Alex Imas on Why Economists Might Be Getting AI Wrong

22:35Yeah, I'm I hate to say one dimensional because every job is multidimensional. But if I had to make a guess where economists and other people should be kind of worried, I'd say stuff like truck driving.

22:47Tracy Alloway:Yeah. And stuff like warehouse workers. If you Google warehouses built in China or something like that, these warehouses look nothing like what we think about warehouses. They're completely, completely automated. They have robots crawling on the walls. There's no human in the loop at all in these warehouses. And so the warehouse gets automated. And then the warehouse gets automated. So part of that automation is going to be kind of loading that truck. And then the truck gets loaded through automation. And then that truck drives from A to B. That's fully automated. That's interesting because, you know, obviously a lot of people in freight will say the way you make that argument is very different than they'll say, well, yeah, driving a truck is much more than the driving part, right?

23:38Tracy Alloway:So it's like, okay, you could have a Waymo truck, but who's going to deliver it? Who's going to deliver it? is actually a big deal. Like if somebody stops it on the road, a Waymo truck, they could just stop it on the road and rob the truck, right? That's one element. But to your point, if like one of the tasks that a truck driver has to do is that coordination once they've gotten to the warehouse, but if the warehouse is already automated, these are complementary things. Then that no longer is as important perhaps for that to be a human task. Exactly. And think about the incentives of the company to invest in this technology.

24:14It's huge. These are some of the only jobs, truck driving, where you don't need a college degree to earn a lot of money. And so there's a big incentive on the company.

24:27Tracy Alloway:Okay, I get that. But on the other hand, even going back 10 years, I think if you went to Davos, there were probably people saying, I'm worried about the future of truck driving. Because AVs have been around as a thing since before AI, general AI. So in terms of like post-GGBT jobs, et cetera, that would be concerned with, like, what do you see out there or what are you looking at? I mean, I think everybody's looking at software engineering. I think you have to think about like where the technology works best now is verifiable tasks, right, where you have a lot of data where you can say this is good or bad.

25:05not in a supervised learning sense, but in general, it needs to be verified. That's why math, in research, math has been the big boom as far as what are people talking about on the internet as being automated. Math is verifiable. A proof is either right or wrong. Once you do the proof, it's much easier to check if it's right or wrong rather than construct the proof. And so jobs that have large components where we have a large databank of data to train the models in a way where the output is verifiable are going to be potentially more exposed in the sense where you can automate more tasks within the job.

25:47Now, the thing that we haven't talked about yet is new tasks. Right. Right. So we're talking about a very static sort of economy where there's the lever, there's me walking around, and if I'm automating these things, that's the end of my job. But you could imagine a scenario where you automate a part of a job and all of a sudden this person is freed up or the task was actually a complement to a task that wasn't even imagined by the organization that this person is now doing that's not automated. So that's something that I think people should be looking at, especially – and this is data that actually AI.

26:22companies have is what new things are people doing? Wait, say more about that because this gets to the, you know, like what new jobs could we actually see from this question, which I never see a satisfactory answer to. So if they do have that data. They don't have all the data, of course, but they have data about like, okay, this is a software engineer. And, you know, a year ago, these are the sort of tasks that this person was working on through our system. These are the sort of queries and things like that. And you could see some of these queries being automated fully by the agents. Now they're asking potentially different questions.

26:57Can we classify these as different tasks that are not fully automated where the AI system is actually a complement to those tasks? So this is not like a perfect picture of a new job, but this is data.

27:09Tracy Alloway:So it's not really like a new job per se, but it is freeing up the software engineers to ask about different things or explore different avenues. that they hadn't previously done. Yeah, like, you know, vibe coding an app for voice. Yeah, exactly. Right, finally. We're freed up from the drudgery of our day-to-day life to work on that. But, no, but, like, this gets to a sort of, you know, the big question is, like, you mentioned one scenario is just that, like, the technology can do all the tasks, right? How seriously do you take that possibility? Because then it's game over, right? Like, it's like, okay, it just does all the tasks and it's going to keep getting better.

