What remains scarce after AGI? – Alex Imas and Phil Trammell

1. 0:00 – 19:36

   Will capital share increase?

   1. DPDwarkesh Patel0:00

      Today, I'm chatting with Alex Imas, who is director of AGI economics at Google DeepMind and professor of economics at University of Chicago, and Phil Trammell, who is head of economics at Epoch and research scholar at Stanford. In general, in this interview, what I want to understand is what economics tells us about what we can expect in a world with more and more automation, more and more advanced AI, what that tells us about what will happen to wages, to labor share, uh, what the best way to tax and redistribute the wealth that we generated as a result of AGI will be, um, and what kinds of things will be scarce. Because what is scarce kind of tells you where the value will accrue. So I wanna start there. What are some plausible candidates of what will be scarce?

   2. AIAlex Imas0:44

      Something like the relational sector, which is what I defined as, um, you know, basically services and goods, where the fact that the human was in the loop was actually part of the value of that product. So because humans are naturally scarce, if we have automation where a lot of other things stop being scarce, uh, we will still have scarcity in things that humans are kind of involved in and in the loop for.

   3. DPDwarkesh Patel1:04

      Hmm. I'm curious to understand whether humans doing services for other humans can ever be a big part of the economy. And here's maybe one intuition pump. So in a world where AI can physically do anything humans can do, you know, there's this whole machine economy where they're, like, building factories and doing research and coming up with new ideas, and humans may or may not be involved in the physical production of those things, but probably not, given that in the ultimate limit, if robotics is solved, if you don't care about humans being involved in that process, why would humans be involved in that process? But then there's these other things which you point out where, well, we actually maybe in some cases do want the ballerina or the barista or whatever to be a human. That's part of the value of going to a cafe or a performance. But only humans have that preference, so there's this human economy where, like, humans are doing services for each other and part of their wealth is flowing to other humans, but part of their wealth is also, like, they, they will want some of the automated goods this, like, machine-only economy is creating, and so part of that wealth is flowing out. And so if you just think of this as like, this is not a closed loop, but a lot of things in the machine-only economy are a closed loop because the machines don't care about, like, getting the human barista to make them a coffee. And so w- within that model, isn't, isn't it intrinsic that, like, the human-only economy will become a smaller and smaller share?

   4. AIAlex Imas2:24

      I would like to pitch kind of a rephrasing of that question.

   5. DPDwarkesh Patel2:27

      Sure.

   6. AIAlex Imas2:27

      So I think my view is that kind of forecast that economists like us would make are not necessarily as individual forecasts like me and, me and Phil are talking right now, are not necessarily very useful.

   7. DPDwarkesh Patel2:38

      Hmm.

   8. AIAlex Imas2:38

      The reason I think that, so there was this blog post by Andrei, uh, Fredkin, Brian Dabirian, and Andrew Koh that came out yesterday actually, uh, that looked at, like, kind of people's forecasts, economists' forecasts about the labor market. And what they found is that there's a ton of disagreement, like, in every single direction. So what they advocate for, and I think I'm, I'm in agreement here, is rather than thinking about individual forecasts like what me and Phil are gonna do, rather than looking at kind of like basically generating prediction markets where you get aggregate forecasts, where you get, like, kind of, uh, uh, wisdom of the crowd effects. And kind of the reason that I think this is because we have been famously terrible at forecasting. And so let's, let's take, let's go all the way back to 1820. Um, this sort of debate that we've been having actually is like two hundred years old. So David Ricardo is one of the classic economists, not neoclassical, classical economists, and he, when Industrial Revolution started happening, he was... wrote a bunch of stuff saying like, "Look, this is be, gonna be great for everybody. Prices are gonna come down." But then he turned around and he's like, "Wait, I can actually see all of these jobs that are creating value. They're gonna be automated by these machines. This is gonna be really bad. Everybody's gonna become unemployed, and there's gonna be political unrest and things like that." And if you look at Ricardo's predictions, they're actually right. If you look at all those jobs that made money in Ricardo's time, they got automated. So if I was David Ricardo and I woke up and somebody told me all those jobs did get automated, and you asked me, Dave Ricardo, like, "What do you think the, uh, prime age employment rate is in twenty twenty-six?" I think he would be surprised if you told him it was the highest it's ever been other than 2000. We have the highest number of employed people that could potentially be employed since 2000. That was like the peak, and now it's like the second peak, basically. So what David Ricardo ended up missing is the fact that, you know, essentially you have these economics of structural change where basically everything that got automated became cheap. People had more money to spend on things, and then they started spending money on services. And you know, this is kind of like the lump of labor fallacy. That's what they call it. David Ricardo didn't think, "Hey, I should have, you know, considered the fact that new jobs would be created." But it's kinda not obvious that, like, money would go to services. Like, why wouldn't they go to more automated goods or-

   9. DPDwarkesh Patel4:58

      Right

   10. AIAlex Imas4:58

       ... and something like that? And I'm not saying that, like... I'm not using this anecdote as to say like, "This is what's gonna happen now. We're gonna have full employment." I'm using that anecdote as to say it's really hard to make predictions. And what I think may be a, a really useful tool that econ-economists have is instead start with a premise. Like maybe we'll start it today. Look, labor share is zero. Like, labor share has gone down. What could possibly explain this? Let's write down an economic model of what happened. Phil will talk about this later today.

   11. DPDwarkesh Patel5:28

       Hmm.

   12. AIAlex Imas5:28

       Or you can start to write down a model to say, "Hey, what if labor share just stays the same? What can make that happen?" And here's my main-- Here's, if you don't take anything out of this conversation from me, we don't have any data. I've been kind of saying we need a Manhattan Project for data. We don't have data on basically consumer demand elasticities. We don't know what they are. We don't know, um, we're not really tracking what jobs are getting created or destroyed. Like the O*NET database with all of the tasks and different jobs, that's been rarely updated. It's super low quality. And so what I think is really useful is to think about, like, what are the potential scenarios, and we'll be talking about a lot of these scenarios, mapping them out and to say what type, what dimension of scarcityWill generate that scenario.

   13. DPDwarkesh Patel6:14

       Yeah.

   14. AIAlex Imas6:14

       So if there's full employment, we could talk about the relational sector or something like that. If there's n- you know, very labor share collapses, we can talk about other sorts of scenarios, and then that will tell us what data we should be collecting.

   15. DPDwarkesh Patel6:27

       It's probably worth defining labor share and capital share-

   16. AIAlex Imas6:29

       Mm-hmm

   17. DPDwarkesh Patel6:29

       ... real quick. So the whole economy, like, the, the total sum of goods and services sold, um, is either paid out to people in wages-

   18. AIAlex Imas6:38

       Yeah

   19. DPDwarkesh Patel6:39

       ... or it's paid out to, uh, capital, which is to say that there's, like, rents on buildings, and then there's shareholders of companies that would get paid out. And, um, for many hundreds of years in the economy, m- uh, 60-something percent of the economy, or all the things that are sold in a given year, basically gets paid out to humans in wages, and the other 30, 40% gets paid out to people who own machines and land and, uh, claims on companies and whatever. Um, and the question is, well, right now, if 60% is going to wages, does that shrink as automation, um, or as AIs get smarter and smarter and better and better?

   20. AIAlex Imas7:14

       Yeah. And it's like it really, this is a call to a fact. Like, right? So it's incredibly-- We should stress this. It's incredibly surprising-

   21. DPDwarkesh Patel7:22

       Yeah.

   22. AIAlex Imas7:22

       ... that it's over 60% after the Industrial Revolution, after all the automation we've ever seen. The fact that it, it's almost like some people are worried it's an accounting error or something like that, that it's kept being, been-

   23. DPDwarkesh Patel7:33

       Right

   24. AIAlex Imas7:33

       ... so constant, and, uh, the fact that it's, like, been over 60%. And, you know, there's, there's even a controversy right now. Uh, so some might say, like, you know, labor share has been falling in the last 20, 30 years. But, you know, depending on how you d- there's been a lot of accounting changes-

   25. DPDwarkesh Patel7:48

       Right

   26. AIAlex Imas7:49

       ... in the last 30, 40 years. So, for example, Andy, uh, Atkinson have this, has this paper showing that act- actually, if you keep the accounting constant over the years, labor share hasn't even fallen ever.

   27. DPDwarkesh Patel8:00

       But it's not, it's not that surprising, right? I mean, if, uh, Phil, you made this point that if labor and capital are complements, you need both to do anything. It would kind of make sense that you'd kind of need to pay both of them to get something done.

   28. AIAlex Imas8:10

       You, you have had stuff com- be completely automated.

   29. DPDwarkesh Patel8:13

       Although, uh, you had the post where you were pointing out that actually-- Sorry, you sh- Oh, yeah. Well, I was gonna say, there's a sense in which so- nothing's yet been completely automated. If you look at the network-adjusted factor shares of a good-

   30. AIAlex Imas8:26

       Mm-hmm
2. 19:36 – 25:57

   Messy Middle scenario

   1. DPDwarkesh Patel19:37

      There's one possibility which, um, Molly Kinder has written something about, this Messy Middle scenario, and there's-- That possibility made me think about whether it might be better to have, at least as far as wealth distribution, um, and redistribution goes, it might be better to have much faster AI takeoff.

   2. AIAlex Imas19:55

      Yeah. Yeah.

   3. DPDwarkesh Patel19:55

      And I wanna ask you whether the following possibility is at all likely, or there's any set of assumptions that this can make it so. Which is that AI makes it possible to automate jobs such that, like, many people are losing their jobs, but it doesn't create enough wealth while the process of automation is happening to pay off, basically, the people who are getting laid off. So there's, like, a Pareto improvement. Um, everybody's getting better as a result of AI automation. And of course, there's a trivial sense in which that must be true because whatever money you're saving, whatever money the company is saving by not paying the humans instead of just paying the AIs, those resources still exist in the economy, and they can just be paid off to people. But there's gonna be some allocative inefficiency. Like, the government doesn't know exactly who got laid off because of AI. There's some political problem of, like, if the meta worker gets laid off first, and they were making 200K a year, um, i- is there a politically sustainable situation where you give them a 200K check a year, uh, when there's many people who are working who are making much less? Um, so do you at all find this scenario plausible, where AI is actually automating a bunch of things, but there isn't enough wealth creation as there is automation?

   4. AIAlex Imas21:06

      Uh, I think it's-

   5. DPDwarkesh Patel21:06

      Is that plausible?

   6. AIAlex Imas21:07

      ... possible. To me, it does seem like a pretty narrow window. M- my guess is that if we have the technology to automate so many jobs that it becomes, like, a new kind of political problem, then the pie will also be growing really fast.

   7. DPDwarkesh Patel21:19

      Well-

   8. AIAlex Imas21:19

      Unless, in all of those professions that it's automating, it's just a hair more productive. So, like, the cost of all the capital to, uh, [laughs] to, to replace all the software engineers is just, you know, a hair less than the cost of-

   9. DPDwarkesh Patel21:31

      Right

   10. AIAlex Imas21:31

       ... what we've been paying the software engineers.

   11. DPDwarkesh Patel21:33

       And why, why is it implausible that it just, like, a company can save money by laying off a bunch of software engineers, but-- And in the long run, there's a Jevons paradox thing, and, you know, we can't anticipate in advance what we'd do with more software, and surely there's gonna be more uses. But in the short run, the effect is just that a lot of people are laid off, and they still need to figure out how they can use a million X more-

   12. AIAlex Imas21:53

       Yeah

   13. DPDwarkesh Patel21:54

       ... JavaScript tokens.

   14. AIAlex Imas21:54

       I think the thing that, like, is, in either, like, you know, Phil, Phil and I have been, like, writing about these things-

   15. DPDwarkesh Patel21:59

       Yeah

   16. AIAlex Imas21:59

       ... and we have mathematical models on the back of these things. We don't have any political economy in any of our models.

   17. DPDwarkesh Patel22:04

       Mm.

   18. AIAlex Imas22:04

       Andy Hall wrote a really nice blog post about the politics of AGI, and he made a really interesting observation. If there's a two percent increase in unemployment, the political winds completely change. Like, unemployment, it has a huge effect on what happens politically. So, you know, to, to Molly's, uh, excellent essay, by the way, um, I think in some ways, like, one of the worst scenarios is a drip scenario because of the political economy piece, right? Because, like, you know, people... Essentially, what you, what you might see is, like, people not really being unemployed en masse, but kind of, like, moving into sectors that pay them less money, kind of basically getting, uh, what happened with phone operators in, in, in the mid-century of the, of the, uh, between 1920 and 1940. Phone operators were completely automated, right? But it took twenty years, even though the technology existed. And therefore, there was this drip. It wasn't like this giant sector just disappeared.

   19. DPDwarkesh Patel23:00

       Yeah.

   20. AIAlex Imas23:01

       And what ended up happening, you-- There's a really nice, uh, QJE paper on this, basically showing that a lot-- They got reabsorbed into the economy, but at lower salaries, and they were mostly underemployed. And I think that's the scenario that Molly was, w- was writing about, this, like, kind of Messy Middle where, like, things aren't a disaster. Because we saw with COVID, like, the fiscal, fiscal response can move quickly if there's an emergency.

   21. DPDwarkesh Patel23:25

       Right.

   22. AIAlex Imas23:26

       And an emergency is a quick uptick in, uh, unemployment-

   23. DPDwarkesh Patel23:28

       Right

   24. AIAlex Imas23:28

       ... which could even look like two or three percent. That's like a national... That becomes a national emergency if it becomes fast.

   25. DPDwarkesh Patel23:34

       Um, the concern is that suppose what- whatever you're saving on those white-collar workers, if that's not growing the economy but is just creating some, you know, saved resources that can be allocated elsewhere, is that enough to do, um, a broad-based redistribution scheme? Because then you, you have, like, the money you saved off a couple of people.

   26. AIAlex Imas23:56

       Yeah.

   27. DPDwarkesh Patel23:57

       And unless you can figure out exactly how to get it to them specifically, you have the problem of can, can I do, like, can I do a UBI off the money I saved by laying off-

   28. AIAlex Imas24:04

       Yeah. So you're basically saying, like, look, the pie did not grow that much.

   29. DPDwarkesh Patel24:07

       Yeah.

   30. AIAlex Imas24:07

       You're just basically sa- you're just basically displacing a bunch of people, but, uh, that actually didn't grow the, the technological frontier of what the economy can produce.
3. 25:57 – 30:02

   How to tax and redistribute AI wealth

   1. AIAlex Imas25:57

      I have some thoughts. I thi- I think, I think it's, uh, just really important to outline the costs and benefits. Like, um, it's also important to know that there... So first, there's differential complexity in, like, implementing these things. Uh, two, they differ in the timeline of, like, being actually helpful. So, like, something like universal basic capital, that's not like, that's not gonna generate returns for something that happens in six months. So you probably are going to end up with a layer of things. So, like, for example, like, a negative income tax. Like, you implement it, and it, it, the day it turns into law, that is already, you already have this sort of insurance that, like, you know, there's a floor for which, you know, everybody, e- everybody gets a certain amount of money and then, you know, if you earn more money, you get taxed more and things like that. Um, and but, you know, there's positives and negatives to negative income tax. With UBI, the, for example, the, I, I, I, I w- worry a lot about, like, the political economy implications. Like, for example, like, if people are just kind of dependent on a check, it really matters who's in power. Like, right now, we're endowed with labor that can turn into, uh, that could turn into income. When that is no longer the case and we are now at the mercy of the polit- uh, of, uh, of the elected official for, like, basic needs, right? So that, to me, feels like a power-sharing arrangement that's really dangerous.

   2. DPDwarkesh Patel27:21

      But wouldn't that be true of any sort of government redistribution program?

   3. AIAlex Imas27:24

      So something like universal basic capital, where you have, like, an ownership share, and you have property rights for capital, then you just, you're just, you're just a share.

   4. DPDwarkesh Patel27:32

      You're, you're a normal shareholder.

   5. AIAlex Imas27:32

      You're just a normal person.

   6. DPDwarkesh Patel27:34

      And, uh, but this goes back to the question of indexing, because if indexing is hard, then universal-

   7. AIAlex Imas27:37

      Yeah, yeah

   8. DPDwarkesh Patel27:37

      ... basic capital is hard.

   9. AIAlex Imas27:38

      And that's the, that, that's the problem of universal basic capital is targeting.

   10. DPDwarkesh Patel27:41

       Right.

   11. AIAlex Imas27:41

       Right? Wh- what do you target to put into people's portfolios?

   12. DPDwarkesh Patel27:44

       Yeah. Like, what if Anthropic goes to zero, but some random robotics company takes all the surplus?

   13. AIAlex Imas27:47

       Exactly, exactly. So that's the risk of universal-

   14. DPDwarkesh Patel27:49

       Right

   15. AIAlex Imas27:49

       ... basic capital. With a negative income tax, you have the same sort of issues that with UBI, where, where, like, you know, somebody comes into power and says, like, "This is, we're not gonna do that anymore," and people can't work, and then, you know, you, you have the issue of the, the floor being lowered.

   16. DPDwarkesh Patel28:02

       Right. One concern with the wealth tax is that, you know, you, it, there's no political sus- politically sustainable equilibrium at, like, 0.5% wealth tax and, you know, it, it... I mean, this happened to the income tax, of course, right? It starts low. It's like for war or something, and then it slowly and slowly escalates until the marginal tax rate in the US is probably on the order of income tax rate is, like, forty percent or something, um, and in certain states, upwards of fifty percent. Um, with a capital tax, is there a reason to worry would, would that distort investment? Because people would just be like, "Why would I invest in Anthropic or Intel? The government is gonna take larger and larger shares of it and dilute my share."

   17. DPDwarkesh Patel28:40

       Well, well, hold on. So, um, I think it's worth separating, like, how the revenue is raised, like, what's taxed, and then how it's distributed. It could be that the government hands out shares of Anthropic to everyone by, by-

   18. DPDwarkesh Patel28:52

       By buying it

   19. DPDwarkesh Patel28:52

       ... a broad-based tax and then buying Anthropic.

   20. DPDwarkesh Patel28:54

       Yeah, okay, fair.

   21. AIAlex Imas28:55

       Mm-hmm, mm-hmm.

   22. DPDwarkesh Patel28:56

       Which would probably be the right thing to do.

   23. DPDwarkesh Patel28:58

       Yeah.

   24. DPDwarkesh Patel28:58

       I mean, hopefully some, like, populist proposal doesn't interfere with that [laughs] and, like, expropriate some, like, particular company that everyone happens to know about.

   25. DPDwarkesh Patel29:06

       Yeah.

   26. DPDwarkesh Patel29:06

       Um, but how... So you're, you're suggesting there could be a tax that is some sort of optimal tax. B- uh, it's we're taxing externalities or we're taxing land or we're... I, I guess we probably need to tax something other than just those two things, but that tax is-

   27. DPDwarkesh Patel29:20

       Or consumption, yeah.

   28. DPDwarkesh Patel29:20

       Huh?

   29. DPDwarkesh Patel29:21

       Or consumption.

   30. DPDwarkesh Patel29:21

       Okay. So a consumption tax, like a European value-added tax-
4. 30:02 – 39:26

   Why demand collapse is unlikely

   1. DPDwarkesh Patel30:02

      All right, I'm, I'm curious to understand. People talk about whether there's a white-collar apocalypse already. Is there any evidence, uh, that suggests that there is mass automation or unemployment as a result of AI already?

   2. AIAlex Imas30:17

      Um, I think there's a, a lot of people are looking at it, so this is an area where there's, like, a lot of eyes and a lot of data-

   3. DPDwarkesh Patel30:23

      Yeah

   4. AIAlex Imas30:23

      ... being produced. Uh, so the Budget Lab, uh, over at Yale is doing really good analysis on this. They just recently released a report, and I think, like, you really have to squint to see anything happening. Like, basically, if you wanna take kind of like, uh, a, an approach across the entire economy and looking at, even looking at, like, software engineering, like, the most-

   5. DPDwarkesh Patel30:42

      Yeah

   6. AIAlex Imas30:42

      ... exposed sort of sectors, there's just, like, not really anything going on. There might be a little bit of a signal about, like, junior developers getting jobs less than before and, but that's like a-Less than before rather than a level shift, is then there's actually an increased demand for senior manager, for, for senior software engineers, if anything. Um, and so if you look at trend, it's b- kind of like for junior managers, it's a bit below trend.

   7. DPDwarkesh Patel31:08

      So, as in you're saying the growth is slower than before.

   8. AIAlex Imas31:10

      Yes.

   9. DPDwarkesh Patel31:11

      But there is still growth even on entry-level software engineers.

   10. AIAlex Imas31:14

       Yeah. Yeah, exactly.

   11. DPDwarkesh Patel31:14

       And what do you think is going on with the anecdotal evidence of graduating college students saying that they're finding it harder to find CS jobs or something?

   12. AIAlex Imas31:22

       I think that's anecdotal evidence.

   13. DPDwarkesh Patel31:24

       You think it do- it, it's always been hard to get jobs for some people, and now it's getting turned into an AI narrative. Same with the layoffs, where it's probably just a normal layoff, and they turned it into an AI layoff.

   14. AIAlex Imas31:33

       Yeah, I mean, you, you have to be careful with all of this. I think, like, there are these, like, you know, there are these, like, coordina- public coordination devices for, like, let's say we get into a narrative where, like, if you're a firm and you're not laying people off, then you're seen as, like, not adapting AI enough.

   15. DPDwarkesh Patel31:48

       Yeah.

   16. AIAlex Imas31:48

       So, like, then you g- you're gonna just get a cascade effect-

   17. DPDwarkesh Patel31:51

       Right

   18. AIAlex Imas31:51

       ... of firms, like, just needing to keep up with the Joneses in terms of, like, starting to lay people off. And that's kind of like, that, that's super worrying where, like, actually the firm might be actua- worse off-

   19. DPDwarkesh Patel32:01

       Right

   20. AIAlex Imas32:01

       ... after the layoffs than before the layoffs, but it's just doing the layoffs to have the perception that, "Look, look, we're not behind the times. We're, we're inc- we're, you know, using AI." Like, you have the-- You probably heard these anecdotal stories of, like, these token counters that, like, you have to maximize tokens-

   21. DPDwarkesh Patel32:16

       Right

   22. AIAlex Imas32:16

       ... and things like that. So again, like, right now, we have, we don't really have, uh, any evidence of a white-collar bloodbath.

   23. DPDwarkesh Patel32:23

       And is that surprising at all? I, I feel, given the fact all these things AI can do, it was just like this is a story as old as time. If you automate some complementary task, the overall bucket of things that the, the human labor which complements the automation-

   24. AIAlex Imas32:37

       Yeah

   25. DPDwarkesh Patel32:38

       ... will, will increase in value.

   26. AIAlex Imas32:39

       So this is, this is one of the statistics that's really important for that argument is elasticity of demand.

   27. DPDwarkesh Patel32:44

       Yeah.

   28. AIAlex Imas32:44

       So, like, the-- you take the O-ring model of jobs. So, like, again, jobs is a series of tasks. Let's say the AI automates, like, nine out of ten, uh, nine out of ten, uh, tasks. One task is not automated. If that person can now kind of focus in on that task and a- the, the job will become more productive. If that translates into a price effect where the product is actually cheaper, if the cons, uh, demand responds enough where now there's, it's being bought more, it's being used more, the service is being used more, that could actually lead to more hiring.

   29. DPDwarkesh Patel33:16

       Right.

   30. AIAlex Imas33:16

       And a lot of people on the internet have been, like, kind of making that argument kind of very generally, saying, like, "Look, we're seeing, if anything in the data, we're seeing an uptick in software engineering."
5. 39:26 – 43:08

   Human employees would be hard to integrate into the machine economy

   1. DPDwarkesh Patel39:26

      We're talking a second ago about why there isn't more automation as a result of LLMs, and one plausible mechanism could be that, as you were saying with the O-ring, uh... So O-ring theory refers to this fact that the Challenger shuttle blew up because, uh, there's, there's one component that malfunctioned, and it destroyed the whole thing. And maybe that's a g- more general model of how goods are produced in the economy, that you gotta make sure everything is reliable and works well, and you can't automate entire job to an AI right now, um, even though it might be able to perform it at some probability. You need extreme reliability in order for it to not destroy the finished good. Um, I think this is-- Yeah, this, this might explain why there's a lot less automation now than there otherwise could be, but I think it works in the other direction once AIs get advanced enough that integrating humans into the production flow of future goods, even beyond the, uh, even beyond the arguments about how humans will be more expensive or dumber or whatever. Even beyond that, just there will be whole production flows that are d- uh, organized for AI labor, where they're talking in neuralese, they're thinking many thousands of times faster. So even if there's some comparative advantage where it makes sense to hire a human, there will be, like, transaction cost and worries about reliability that will actually make it hard to integrate humans into future production flows.

