Nobel-Prize Winning Economist Daron Acemoglu on Musk, AI and Democracy — The Mishal Husain Show

[00:00:00] Daron Acemoglu: We need to be radical in the way that we need to think about economic solutions, about AI. I think this is not a time to be timid and say, oh, well, let's wait for another five years and see what Elon Musk and Peter Thiel and Mark Zuckerberg and Sam Altman come up with. [00:00:14] Speaker 2: Daron Asimoglu, Nobel Prize winning economist who wants to save the center left from itself. Is there really no one that you'd point to as that knight in shining armor? [00:00:25] Daron Acemoglu: No, not in the knight in shining armor. I mean, in the US, I can point to better models as opposed to less good models. [00:00:31] Speaker 2: But not the democratic socialists, I think, not the Mamdani, Alexandria Ocasio-Cortez group. [00:00:36] Daron Acemoglu: Well, look, again, they divide. I think they have some good ideas, but they divide. [00:00:40] Speaker 2: From Bloomberg Weekend, this is the Michelle Hussain Show. We're in the midst of a pretty intense period right now for the democratic process in a number of parts of the world. The US midterms are upon us. We've just had elections in Brazil, Serbia, Israel. There's unrest in France that shows the pressure leaders are under to deliver for the people, especially the young. And in the UK, there's been a budget which shows how the government there is grappling with how to deliver amid economic realities. Against that backdrop, there is no shortage of theories on what politics could or should do better. But the one that caught my eye and led to this episode is from a new book by a Nobel Prize-winning economist who drills right down into the actions he'd like to see politicians take. He is the MIT professor, Daron Acemoglu, whose latest book is called What Happened to Liberal Democracy? Remaking a Politics of Shared Prosperity. And he tells me from the outset of this conversation that he's from the center-left. What he's really interested in is what he calls the left-leaning corner of the big tent of liberalism. What he means by that is the kind of politics that led to Roosevelt's New Deal or social democracy in Europe. He has got plenty to say about AI, tech titans, algorithmic social media, and even The Economist, which recently had a go at him. What I found really compelling is that he is a celebrated economist known for work on institutions and shared prosperity and automation versus human labor. And he feels strongly enough about our times to come out with what is essentially a political prescription. And that's what you'll hear. Remember, you can read my notes on this episode, as always. It's where you'll get the added context and a sense of my reading list before and after we recorded. There's a link in the show notes. And let's dive into this, exactly at the point that Daron Acemoglu settled into his seat here in the studio. So, I'll tell you one thing I wondered about. I mean, congratulations on the book, obviously. I wanted to ask you about the dedication, Irma and Kervork. Yes, those are my parents, yes. Your parents, who I guess are no longer alive. They are not, yes. To Irma and Kervork, with whom I would have so much liked to share this book. It is your first solo book, right? Yes, correct. [00:03:26] Daron Acemoglu: Oh, well, I did write a textbook at the graduate level. [00:03:29] Speaker 2: Why particularly do you say you would have liked to share this book with them? [00:03:33] Daron Acemoglu: Oh, because that's my baby right now. When I write the next one, perhaps, I'll think, oh, well, this is the one I really like. [00:03:41] Speaker 2: Did they see you win the Nobel? Were they still alive? [00:03:43] Daron Acemoglu: No, no, they died when I was 19 and 21, so a long time ago. [00:03:48] Speaker 2: Long time ago. Gosh, you were very young, so you were still a student when you lost your parents. And they were still living in Istanbul at the time. [00:03:54] Daron Acemoglu: Yeah. [00:03:55] Speaker 2: You lost them within a very short space of time. [00:03:57] Daron Acemoglu: Yeah. Such is life. [00:03:59] Speaker 2: Yeah. Thank you for coming in to talk to us about the new book, which is clearly making waves. One thing I was curious about before we talk about the ideas you put forward, it's whether your heritage has played a role at all in your work on inclusivity and shared prosperity. Because I realize, of course, that you grew up in Turkey as part of a minority community, the Armenian community. Correct, absolutely. Is there a link? [00:04:22] Daron Acemoglu: There is definitely a link to my early work. My interest in dictatorship, democracy, and the economic consequences that they create, I think, was very much shaped by my early and late childhood, where Turkey was under a military dictatorship and was having lots of political, social, and economic problems. [00:04:44] Speaker 2: You were about 13, I think, when the coup happened. [00:04:46] Daron Acemoglu: I was about 13, yes, exactly. And especially when I was in middle school, so between the ages of 13 and 16, the military presence was very palpable. So those really were shaping experiences for me. I think once you start developing your own ideas, so from, say, the time I finished my PhD, the next three, four years, I think that formation probably was also influenced. But after that, I don't know whether it has had as much of an impact, but perhaps it has. I don't know. [00:05:22] Speaker 2: That period of Turkey in the 1980s, the tensions, I think you felt them more as a minority community. Your father ended up leaving his job at the university, and you said that that played a role. [00:05:33] Daron Acemoglu: Correct, absolutely, yes, I did. I was fortunate. I was in a reasonably well-off family, not rich at all, but after especially my father left the university, you know, we could at least afford stuff, but I felt the repression more, and there was animosity towards Armenians at the time in some quarters. So you could sort of feel that, but I never, you know, identified with any ethnic or religious group as a child. And that probably has played some role in me being attracted to liberal ideas that sort of emphasize and celebrate plurality. [00:06:14] Speaker 2: And you've lived a very international life, a student in the UK, and now you teach in the US. Further back, would your grandparents have lived through the period at the beginning of the 20th century? [00:06:24] Daron Acemoglu: Not my grandparents, but their parents, yes. [00:06:28] Speaker 2: So does that become part of the family story? [00:06:31] Daron Acemoglu: Yes and no, because it was a sensitive topic. [00:06:35] Speaker 2: In Turkey? [00:06:36] Daron Acemoglu: In Turkey. In Turkey. And as a result, often, even within families, you don't discuss the issues. So I didn't actually know about my family's history until one of my cousin's father wrote something about that history and shared it with me. So it was much, much later. I mean, I knew about the Armenian genocide and the issues surrounding it, but it wasn't discussed at home. [00:07:05] Speaker 2: And when you did discover, what did you discover about, was there a narrow escape and what was the impact on you learning about that time? [00:07:13] Daron Acemoglu: Yeah, it was, it was a narrow escape, but it was so long ago that I don't think it touched me as much as it would have done, for example, if I was 10 at the time. Yes, indeed, it was a narrow escape, indeed. [00:07:23] Speaker 2: How old were you when you finally put the pieces together? [00:07:26] Daron Acemoglu: I mean, I've started reading about the history of the Ottoman Empire and all of the events around then, when I was in, you know, my late teens, early 20s. But when I received that document and I read it, I was, I think, in my late, mid to late 20s. So it's quite a bit. [00:07:47] Speaker 2: Okay. So well, well into your academic career. [00:07:49] Daron Acemoglu: Yeah. So that's why, that's why it's like its impact was a bit less than it would have been had it been, you know, say 10, 15 years ago. 10, 15 years ago. [00:07:54] Speaker 2: I still don't understand that you've seen politics and economics as interwoven. