven't thought it through, but.

• 30:58Noah Smith?

  Are you dubious of this?

• 30:59Speaker 00

  Yeah, I wouldn't want the government influencing where that investment happens, but I want the government taking a significant share of the returns of that investment.

• 31:07Noah Smith?

  Yeah. Are you dubious of the trope that labor provides meaning? And if people don't have a clear sense for labor, then it will be very difficult for them to obtain alternative sources of meaning? Or is that kind of a capitalist sort of trope that isn't necessarily true?

• 31:21Speaker 00

  My suspicion is that humans have just adapted to so much, like agricultural revolution, industrial revolution, the growth of states, like once in a while, like a communist or fascist regime will come around or something like the idea that being free and having millions of dollars is the thing that finally gets us.

• 31:39Noah Smith?

  Yeah.

• 31:39Speaker 00

  I'm just suspicious of. By the way, do we not disagree about the thing? I'm saying, once we get AGI, humans will not have high paying jobs. Do we disagree about this?

• 31:52Noah Smith?

  I think humans may have high paying jobs.

• 31:54Speaker 00

  Okay.

• 31:55Noah Smith?

  Because of comparative advantage. The key here is if there's some AI specific resource constraint that doesn't apply to humans, then comparative advantage law takes over and then humans get high paying jobs, even though AI would be better at any specific thing than human. Because there's some sort of aggregate constraint. The example I always use, of course, is Marc Andreessen, who is the fastest typist I have ever seen in my life and yet does not do his own typing. And so because there's a Marc Andreessen specific aggregate constraint on Marc Andreessen's, there is only one of him. Who? So he hasn't taken all the secretary's typist jobs, but because he has better things to do. And so if there's some sort of AI specific resource constraint that hits, then humans could have. I'm not saying there will be.

• 32:34Speaker 00

  Yeah.

• 32:34Noah Smith?

  And I'm not saying there won't be.

• 32:35Speaker 00

  Yeah.

• 32:35Noah Smith?

  I'm saying, I don't know if there is.

• 32:36Speaker 00

  Yeah. The reason I find that implausible is that I think that will be true in the short term. Because right now, there's 10 million H100 equivalents in the world. In a couple of years, there might be 100 million. Like, H100 has the same amount of flops as a human brain. So theoretically, they're like as good as a brain if you had the right algorithm. So there's like a lower population of AIs, even if you had AGI right now, than humans. But the key difference is that in the long run, you can just keep increasing the supply of compute or of robots. And so if it is the case, so if an H100 costs a couple thousand dollars a year to run, but the value of an extra year of intellectual work is still like a hundred thousand dollars. So you're like, look, we've saturated all the H100s, and we're gonna pay a human a hundred thousand dollars because there's still so much intellectual work to do.

• 33:20Speaker 00

  In that world, the return on buying another H100, like an H100 costs forty thousand dollars, just like in a year, that H100 will pay you over 200% return, right? So you'll just keep expanding that supply of compute until basically the H100 plus depreciation plus running cost is the same as an extra year of labor. And in that world, that's like much lower than human subsistence. So comparative advantage is totally consistent

• 33:45Noah Smith?

  Okay.

• 33:45Speaker 00

  with human wages being below subsistence.

• 33:48Noah Smith?

  It is, but that comes from the common resource consumption. So if basically all of the land and energy that could be used to feed and clothe and shelter humans gets appropriated by H100s, then that is the case. However, if you pass a law that says this land is reserved for growing human food, if we actually were to just pass a simple law saying that you have to use these resources, these resources are reserved for humans.

• 34:16Speaker 00

  But at that point, the comparative argument-

• 34:17Noah Smith?

  Then comparative advantage-

• 34:18Speaker 00

  At that point, human labor has nothing to do with this. The only reason the system works is that you are basically transferring resources. You've come up with a sort of like intricate way to transfer resources to humans. It's just like, this resource is for you, you have this land, and therefore you can survive. And this is just like an inefficient way to allocate resources to humans.

• 34:36Noah Smith?

  It's true that it is an inefficient way.

• 34:38Speaker 00

  I think people will like cure this argument of comparative advantage and be like, oh, there's some intrinsic reason that humans will-

• 34:41Noah Smith?

  We take UBI instead.

• 34:42Speaker 00

  Yeah.

• 34:43Noah Smith?

  Okay.

• 34:43Speaker 00

  Yeah.

• 34:44Noah Smith?