   2. DPDwarkesh Patel40:41

      Yeah. That, that seems right to me. In particular, I just wanna distinguish between the, the, the point that if you automate, like, nine-tenths of a job, then people might kind of shift over to the last tenth, but, like, there might be ten times more work-

   3. DPDwarkesh Patel40:55

      Mm-hmm

   4. DPDwarkesh Patel40:56

      ... demanded of them from the model of O-ring automation from, like, uh, Gans and Goldfarb recently, which was that if you can only automate nine-tenths of the job but you can do it to a lower standard of quality than the human could do it, you might not want to automate even-

   5. DPDwarkesh Patel41:10

      Right

   6. DPDwarkesh Patel41:10

      ... those nine-tenths. And that's the thing that could totally port over to, um, uh... Like, the, symmetrically, it could be a reason why we don't use a human for one-

   7. DPDwarkesh Patel41:22

      That's right

   8. DPDwarkesh Patel41:22

      ... tenth of the job anymore.

   9. DPDwarkesh Patel41:23

      Yeah.

   10. DPDwarkesh Patel41:23

       Because a human just can't perform it to the level of quality that the AI can perform the other parts of the job.

   11. DPDwarkesh Patel41:28

       Yeah.

   12. DPDwarkesh Patel41:28

       Or the level of speed or whatever, and they end up pulling down the, uh, quality or speed of the finished product.

   13. DPDwarkesh Patel41:34

       By the way, the model you're talking about seems extremely plausible to me of why more lawyers or accountants or whatever are not automated. Like, there are cases in-- or even software engineers, where there's a pretty good probability that the thing worked as you expect, but the, the thing you're paying a lawyer for is like, "No, really, my company's not gonna go under because, um-"

   14. AIAlex Imas41:52

       You're also paying for a lot of, like, regulation-type stuff.

   15. DPDwarkesh Patel41:55

       Right.

   16. AIAlex Imas41:55

       Right? So, like, with lawyers particularly, you need some entity to back up the product. You need kind of like an ownership of the product. You need somebody to be able to fire or, or, or hire, like licensing issues. There's a lot of like, sort of like regulatory layers that are, like, also going to be keeping, even if there's no relational element, human in the loop that have nothing to do with, like, the ability of the human to actually perform the service.

   17. DPDwarkesh Patel42:23

       Yeah.

   18. DPDwarkesh Patel42:24

       Yeah, I mean, you know, all of these frictions on, um, the political-type decisions that we are accustomed to only trusting humans, you know, only having humans for, like legislation and being a judge, being a jury, or all the licensing that keeps certain professions human, that all strikes me as transitional, right? I mean, uh, what we expect to come from a human and, like, how we organize our politics, that's changed so many times throughout history, right? From little hunter-gatherer bands to empires-

   19. DPDwarkesh Patel42:55

       Right

   20. DPDwarkesh Patel42:55

       ... to whatnot. AndYeah. Once an AI-run political system is much more efficient than the alternatives, then those will probably tend to outcompete the others and, you know.

   21. DPDwarkesh Patel43:07

       So speaking of

6. 43:08 – 1:01:28

   What if some humans (or AIs) value wealth accumulation intrinsically?

   1. DPDwarkesh Patel43:08

      which, we've been talking about what preferences humans currently have and what impact that has on what kinds of goods will be scarce in the future. But of course, we'll have different kinds of entities in the future, uh, AIs, right? There was a time when there were no humans on Earth, but evolution selected for agents that have specific drives and preferences because those tend to survive the most, and those preferences now basically determine how a hundred trillion dollar world economy, uh, what it produces. And so why not expect the same thing of AIs in the future? This is not, this is not even a world with, um, catastrophic misalignment. That is to say they just kill everybody. But there will, there will be evolution of even, even if not individual AIs, then firms which have AIs as part of them. And what will that evolution favor? Well, it will fa-favor probably firms or agents that grow, right? There's, there's like a selection argument that things which grow will be more prevalent. And maybe just based on that, you can make some predictions about what their preferences will be. But is the kind of entity which prefers to have human intrinsic goods going to be the kind of entity that accumulates resources the most? Probably not, right? Probably it like saves more. It like has unsatisfiable demand for things like whatever the relevant resource happens to be. Compute is an obvious one. Um, and can we use that to make some prediction about what, m-m, the non-human preferences that will be guiding the future?

   2. AIAlex Imas44:34

      Yeah. So I think if there's like an AI that's like has its own welfare and it's, it's fully autonomous and it's like making its own decisions and that they're, that are welfare relevant, to be honest, I, I, I have absolutely n-no prior that they-- it would like at all prefer hu-- other hu-- to, to like deal with humans.

   3. DPDwarkesh Patel44:51

      Right.

   4. AIAlex Imas44:51

      There's like no reason. But let's, but, but let, let me take like a qui-- uh, the other side of that argument. Will humans' preferences to be interacting with one another and to trust and empathize and all of these sorts of like things with other humans versus a simulated AI? I think it's a really important question whether those will change, right? So I've heard a lot of arguments saying like, "Look, you know, right now we're just not used to the technology," and at some point, like what you're thinking of r-relational or something like that, people are just gonna see like an AI therapist as su-a superior product.

   5. DPDwarkesh Patel45:26

      Right.

   6. AIAlex Imas45:27

      And they're not gonna need the sort of like empathy or whatever-

   7. DPDwarkesh Patel45:30

      Right

   8. AIAlex Imas45:30

      ... that the human is pr-- uh, providing. I think this is actually a really complicated question. Um, here's one argument for why it's not gonna go away, and that, that has to do with evolution. So let's say there's two types of people. One person doesn't really have this preference. They ju-- can just interact with other AI, whatever can simulate it better. The other one has almost like a, like a moral emotion, like from the, using Jonathan Haidt's framework, moral emotion against interact-- like offloading those sorts of social interactions to an AI. Which of those two people are going to reproduce, find a mate, all of these sorts of things? I think the answer is kinda clear, right? It's the second one that has the preference for other people.

   9. DPDwarkesh Patel46:10

      It depends on how the reproduction is happening.

   10. AIAlex Imas46:12

       Fair.

   11. DPDwarkesh Patel46:12

       Yeah.

   12. AIAlex Imas46:12

       But if, if we're, if we're in, you know, the world where like, um, reproduction is still happening the way that it's happening, I think... And this is a big question. I'm not even like, I'm not making a prediction.

   13. DPDwarkesh Patel46:22

       Right.

   14. AIAlex Imas46:22

       Again, I'm just saying like if we're thinking, you know, you had David Reich on, on the show, like his point on the last podcast was that, you know, we're buzzing with natural selection.

   15. DPDwarkesh Patel46:32

       Right.

   16. AIAlex Imas46:32

       So even if like you get some sort of indifference now, you might get selection to point into like an even stronger preference for other humans.

   17. DPDwarkesh Patel46:40

       Yeah. Here's one way to think about it. How is the wealth of the richest people in the world instantiated? Of course, they can, um, m- as you were, we were having a call earlier, and you were making the point that their consumption is more geared towards relational goods, like Mark Zuckerberg is hiring MMA instructors and dancers for his wife's birthday and so forth. But most of his wealth is just stock and meta, and he, as a controlling shareholder, could say, "Hey, Meta, just give me all this income," uh, or, "Turn all this wealth into dividend income, um, and I will just spend that on consumption." But instead, he rather would have his wealth compound, um, and Meta to build more data centers, basically. So you, you don't even have to change humans for this to be the case. It is just the case that the hu-humans which are wealthiest and are growing wealthier because their wealth is compounding, just have this like almost Nick Landian preference for like accelerating capital. Um, and that, that does seem to suggest that, yeah, is, is that an important determinant of what kinds of things are produced in the future?

   18. DPDwarkesh Patel47:41

       Yeah. Could kind of just say like there's two ways you could get the two kinds of people, one of whom prefers a human therapist and one of whom is fine interacting with the AI. If they both satiate equally quickly in capital, right?

   19. AIAlex Imas47:53

       Mm-hmm.

   20. DPDwarkesh Patel47:53

       But the one who likes the human therapist just also likes having some human intrinsic, uh, services, um, then the marginal value, like how the marginal value of capital in the future for, compares to the marginal value of capital today for each of them, if they start out equally rich, should be basically the same. I mean, there could be interactions and whatnot, but bas-basically, that should be the same. If what's driving the difference is that one person just doesn't satiate in capital because they're, they're engaged by the prospect of, you know, exploring the universe-

   21. DPDwarkesh Patel48:28

       Right

   22. DPDwarkesh Patel48:28

       ... and turning their head into a galaxy brain or whatever, and the other one satiates-

   23. DPDwarkesh Patel48:31

       Yeah

   24. DPDwarkesh Patel48:32

       ... then the person who doesn't satiate in capital is going to, if they're being rational, they're gonna have a higher savings rate.

   25. DPDwarkesh Patel48:38

       Yeah.

   26. DPDwarkesh Patel48:38

       Okay? So in the long run, they're going to have most of the wealth, and the overall capital share will basically be the capital share of that person's spending, which is-

   27. DPDwarkesh Patel48:48

       Right

   28. DPDwarkesh Patel48:48

       ... gonna be one.

   29. DPDwarkesh Patel48:49

       It's important that this is-- we're not talking about a hypothetical future.

   30. DPDwarkesh Patel48:52

       Yeah, yeah.
7. 1:01:28 – 1:16:07

   What should developing countries do?

   1. DPDwarkesh Patel1:01:28

      Do economists have any advice for countries which are not in the AI production chain? If you're, if you're not either producing the AI models, you're not producing, um, the hardware that goes into AI models, if you're not Korea making HBM or Taiwan making, um, with the FAS or not, um, uh, Netherlands with ASML. Like, what is India or Nigeria, what should they be doing right now?

   2. AIAlex Imas1:01:50

      Yeah.

   3. DPDwarkesh Patel1:01:50

      If you're talking to Modi right now, what do you say?

   4. AIAlex Imas1:01:52

      I, I think the biggest lack of resources that we have allocated in the economic profession is thinking about middle income developing countries in the, in the age of AI. And, I mean, this is, this is my fault, you know, this is something I fault myself with, uh, as well. There's not enough people thinking about this question. Like, there are scenarios where, you know, you get, like, AI technology, um, you know, uh, uh, being allocated and dissipating to, to Nigeria and, and developing countries and things like that, and, like that, uh, leveling the playing field, like essentially, like giving them a, like a level up as far, as far as capabilities. Um, but there's another world where, like, because they don't have enough resources, they're not making, uh, they're not training the models, they don't have the hardware, where they just completely get left behind.

   5. DPDwarkesh Patel1:02:36

      Yeah.

   6. AIAlex Imas1:02:36

      And because of, you know, um, automation, we can produce commodities in developed countries now.

   7. DPDwarkesh Patel1:02:43

      Right.

   8. AIAlex Imas1:02:43

      Then we don't even have, you know-

   9. DPDwarkesh Patel1:02:45

      Yeah, exactly

   10. AIAlex Imas1:02:45

       ... the consumer market, and then that, that world looks pretty, pretty bad.

   11. DPDwarkesh Patel1:02:49

       Yeah. This seems to me like an extension of the Messy Middle case, right?

   12. AIAlex Imas1:02:54

       Mm-hmm.

   13. DPDwarkesh Patel1:02:54

       Um, one of the ways in which the Messy Middle might, uh, only be bad in a narrow range of scenarios isn't just that, like, it would be easier to redistribute because the pie would be bigger, but because, um, the interest rate would be way higher and/or sort of equivalently the price of everything except the human intrinsic goods would be, would be falling really rapidly. Uh, sort of two sides of the same coin. Um, a little bit of savings would turn into a lot of consumption next year-

   14. AIAlex Imas1:03:20

       Mm-hmm, mm-hmm

   15. DPDwarkesh Patel1:03:20

       ... right? So things have to go really wrong for us to, like, just get over the threshold of, uh, you know, capital being productive enough to automate lots of work, but not be productive enough that, that the interest rate is high and/or the price of capital-produced goods is falling a lot, okay? So even without redistribution, a little bit of savings will save a lot of people. And sorry, you're saying if the developing countries have some savings-

   16. AIAlex Imas1:03:41

       Yeah, yeah, yeah

   17. DPDwarkesh Patel1:03:42

       ... i- in the developed world, that will be enough to produce a lot of surplus that they can then-

   18. AIAlex Imas1:03:46

       They will now be able to consume a lot-

   19. DPDwarkesh Patel1:03:47

       Right

   20. AIAlex Imas1:03:48

       ... using their savings.

   21. DPDwarkesh Patel1:03:48

       So, so but, but, but I mean, uh, the Messy Middle could be, like, wider in this ca- I mean, they're starting from such a lower level-

   22. AIAlex Imas1:03:55

       Right

   23. DPDwarkesh Patel1:03:55

       ... in terms of, like, how much-

   24. AIAlex Imas1:03:57

       They've saved and, uh-

   25. DPDwarkesh Patel1:03:57

       ... how much they have and how, how much it's, like, actually indexed to the global economy.

   26. AIAlex Imas1:04:00

       Right, yeah.

   27. DPDwarkesh Patel1:04:01

       Um, and I think it's important for them to get on it now, and I don't have strong feelings about whether it should take the form of, like, sovereign wealth funds that invest in-

   28. AIAlex Imas1:04:09

       Right

   29. DPDwarkesh Patel1:04:10

       ... the, the right supply chains or, uh, or just, you know, subsidies to their own citizens to buy a little bit of-

   30. AIAlex Imas1:04:16

       Th- this, this is actually, I think, a crucial point. We were talking earlier about why the Rockefellers or whatever of the world, why their descendants don't control everything if our argument about the selection of these ki- kind of greedy optimizers hold. And one argument is just that it's, like, very hard to index the economy, and maybe they would have just decided to in- have their heirs index the economy and have it grow at the rate of economic-- have their wealth grow at the rate of economic growth, and they would be, you know, trillionaires. Their heirs would be trillionaires by now. But it just histor- uh, before index funds existed, it's just very hard to just get, um, get it represent-- Uh, it, it just a very small fraction of the economy going back a hundred years accounts for a majority of the value created now, and if you miss those particular things, you would have basically, uh, your wealth would have just kind of stagnated. Um, and maybe there was a brief golden window from the creation of index funds up until, m- I don't know, five years ago, where actually you could index the economy and you could have your wealth grow at the rate of the economy grows. But now that we're in this world with very concentrated returns, especially to private companies, which is capital that is, as we were making a point in our blog post, um, the average person has disproportionately less access to, uh, as opposed to, you know, s- most of their capital is, like, having a random house, at least in the US.

Episode duration: 1:16:08