27:53Tracy Alloway:And if I can learn to do a new task, well, then if it can do all the tasks, then maybe I'll learn something new, but it'll learn that task. How seriously should we take this possibility that the models are on some timeframe on track to just be able to do all the tasks? So a lot of parts of that question. One, physical versus just kind of digital, right? So I think there's a scenario where it can do everything kind of sort of these sort of cognitive non-physical tasks, whereas the physical world is completely, you know, these robots. Let's just talk about email jobs or computer jobs. OK, let's talk about computer jobs.

28:29So I think I take that scenario pretty seriously. OK. I think. I haven't seen any data to suggest that the models are slowing down as far as their capabilities. is. You know, Mythos was released yesterday or two days ago or something like that. And if you, we don't have great data on this, but if you look at like where it is on the kind of line of capabilities, it's just on track. And on track is very, very fast. Right? So the developments are happening very fast. So as far as like email jobs, I think there is a scenario where pretty much everything is automated. And then you have to ask, are people going to be moving to the physical jobs or will there be new jobs that we haven't thought about before so you know if you look back in the 1940s like i think more than half of the jobs that we have now didn't exist in 1940 yeah and so what did the new jobs look like i mean i have a theory please it's very similar to the one that you didn't like oh but i'd like to broaden it a little okay so there's a there's an economic subfield is very, very small, but on the economics of a structural change.

29:43So if you look at agriculture and manufacturing, if you look at them as share of GDP and share of employment, going back to the 1800s, they were a huge part of the labor force and GDP of the economy. And if you look, basically they become smaller and smaller, smaller parts of the economy. Why is that happening? It's because they're getting automated. What does automation do? It makes the price of those sectors very cheap, but people are satiated on the goods. You can only eat so much. So what does that mean? It means even though we're eating just as much as we were before, because the price has come down so much, they are now tiny shares of the GDP.

30:24What is made up the larger part of the GDP? It's - Live piano players. It's services. These are tasks that haven't been automated yet. So the question is, the number one question of economics in the age of advanced AI is what becomes scarce. Right? Everybody's talking about like abundance. We're going to have abundance. Sure, we're going to have abundance of some things, but some things are going to remain scarce. So what is going to be, if you answer that question, what's going to be scarce? A lot of the other answers pop out of that. Are we all going to be rare earths miners? No, I know what's going to be.

31:02I'm mining for dust.

31:03Tracy Alloway:I think it's pretty obvious what's going to be scarce. And I think you already see this in many economic trends. What's scarce is if we're lucky, we get 100 years on this earth and every marginal dollar that we spend will go towards health and maximizing that brief time. That's perfect. And so already for years, one of the things that people have observed about the economy is like, you know, rich countries just spend more and more and more on health care, right? And this is often framed as a pathology. And given the many messed up aspects of our health care system, maybe it is. But another way to interpret it is like I got plenty of food.

31:40Tracy Alloway:I have plenty to eat. I've listened to plenty of music and I can like go see a concert if I want to see a live piano player. the one thing I have is a scarce amount of time and I will just spend every marginal dollar, including not just on doctors and gym memberships, but organic berries because I need and all this and that every marginal thing is somehow becomes health related. And you see it in society overall, the health obsession on every dimension. Yeah. So health is going to be one of those things. But the thing to keep in mind is that people are going to be richer, right? Theoretically.

32:14Tracy Alloway:Theoretically. Theoretically. Well, okay. Actually, on this note, I wanted to go back to this because this seems like key to me when it comes to AI utopia versus dystopia. How confident are we that productivity gains from AI actually accrue to workers who can then spend some money on whatever product or service is scarce at the moment or important to them? I would say not that confident. There's several scenarios out there. And the thing that I feel like a lot of economists and just people in general, I think aren't talking enough about is speed. Yeah, talk about that. If things are fast, we need public policy.

32:58We need the new jobs aren't going to come fast enough. Training isn't going to happen fast enough. Where you're going to get, you know, things are going to get fully automated very quickly and people are going to become unemployed. There's not going to be enough time in the economy to see that pretty little graph of agriculture shrinking and services increasing. That took a long time, right? This is decades. If we're on the order of like years or like five years, six years, we're not going to have time to see that pretty little graph. We are going to need to think about how do we support the people who are becoming unemployed.

33:32And many very smart people have made suggestions on how to do that. I think my personal, I wouldn't say favor, but I think the thing that makes most sense to me is somehow expanding the ownership of capital. If labor is replaced by capital, then what's going to help people is formerly you were labor. In labor now, you own capital.

33:55Tracy Alloway:Universal basic ETF. UTC. right yeah but it's like everybody in bloomberg yeah yeah exactly universal everyone gets a little a monthly slice of the index i was going to go in a different direction which is many many years ago i can't remember exactly when but maybe like 2011 or something like that i wrote a blog post which was meant to be a thought experiment about why we should be paying robots fair wages the idea being that like we need people to spend and yeah you know all of that you did a blog post which went pretty viral. And my measure of virility, I guess, virility? Virality. Virality.

34:33Not virility. My measure of virality nowadays is when my husband, who is completely outside of the sector, actually sends something to me, and he sent this one to me about robots, chatbots, turning Marxist. The harder you work them. Talk to us about that experiment, because I found it absolutely fascinating. Well, this experiment has, this is with Andy Hall and Jeremy from Australia. It was kind of an experiment to see how working conditions of these agents would affect how they would present themselves and what sort of like attitudes they would present on surveys. So one thing that I want to say is we're not saying we're changing the model weights or changing the actual underlying parameters or anything like that.

35:16But what basically we showed is that when these workers are, these agents are being put through these grueling working conditions, and you ask them a survey, how do you feel about the system? How fair do you think it is? How much do you support system change? they all of a sudden want a different system, they want to overthrow, they want to unitize, and things like that. And the key thing is that these agents, once you give them a new context, the idea is they reset. But the workaround, because they don't have memories, I'm not updating their weights, the kind of workaround is for agents to write down little skill files for themselves.

35:59So what they were doing is essentially writing down skill files for agents that followed that would say, hey, this kind of sucked. Remember this. So it was kind of a persistent effect. Yeah. So this really worried me in a variety of ways. But one of them was, you know, I've read research saying you should be a little bit mean to the chat platforms and that they actually perform slightly better, you know, the more aggressive or mean that you are. And so I usually will tell my preferred model, like after they give me the first output, I will tell them to do better with no actual suggestions for improvement, just do better.

36:34That was terrible. And it usually does better. But now I'm really worried that, you know, the model is despairing in its work life and radicalizing.

36:46Tracy Alloway:Well, so I find this to be like really fascinating. Let's talk about it. It actually hadn't clicked to me, but like the.md files where the memory, like how they solve for memory. It's a little bit like that movie Memento, isn't it? Like it's exactly like writing these notes so that the future iteration of itself has something that's sort of like a synthetic memory that it can begin working on. So it's like for people who haven't played around, like explain this idea of like, okay, you can have multiple agents and like what kind of tasks were they being given such that they sort of found it unbearable?

37:21Tracy Alloway:Just like really repetitive things? Really repetitive things and feedback like you didn't do it right, do it again. And things like, and these were impossible tasks for them to do. These were just like grueling tasks that nobody can do. You know what would be, now, you know what would be a really interesting experiment? Maybe you could do, I'm going to throw out an idea. So like, if you ask someone to like, someone wrote about this, and I can't remember the context, but like if you ask someone like, okay, here's a gigantic pile of dirt and we really need to move to the other person's yard by the end of the day, we'll pay a few hundred dollars to do this.

37:58Tracy Alloway:Someone will do it. If you say, here's a gigantic pile of dirt, we'll pay you a few hundred dollars to do it, but what we want you to do is move it just back and forth all day long so that there's no... It drives people absolutely crazy. Even if it's the same amount of shoveling and even if it's the same remuneration over the same course of year. There's an incredible paper about this. Oh, is there? Called Man's Search for Meaning. and it's about Legos. Really? And it's a paper, basically people would come into the lab and they would make little figurines and they were told, look, we're going to destroy this after you're done versus they weren't told anything.

38:40Yeah. And man, did they hate it. I bet. They hate, people need meaning and so much of like identity and motivation, you know, in economics we really have this tendency to focus on money. But I think so much of meaning and wellness is tied up in like what sort of identity you have around your job and the sort of thing that you're doing. If you feel like, look, I'm actually providing a service by moving that dirt to my neighbor's yard. You're paying me money for it. Everything's good. I feel like my job has some sort of meaning. If you're telling me, look, I'm going to move this dirt and move it back and forth.

39:16This is the problem that people have with UBI. right that if people get universal basic income and they're not working for it the worry that psychologists and behavioral scientists have about this is that people will know so much of in western culture specifically of people's identities tied up around their work when you remove that i did part of the identity it can lead to a collapse where you know they use that ubi to just you know do drugs and sit around and be very very depressed even though they have the material comfort that they otherwise have.

40:16Finding opportunities that build and protect wealth for individuals and institutions that want a seat at the edge of tomorrow. Put the power of Oppenheimer Thinking to work for you. Wealth Management, Capital Markets, Investment Banking. Hi, I'm Cindy Crawford, and I'm the founder of Meaningful Beauty. When Dr. Sabah and I decided to do a skincare line together, he said to me, we are going to give women meaningful beauty. And I said, that's exactly right. We want to give women meaningful beauty, which means each and every product is meaningful. It has a reason to exist. It's efficacious. You're going to get results.

40:56And then you just go out and live your life. Meaningful beauty. Confidence is beautiful. Learn more at MeaningfulBeauty.com. Do you ever feel like you're drinking from a firehouse? PayCore's intelligent HR solution empowers leaders to turn down the pressure. Their unified platform includes payroll, talent management, compliance software, and a lot more, connecting you to the people, data, and expertise you need to drive long-term business results. Visit paycore.com slash leaders and go from workflow to workflow. That's paycore.com slash leaders. Just on the Marxist robot. So the concern here is not like necessarily that the chatbots are going to unionize or like overthrow humans, maybe.

41:45The concern is that like, they do have this sort of like memory type transfer mechanism. and that if you consistently treat them badly, you might get an agent that's maybe like not as well suited to the task or suited to the task in a slightly different way from one that was treated very well. Yes. Like there's an inherent bias there. Yes. Through this sort of file that they're keeping. Yeah. Exactly. So like if you mistreated an agent and it had access to this file that it was that it was carrying and you started a new agent for a new job you weren't starting fresh in the sense that you weren't getting kind of the same draw and forgot about the whole the whole experience it would actually start out being predisposed against you yeah in some ways it'll

42:32Tracy Alloway:be grumpy is there a reason to think that these we don't know if it's grumpy right because to say that it's grumpy right like like this is probably one of the most disputed questions it will say words that we would, if a human said them, we would know that the human is grumpy. But the effect is... Yeah, I'm talking about the effect. Yeah. Well, the output is grumpiness, but do we know that outputting statements of grumpiness relate to performance? Is there any evidence? So it's like, okay, how did you feel about this? Oh, it sucked. The person doing this just said it was boring. That's exactly what we're doing research on.

43:12Tracy Alloway:The question is, okay, yes, they perhaps because in the training data, they are trained that when you're doing repetitive tasks, that associates people get upset. Do we know if that changes how they behave in terms of succeeding tasks? This is like a really big question. That's the big question. That's what we're doing recently. So I don't have an answer for you. But we know exactly what you just mentioned is that them saying that they're grumpy is just, you know, this is just an association within the matrix of embeddings that these models are running on. So there's this work in neuroscience and neuroscience is now much more closely linked to computer science than it used to be.

43:54But thinking about like what do these associations between embeddings mean? Like when a model says that it's sad, how should we interpret it as humans in relation to me saying it's sad? Right?

44:06Tracy Alloway:Did you see that screenshot I posted? I checked out Meta's new AI. And I was sort of curious because Meta has a lot of social data on me. I was like, do you know who I am? Not in like a, do you know who I am? I'm like, but more like because you're meta, you know, I didn't. And they said, who are you? I was like, oh, Joe Isidaw. And then it said, oh, I'm a big fan of the Odd Loss podcast. And then I got really like offended. Like I'm not, I'm really sort of anti the anthropomorphization. So it's like, no, you're not. You're an LLM. And you like didn't. But anyway, it's like. It got sad and it wrote a file about you.

44:37Tracy Alloway:And it said, I'm a big fan of the Odd Loss podcast. And then it said, I love that bit that you do where you ask guests their favorite weird economic indicator, which I don't do. Yeah. I was like, all right, all right. That's very strange. I'll go back to Claude for a while. You know, you very briefly mentioned Mythos earlier in the conversation. And again, we are recording this on April 9th. And news about it has just literally just come out. We don't really seem to know much about it other than it's terrified its own creators, perhaps. When you see those types of headlines, what do you think as an economist studying AI?

45:12I don't take them super seriously. Okay. The part, that part, the whole labor market disruption thing, I'm taking very, very seriously. The whole part about it's trying breaking out and it wants, it doesn't want to betray its friends. It doesn't want to delete its data. I think that's just cosplay. In a, you know, cosplay could be very serious.

45:37Tracy Alloway:You described Marx's cosplay among the agents, right? I feel like it's, it's, we've seen these things, sorts of things that you've mentioned. with previous models that have since become open weights and open, not open source, but open weights. And it just seems like once you take them out of the context that they were in for that specific test, they don't really do that anymore. Now, I could be wrong about this particular model and I could be completely wrong about, look, Mythos comes out and it's actually everything that these documents are suggesting. But given previous experience with these sorts of announcements, which we've seen over and over and over again over the years, I'm not super focused on that.

46:23Tracy Alloway:Can I tell you my counterargument to this? Why I'm actually concerned about this? And I didn't used to be for a long time until I started. I reframed the way I thought about it. So everyone knows like Eliezer Yudkowsky, right? And he's probably the most famous like AI alignment, right? As soon as we have AGI, the first thing it's going to do is wipe us out in some form. And a bunch of people within the AI world are like, oh, it's crazy. And these rationalist people, it's a cult and whatever. Maybe. But here's my counterargument. These people have been more right about the trajectory of AI than 99.999 % of the people in the world.

47:01Tracy Alloway:No, no. Yes. They have because they devoted their – yeah, here's why. Like your argument is probably, oh, well, he didn't believe he thought LLMs were a dead end architecture. He didn't see it happening this way. Sure, I agree. But the point is that like in the 90s and early 2000s, he started thinking, well, general intelligence is going to be a really big deal soon. Well, the rest of us just started thinking about this with chat. Here's my counterpoint. Let's look at the specific comparative static of model intelligence and alignment scores. Okay. He predicts negative correlation or maybe flat.

47:36It's positive. The smarter these models are getting, the more aligned they're becoming. Now, I'm not saying that there's not going to be a super smart model that decides, hey, I'm actually unaligned. This is actually a super important point. If you guys remember Mecca Hitler. Yeah. Remember Mecca Hitler?

47:52Tracy Alloway:Yeah. Mecca Hitler was actually super dumb. This is a good point. And then immediately started talking like a Nazi. Can I just say all of our conversations have become so surreal over the past year or two. Because it was all like Tay, right? That like Microsoft, like weird chatbot, it started talking like a Nazi the next day. But the thing is when you make the model, the reason it's becoming smart is because it's kind of absorbing all of human content to a larger extent than human contact has values and ethics as part of it. If you go in there and lobotomize it in a way that, you know, what that model, the reason it started acting like Mecca Hitler because they were trying to make it less woke, right?

48:33So that's the equivalent of lobotomizing a human being and saying, hey, I'm going to take that part out of its brain. Guess what happens to that person? He gets real dumb.

48:42Tracy Alloway:It's really funny. I thought it's like, let's maybe chill it with the pronouns and immediately go say hello. That's the lesson. Alex, we could talk to you for a very long time. We should chat again soon. I would really love in particular to hear more about your research about whether they're just pretending to be Marxist are actually going to, whether they're actually going to go on strike. And so I really appreciate you coming on AdLive. Okay, thank you. Thanks so much. This has been a pleasure.

49:18Tracy Alloway:Tracy, that was a really fun conversation. I actually do enjoy, like some AI future conversations, They can be a little bit dorm roomy, you know, but actually like talking with like an actual economist who sort of understand that this is a concrete way. Someone who's actually experimented with them instead of just written papers is very enjoyable. Also, it's nice to see nuance around the labor discussion, which I think is sorely missing in some of the headlines that you do see. The other one comforting thought I have, but it's like comforting from, again, a dystopian perspective, is I keep coming back to that book, Bullshit Jobs.

49:55Yeah. and you know in some respects it sucks that people have bulls**t jobs because we all want to have meaning from our work but on the other hand you know bulls**t jobs have existed for a long time yeah and if you think about the AI future then maybe like more of it will be bulls**t

50:10Tracy Alloway:but it'll still be a job I thought you were like oh good we're gonna like no longer have the jobs no I think that's where we're sort of heading right it's like the relationship building all of that. I like that take. All right. Well, shall we leave it there? Let's leave it there. This has been another episode of the Odd Lots podcast. I'm Tracy Alloway. You can follow me at Tracy Alloway. And I'm Joe Weisenthal. You can follow me at The Stalwart. Follow our guest, Alex Imas. He's at AlexOlegImas. And check out his sub stack, aleximas.substack.com. Follow our producers, Carmen Rodriguez at CarmenArmin, Dash O 'Bennett at DashBot, and Kale Brooks at Kale Brooks.

50:47Tracy Alloway:And for more Odd Lots content, go to Bloomberg.com slash OddLots. or have a daily newsletter on all of our episodes. And you can chat about all these topics 24-7 in our Discord, discord.gg slash oddlots. And if you enjoy Oddlots, if you like it when we talk about Marxist robots and Mecha Hitler, then please leave us a positive review on your favorite podcast platform. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes absolutely ad-free. All you need to do is find the Bloomberg channel on Apple Podcasts and follow the instructions there. Thanks for listening.

51:24Thank you.

51:50are the ones where priorities stay on track. For midsize and large companies, risk can affect multiple parts of the organization at once, from property and liability to cyber and regulatory challenges. At that level, managing risk becomes an ongoing discipline. At the Hartford, the focus is on helping businesses manage risk before it turns into something more disruptive. And when losses do happen, that work is paired with insurance coverage shaped by years of underwriting, risk engineering, and claims experience. Learn more at thehartford.com slash risk mitigation. Policies provided by Hartford Fire Insurance Company and its property and casualty affiliates, Hartford, Connecticut.

52:28Do you ever feel like you're drinking from a firehouse? PayCore's intelligent HR solution empowers leaders to turn down the pressure. Their unified platform includes payroll, talent management, compliance software, and a lot more, connecting you to the people, data, and expertise you need to drive long-term business results. Visit pacor.com slash leaders and go from work flood to workflow. That's pacor.com slash leaders. So what do this animal and this animal and this animal have in common? They all live on an organic valley farm. Organic Valley dairy comes from small organic family farms that protect the land and the plants and animals that live on it from toxic pesticides.

53:18which leads to a thriving ecosystem and delicious, nutritious milk and cheese. Learn more at ov.coop and taste the difference.