  Yeah, I mean, sure. But then again, we typically do not see the first best, most efficient political solution implemented for things like redistribution. In the real world, redistribution happens via things like- The minimum wage or like letting the AMA decide how many doctors there's gonna be. So redistribution in the real world is not always the most efficient thing. So I'm just saying that like comparative advantage. If you're talking about will humans actually continue to get high paid work? Yes or no? It depends on political decisions that may, and it depends on physical constraints that will happen.

• 35:16Speaker 00

  But the high paid jobs are literally because like you have said that there must be high paying jobs politically like.

• 35:22Noah Smith?

  I understand you. In this case, you've said it in an indirect way, but you still said.

• 35:25Speaker 00

  Right, yeah.

• 35:25Noah Smith?

  You're absolutely right.

• 35:26Speaker 00

  Yeah.

• 35:27Noah Smith?

  You're not wrong.

• 35:27Speaker 00

  Yeah. I guess it's incredibly different from what somebody might assume.

• 35:31Noah Smith?

  Right.

• 35:32Speaker 00

  Like it has almost nothing to do with the comparative advantage argument.

• 35:35Noah Smith?

  Okay, sure, but that's true of a lot of jobs that exist now. Like a lot of jobs that exist now, I'm not sure what like university professors, there's a lot of those jobs. Or like credit rating agencies or you know, there's a lot of things where probably we could wring out some significant TFP growth. More or less by eliminating those things, but we don't because our politics is a cludocracy. I think this is one of Tyler's points.

• 35:57Speaker 00

  Yeah. I mean, I do think it's important to like point out in advance, like basically it would be better if we just bit the bullet about AGI. So that instead of doing redistribution by expanding Medicaid and then Medicaid can procure all the amazing services that AI will create. It'd be better if we just said, look, this is coming. And I'm not saying we should do a UBI today, but like if all human wages go below subsistence. Then the only way to deal with that is through some kind of UBI rather than if you happen to sue OpenAI, you get a trillion dollar settlement. Otherwise you're screwed.

• 36:29Noah Smith?

  Right.

• 36:31Erik Torenberg?

  Some people said the bare case for UBI was something around like COVID as an example. You gave people a bunch of money and what do they go do? Go riot in the streets. I'm teasing, but are people gonna use that money in an effective way?

• 36:41Noah Smith?

  I mean, that was literally what happened.

• 36:44Erik Torenberg?

  Yeah, so is UBI the form that you would think the like, what is the most effective method?

• 36:49Speaker 00

  The reason if there were UBI is like this thing where in a future world with explosive growth, we're going to see so many new kinds of goods and services that will be possible. That are not available today. And so distributing just like a basket of goods is just inferior to saying, oh, if like we solve aging, here's some fraction of GDP. Go spend your tens of millions partly on buying this aging cure, whatever this new thing that AI enables, rather than here's a food stamps equivalent of the AGI world that you can have access to.

• 37:18Noah Smith?

  Of course, I mean, this discussion may be academic because I believe that you said that we got phones and the world looked the same. I mean, no, it doesn't. Phones have destroyed the human race. Like, the fertility crash that's happening all around the world.

• 37:32Speaker 00

  Right.

• 37:32Noah Smith?

  Nobody has replacement level. Fertility is going far below replacement everywhere because of technology. And.

• 37:38Speaker 00

  Is that the phone or the pill or?

• 37:40Noah Smith?

  Well, no, it's a phone. I mean, we'll know the pill and other things like women's education, whatever, like lowered fertility like quite a bit, but some countries were still at replacement level. Some were still around replacement level, but the crash we've seen since everybody got phones is epic and is just unbounded. The human race does not have a desire, a collective desire to perpetuate itself. Yes, we're going to get lonely, but we'll have company through AI and through the internet, social media, until there's just a few of us and we dwindle and dwindle. Yeah, I mean, like, technology has already destroyed the human race. And basically, UBI is just like keeping us around on life support for a little while while that plays out.

• 38:16Speaker 00

  I do think so far there's been a lot of negative effects from widespread TikTok use or whatever that we're still like learning about. I am somewhat optimistic that in the long run, there's some optimistic vision here that could work. Just because right now, the ratio of like, it's impossible for Steven Spielberg to make every single TikTok and direct it in a sort of really compelling way. That's like genuine content and not just video games at the bottom and some music video at the top. In the future, it might genuinely be possible to give every single person their own dedicated Steven Spielberg and create incredibly compelling but long narrative arcs that include other people they know, et cetera.

• 38:57Noah Smith?

  Oh, yeah.

• 38:57Speaker 00

  So in the long run, I'm like, maybe this. I don't think TikTok is like the best possible medium.

• 39:03Noah Smith?

  No, I also don't think TikTok is unique in destroying the human race. I think that interacting online instead of interacting in person.

• 39:09Speaker 00

  I'd make your money.

• 39:10Noah Smith?

  That's a great. How do you make your money? Go ahead.

• 39:13Speaker 00

  I agree.

• 39:14Noah Smith?

  We're all making, we're all making money destroying our species.

• 39:16Erik Torenberg?

  You don't think we get isolated to dating apps and?

• 39:18Noah Smith?

  No, I'm saying like, as long as you can get your, why did humans perpetuate the human species? It was not because they wanted to see the human species perpetuated. It was because it's like, oop, I had sex and there came a baby. And that's done. We've severed that. That is the end. We did not evolve to want our species to continue.

• 39:33Erik Torenberg?

  Right, but you're saying the reasons why we're not having babies is because we can make friends on the internet. But is it that dating apps have created a much more efficient market? And thus there is maybe?

• 39:41Noah Smith?

  Pair, pair, bud. I don't know. I mean, like, people are having less sex. If Elon gets his way, everybody will just sit there gooning to some sort of Grok companion.

• 39:49Erik Torenberg?

  The goonpocalypse seems upon us.

• 39:51Speaker 00

  Is this available right now? What's the website?

• 39:53Noah Smith?

  Oh, no.

• 39:57Noah Smith?

  This podcast got silly, but anyway, I guess the point is that the idea of a humanity that just keeps increasing in numbers and spreading out to the galaxy. I don't see a lot of evidence that is in our future and that we have to go to great lengths to make sure that future is compatible with AGI. Because I don't think it's happening in any case, AGI or none.

• 40:17Erik Torenberg?

  By the way, not to cope too hard, but in a world where AGI happens, how important is increasing population?

• 40:23Speaker 00

  I mean, population has so far been the decisive factor in terms of which countries are powerful. Like, the reason China, if US was not involved, the reason China could take over Taiwan is just that there's 1.4 billion Chinese people and there's 20 million Taiwanese people. Now, if in future your population is, your effective labor supply is like largely AIs, then this dynamic just means that like your inference capacity is literally your geopolitical power, right?

• 40:51Erik Torenberg?

  I want to shift to short term a bit. You've had some people on the podcast, you have the AI 2027 folks who believe that AGI is perhaps two years away. I think they updated to three years away. And then you've also had some folks on who said it's not for 30 something years. Maybe you could steel man both arguments and then share where you netted out.

• 41:07Speaker 00

  Yeah, so two years, if I'm steel manning them, is that, look, if you just look at the progress over the last few years, it's reasoning. Aristotle's like, the thing that makes humans is reasoning. It was not that hard, right? Like, train on math and code problems. And have it like think for a second and you get reasoning like it's crazy. So what is a secret thing that we won't get?

• 41:28Noah Smith?

  Can I ask a stupid question? Why was stuff like O three type models? Why are those called reasoning models? But like GPT four O is not called reasoning. What are they doing different that's reasoning?

• 41:39Speaker 00

  One, I think it's GPT three can technically do a lot of things GPT four can. But GPT four just does it way more reliably. And I think this is even more shrew of reasoning models relative to GPT four O where four O can solve math problems. And in fact, like modern day four O has been probably trained a lot of math and code. But the original GPT four just wasn't trained that much on math and code problems. So like it didn't have whatever meta meta circuits there exist where like, how do you backtrack? How do you be like, wait, but I'm on the wrong track. I gotta go back. I gotta pursue the solution this way.

• 42:08Noah Smith?

  Algorithmically, I have a okay idea of what a reasoning model does that the non reasoning models don't. But in terms of how does that map to a thing that we call reasoning? What is the definition of what it means to reason that these people are using? The operational definition here, because I don't understand that myself.

• 42:25Speaker 00

  I mean, four O can't get a gold in IMO.

• 42:28Noah Smith?

  Okay, but I can reason and I can't get a gold in IMO. But I could reason.

• 42:32Speaker 00

  Yeah, but I can't get a gold either, but I don't think I can reason as well as math Olympiad, at least in the relevant domain. I agree that reasoning is not just about mathematics, but this is true of any word you come up with, like the zebra. What about the thing that like is a mixture of a zebra and a giraffe? And they have a baby. Is that a zebra still? I agree, there's edge cases to everything, but there's a general conceptual category of zebra. And I think there's like a general conceptual category of reasoning.

• 42:53Noah Smith?

  Okay, I was just wondering what it is. Like when you have a checkout clerk, right? That checkout clerk wouldn't would look at an IMO problem and be like, what? But then like you have a checkout clerk and the checkout clerk, you're like, okay, so you put the thing on this shelf and therefore someone has he looked for it and didn't find it. So something else must have happened.

• 43:11Speaker 00

  But I think a reasoning model, I think a reasoning model will be more reliable and be better at solving that kind of problem than four O.

• 43:18Noah Smith?

  So you're still manning the AI twenty twenty seven?

• 43:20Speaker 00

  Yes, so a lot of things we previously thought were hard have just been incredibly easy. So whatever additional bottlenecks you are anticipating, whether it's this continual learning on the job training thing, whether it's computer use, this is just going to be the kind of thing where in advance it's like, how would we solve this? And then deep learning just works so well that we like. I don't know, try to train it to do that and then it'll work. The long timelines people will say, I don't know, there's a sort of longer argument. I don't know how much to bore you with this, but basically the things we think of as very difficult and requiring intelligence have been some of the things that machines have gotten first. So just adding numbers together, we got in the 40s and 50s. Reasoning might be another one of those things where we think of it as the apogee of like human abilities, but in fact, it's only been recently optimized by evolution over the

• 44:03Speaker 00

  last few million years, whereas things like just moving about in the world and having common sense and so forth, and having this long-term memory. Evolution spent hundreds of millions, if not billions of years optimizing those kinds of things. So those might be much harder to build into these AI models.

• 44:17Noah Smith?

  I mean, the reasoning models still go off in these crazy hallucinations that they'll never admit were wrong and will just gaslight you infinitely on some crap it made up. Like just knowing truth from falsehood.

• 44:28Speaker 00

  Yeah.

• 44:28Noah Smith?

  I've met a couple of humans who don't seem to be able to know truth from falsehood. They're weird.

• 44:32Speaker 00

  Yeah.

• 44:33Noah Smith?

  And so, but O3 sometimes does this.

• 44:35Speaker 00

  I mean, I think it's a good question. Do they hallucinate more than the average person? I think no, less. Again, hallucinate meaning like getting something wrong and when they push them on it, they're like, no, whatever. And eventually they'll like accede if there's, they're clearly wrong. I think like, they're actually more reliable than the average.

• 44:50Noah Smith?

  But so the thing about the average human is you can get the average human to not do that.

• 44:53Speaker 00

  Right.

• 44:54Noah Smith?

  With the right consequences.

• 44:55Speaker 00

  Yeah.

• 44:55Noah Smith?

  And maybe AI, we haven't found the right like reinforcement, learning function or whatever to get them to not do that.

• 45:00Speaker 00

  Right.

• 45:01Noah Smith?

  Okay, now let's get to the view that it's 30 years away. What's that view?

• 45:04Speaker 00

  Just this thing of reasoning is relatively easy in comparison to, forget about robotics, which is just going to be, evolution spent billions of years trying to get like robotics to work, but there's like other things involved with like tracking long run state of, you know, a lion can follow up prey for a month or something, but these models can't do a job for a month. And these kinds of things are actually much more complicated than even reasoning.

• 45:28Noah Smith?

  Hmm. And where you've netted out is it's either going to happen in a few years or not for quite some time?

• 45:33Speaker 00

  Yeah. Basically, the progress in AI that we've seen over the last decade has been largely driven. By stupendous increases in compute. So the compute used on training a frontier system has grown four X a year for, I think, like, the last decade. And that just, over four years, is 160 X, right? So that's over the course of a decade, that's hundreds of thousands of times more compute. That physically cannot continue if you just like, okay, what would it mean? Right now, we're spending 1.2% of GDP or something on data centers. Not all of that is for training, of course, but what would it mean to continue this for another decade? For maybe five more years, you could keep increasing the share of energy that we're spending on training data centers or the fraction of TSMC's leading edge nodes, wafers

• 46:18Speaker 00

  that we dedicate to making AI chips, or even the fraction of GDP that we can dedicate to AI training. But at some point, you can't keep this like four X trend going a year. And after that point, then it has to just like come from new ideas of like, here's a new way we could train a model.

• 46:31Noah Smith?

  And by the way, when I was writing that comparative advantage post, and I was thinking about AI-specific aggregate constraints, resource constraints. That's what I was thinking of, actually. That that expansion of compute has to solve that. But I don't know how much it matters.

• 46:45Speaker 00

  That's for training and yeah, for the labor that will be like the inference will also use the same bucket of compute. It is the case that for the amount of compute it costs to train a system, if you like set up a cluster to train a system, you can usually run 100,000 copies of that model at typical token speeds on that same cluster. That's still obviously not like billions. But if we've got all this compute to be training these huge systems in the future, it would still allow us to sustain a population hundreds of millions, if not billions of AIs. At that point, maybe we obviously will still need one more AIs.

• 47:15Noah Smith?

  What does a single AI mean in this instance?

• 47:17Speaker 00

  Oh, like when you're talking to Claude, it's like a single instance that's talking to you.

• 47:21Noah Smith?

  So instances?

• 47:22Speaker 00

  Yeah.

• 47:23Noah Smith?

  So what's going to determine whether it's in a few years or?

• 47:26Speaker 00

  Right now, we're basically riding the wave of this extra compute. That's why AI is getting better every year mostly. In terms of the contribution of new algorithms, it's a smaller fraction of the progress that's explained by that. So if we've just got this like rocket, like how high will it take us? And does it get us space or not? If it doesn't, then we just have to rely on the algorithmic progress, which has been a sliver.

• 47:45Noah Smith?

  Yeah. But you think it might get us space?

• 47:48Speaker 00

  Yeah, I think there's like a chance that like, oh, continual learning is also like, you know, I had this whole theory about, oh, it's so hard and how do you start it in? And they're like, I fucking trained it to do this. Like, what are we talking about here?

• 47:57Noah Smith?

  That leads into another thing that I've thought about, which is how poor our track record for making predictions about the future of AI has been. The first time you and I hung out, I don't know if you remember this, was with Leopold.

• 48:10Speaker 00

  Yeah, oh, really?

• 48:10Noah Smith?

  Yeah.

• 48:11Speaker 00

  Yeah, I remember this.

• 48:11Noah Smith?

  It was at your old house.

• 48:12Speaker 00

  Yes.

• 48:12Noah Smith?

  And Leopold is just pronouncing a whole bunch of pronouncements from the couch. And he released this big situational awareness thing. How long ago was that? A year and a half?

• 48:21Speaker 00

  Yeah.

• 48:22Noah Smith?

  Yeah. I would say that already, most of the things he predicted have been invalidated or made irrelevant.

• 48:29Speaker 00

  Really?

• 48:29Noah Smith?

  In the last year and a half, and especially in terms of like, all the stuff about competition with China, like, it turns out filtration was able to get them a whole lot of things that he never predicted. It turns out that so many of the things, other than just the idea that AI would keep getting better, which he predicts and a lot of people predict, but then I feel like a lot of the specific predictions about US capabilities and Chinese capabilities and what would be the bottlenecks and what would be the things that, you know, here's how we can compete with China, that's all been proven wrong since.

• 48:53Speaker 00

  I think this is actually an interesting trend in the history of science, where like, some of the scientists who are the smartest in thinking about the progression of the atom bomb or progression of physics, just had these like, ideas about, the only way we can sustain this is if we have one world government. I'm just talking about after World War II.

• 49:07Noah Smith?

  Right.

• 49:08Speaker 00

  There's no other way we can deal with this new technology. I do think, relative to the technological predictions, Leo, I think the main way in which he's been wrong is that it didn't take some like, breaking the servers in order to learn how O3 or something works. It was just public, just seeing you being able to use the model.

• 49:25Noah Smith?

  You can talk to it and learn what it knows.

• 49:26Speaker 00

  Just knowing a reasoning model works, and then you can like, use it and you see like, oh, what is the latency? Like, how fast is it outputting tokens? That will teach you, like, how big is the model? Like, you learn a lot just from publicly using a model and knowing a thing is possible. He has been right in one big way, which is like, he identified three key.

• 49:41Speaker 00

  Things that would be required to get us from GPT-4 to BBAGI kind of thing, which was being able to think, so test time compute, onboarding.

• 49:49Noah Smith?

  Did you talk about test time compute in that?

• 49:50Speaker 00

  Yeah, yeah. It was like one of his, one of his three big un-hobblings. Then like onboarding in terms of the workplace. And then I think the final one was computer use. Look, one out of three, and it was a big deal. So I think you got some things right, some things wrong, but yeah.

• 50:05Noah Smith?

  And what's your take on the model of automating AI research as the path to AGI?

• 50:09Speaker 00

  The meter uplift paper, contrary to expectations, they found that whenever senior developers working in repositories that they understood well, used AI, they were actually slowed down by 20%.

• 50:20Noah Smith?

  Yeah, I did see that.

• 50:21Speaker 00

  Yeah. Whereas they themselves thought that they were sped up 20%. And

• 50:25Noah Smith?

  Right.

• 50:26Speaker 00

  So there's a bunch of

• 50:26Noah Smith?

  I'm getting things done.

• 50:29Speaker 00

  This was back to your theory about the phones are destroying us. That is us update towards the idea that AI is not on this trend to be this super useful assistant that's helping us already make the short process of training AI are much faster and this will just be this feedback loop and exponential. I have other independent reasons. I'm like, I don't know, I'm like 20% that like will have some sort of intelligence explosion.

• 50:52Noah Smith?

  One of the other Leopold predictions was nationalization. Is that something you could potentially foresee in the next few years?

• 50:57Speaker 00

  I don't think it's politically plausible, especially given this administration. I don't think it's desirable. First, I think it would like just drastically slow down AI progress because, look, this is not 1945 America and also building an atom bomb is like a way easier project than building AGI.

• 51:14Noah Smith?

  But China's quasi-nationalizing most of its, I mean, China doesn't control BYD's day-to-day decisions about what to build, but then if China says do this, BYD does it, as does every Chinese

• 51:23Speaker 00

  I mean, that's kind of the relationship American companies have the US government as well.

• 51:26Noah Smith?

  You think so?

• 51:27Speaker 00

  I mean, somewhat. Also, the big difference is, what do we mean by nationalization? There's one thing which is like, there's a party cadre who is

• 51:34Noah Smith?

  In your company.

• 51:35Speaker 00

  Exactly. There's another which is that East Province is like just pouring a bunch of money into building their own competitor to BYD in this potentially wasteful way. That like distributed competitive process is seems like the opposite of nationalization to me. Like when people imagine AGI nationalization, I don't think they're saying like Montana will have their AGI and Wyoming will have their AGI and they'll all compete against each other. I think they imagine that all the labs will merge, which is actually the opposite of how China does industrial policy.

• 52:00Erik Torenberg?

  But then you do think that the American government, basically if it says do this, then like XAI and OpenAI will do it.

• 52:07Speaker 00

  No, actually, I think in that way, obviously, the Chinese system and the US system are different. Although it has been interesting to see that whenever, I don't know, we've noticed the way that different lab leaders have changed their tweets in the aftermath of the election. I mean, also.

• 52:18Erik Torenberg?

  Yeah, more bullish up its source.

• 52:20Speaker 00

  Right. And didn't Sam have a thing where, I think previously he said that AI will take jobs. How do we deal with this? And then didn't he recently say something at a panel where I think President Trump is correct that AI will like create jobs or something where I don't think in the long run you believe this.

• 52:34Erik Torenberg?

  But the reason why humans should be excited about even their jobs being taken is just they'll be so rich that why do they even need it?

• 52:40Speaker 00

  Yeah, yeah.

• 52:40Erik Torenberg?

  Much richer than they are now.

• 52:42Speaker 00

  Right. Modular, this redistribution slash not fucking it over with some guild like thing.

• 52:47Erik Torenberg?

  Yeah. You mentioned the atomic bomb and we also mentioned off camera that you don't think the nuke is a good comparison for what happens. How does it play out when a lab figures out AGI? What then happens? Is there a huge advantage if one country has it first or if one lab has it first, do they dominate?

• 53:01Speaker 00

  I think it's less like the nuclear bomb where there's a self-contained technology that is so obviously relevant to specifically this like offensive capability. And you can say, well, like there's nuclear power as well, but like neither of those, like nuclear power is just like this very self-contained thing. Whereas I think intelligence is much more like the industrial revolution. Where there's not like this one machine that is the industrial revolution. It is just this like broader process of growth and automation and so forth. So that being said.

• 53:29Erik Torenberg?

  So Brad DeLong's right and Robert Gordon is wrong. If Robert Gordon said there's only, it's four things, it's just four big things.

• 53:34Speaker 00

  Oh, really?

• 53:34Erik Torenberg?

  And then Brad DeLong is like, no, it's a process of discovering.

• 53:37Speaker 00

  Interesting.

• 53:38Noah Smith?

  What were Rob's four things again? Oh, I mean, electricity.

• 53:41Speaker 00

  Test time compute, not kidding.

• 53:43Noah Smith?

  The internal combustion engine, steam power, and then what was the fourth one? Like, maybe, like, plumbing? Right. I think was the fourth one?

• 53:52Speaker 00

  Yeah, or even in that case, maybe that actually is, maybe that's closer to how I think about it. But then you needed so many complementary innovations. So internal combustion engines, I think, invented in the 1870s. Drake finds the oil well in Pennsylvania in the 1850s. Obviously, it takes like a bunch of complementary innovations before, like, these two things can merge. Before they're just like, we're using the oil for the kerosene to light lamps.

• 54:11Noah Smith?

  Right.

• 54:12Speaker 00

  But regardless, so if it's this kind of process, it was the case that many countries achieved industrialization before other countries. And like, China was dismembered and went through a terrible century because the Qing Dynasty wasn't up to date on the industrialization stuff. And much smaller countries were able to dominate it. But that is not like, we developed the atom bomb first, and now we have decisive advantage.

• 54:34Noah Smith?

  It's because of us. If that had been Nazi Germany or the Soviet Union, it would have gone differently.

• 54:38Speaker 00

  Yeah.

• 54:39Noah Smith?

  How do you see the US-China competition playing out in terms of AI?

• 54:42Speaker 00

  I genuinely don't know. Yeah, I think it's like possible that there could be some positive some, like, it's not like a nuclear weapon where both countries can just adopt AI. And there is this dynamic where if you have higher inference capacity, not only can you deploy AIs faster and you have more economic value that's generated, but you can have a single model learn from the experience of all of its copies. And you can have this basically broadly deployed intelligence explosion. So I think it really matters to get to that discontinuity first. I don't have a sense of at what point, if ever, is it treated like the main geopolitical issue that countries are prioritizing.

• 55:21Speaker 00

  I also from the misalignment stuff, the main thing I worry about is the AI playing us off each other rather than us playing the AIs off each other.

• 55:30Noah Smith?

  You mean AI just like telling us all to hate each other, the way like Russian trolls currently tell us all to hate each other?

• 55:36Speaker 00

  More so like the way that the East India Company was able to play different provinces in India off of each other. And ultimately, at some point, you realize, okay, like, they control India. And so you could have a scenario like, okay, think about the conquistadors, right? A couple hundred people show up to your border. And they take over an empire of 10 million people. And this happened not like once, it happened two to three times. Okay, so why was this possible? Well, it's that the Aztecs, the Incas, weren't communicating with each other. Like, they didn't even know the other empire existed. Whereas Cortez learns from the subjugation of Cuba, and then he takes over the Aztecs. Pizarro learns from the subjugation of the Aztecs and takes over the Incas. And so they're able to like just learn about, okay, you take the emperor hostage, and then this is the strategy you employ, et cetera.

• 56:18Noah Smith?

  It's interesting, the Aztecs and Incas never met each other, and that worked both times, sort of.

• 56:24Speaker 00

  Yeah.

• 56:24Noah Smith?

  That's interesting that these totally disconnected civilizations both had the similar vulnerabilities.

• 56:29Speaker 00

  Yeah, I mean, it was like literally the exact same playbook. The crucial thing that went wrong is that at this point in the 1500s, we actually don't have modern guns, we have arquebuses. But the main advantage that the Spanish had was they had horses. And then secondly, they had armor, and it was just incredibly, you'd have thousands of warriors. If you're fighting on an open plain, the horses with armor will just like trounce all of them. Eventually, the Incas had this rebellion, and they learned, they can like roll the rocks down hills. And the rebellion was moderately successful, even though it was eventually, we know what happened.

• 56:57Noah Smith?

  You could say that the Spanish on their side had guns, germs, and steel.

• 57:02Speaker 00

  So how could this have turned out differently? If the Aztecs had learned this, and then had like told the Incas, I mean, they weren't in contact. But if there was some way for them to communicate, like, here's how you take down a horse. I think what I would like to see happen between the US and China, basically, is like the equivalent of some red telephone during the Cold War, where you can communicate, look, we noticed this, especially when AI becomes more integrated with the economy and government, et cetera. Like, we noticed this crazy attempt to do some sabotage, like be aware that this is a thing they can do, like train against it, et cetera.

• 57:30Noah Smith?

  Right, AI is trying to trick you into doing this, watch out.

• 57:33Speaker 00

  Yeah, exactly. The required level of trust, I'm not sure is plausible, but that's the optimal thing that would happen.

• 57:39Noah Smith?

  At the lab level, do you think it's a multipolar? Or is there consolidation? And who's your bet to it?

• 57:44Speaker 00

  I've been surprised. So you would expect over time, as the cost of competing at the frontier has increased, you would expect there to be fewer players at the frontier. This is what we've seen in semiconductor companies, right? That it gets more expensive over time. There's now maybe one company that's at the frontier in terms of like global semiconductor manufacturing. We've seen the opposite trend in AI, where there's like more competitors today than there were a year ago, even though it's gotten more expensive. I don't know where the equilibrium here is, because the cost of training these models is still much less than the value they generate. So I think it would still make sense to 10x the amount of investment. Somebody new to come into this field and 10x the amount of investment. Do you have a take on where the equilibrium is?

• 58:21Noah Smith?

  Oh, well, I mean, it has to do with entry barriers. Basically, it's all about entry barriers. It's the question of if I just decide to plunk down this amount of money. So if the only entry barrier is fixed costs, I'd say we have such a good system for like just loaning people money that that's not going to be that big a deal. But if there's entry barriers that have to do with if you make the best AI, it gets even better. So, you know, why enter? That's the big question. I don't actually know the answer to that question.

• 58:48Erik Torenberg?

  There's a broad question we ask in general is like, what are the network effects here?

• 58:51Noah Smith?

  Right.

• 58:51Erik Torenberg?

  And what is the visibility? And it seems often to be brand.

• 58:55Noah Smith?

  Yeah, I mean, I'm not sure that's a network effect, but brand like OpenAI, ChatGPT is the Kleenex of AI in that Kleenex is actually called a tissue, but we call it a Kleenex because there was a company called

• 59:06Speaker 00

  Where are you going with this? Are we back at the Brock Dooney thing?

• 59:09Noah Smith?

  Oh, no.

• 59:12Noah Smith?

  Well, what's another example? Xerox.

• 59:14Erik Torenberg?

  Yeah.

• 59:14Noah Smith?

  Use Xerox this thing. Xerox is just one company that makes a copier, right? Not even the biggest, but everybody knows that a Xerox in.

• 59:20Speaker 00

  Right.

• 59:20Noah Smith?

  And so ChatGPT gets massive rents from the fact that everyone just says, I'll use AI. What's an AI? ChatGPT, I'll use it. And so like, brand is the most important thing.

• 59:31Speaker 00

  But I think that's mostly due to the fact that

• 59:32Noah Smith?

  So far.

• 59:33Speaker 00

  this key capability of learning on the job has not been unlocked. And so

• 59:37Noah Smith?

  Right. And I was saying that could be a technological network effect that could supersede the brand effect.

• 59:42Speaker 00

  Right, yeah, yeah. And I think that that will have to be unlocked before most of the economic value of these models can be unlocked. And so by the point they're generating hundreds of billions of dollars a year, or maybe trillions of dollars a year, they will have had to come up with this thing, which will be a bigger advantage, in my opinion, than brand network effects.

• 1:00:01Erik Torenberg?

  Is Zuck throwing away money, wasting it on hiring all these guys?

• 1:00:04Speaker 00

  No, people have been saying like, look, the messaging could have been better or whatever. I mean, I think it's just much better to have worse messaging or something, but then not sleepwalk towards losing. Also, if you just think about like, if you pay an employee a hundred million dollars and they're a great AI researcher and they make your compute, your training or inference 1% more efficient, Zuck is spending on the order of like $80 billion a year on compute. If that's made 1% more efficient, that's easily worth a hundred million dollars. Like a hundred million dollars is below the break even point for this extra researcher. So the real question is like, why haven't we hit that break even point yet?

• 1:00:38Erik Torenberg?

  And if we as podcasters encourage one researcher to join Meta, I mean, what's the? How do you put a price on that?

• 1:00:44Speaker 00

  Yes.

• 1:00:46Erik Torenberg?

  This has been a phenomenal conversation. Noah and Dorkesh, thank you so much for coming on. It's been great.

• 1:00:50Speaker 00

  Awesome. Thanks, Eric.

• 1:00:54Erik Torenberg?

  Thanks for listening to the A16Z Podcast. If you enjoyed the episode, let us know by leaving a review at ratethispodcast.com/A16Z. We've got more great conversations coming your way. See you next time.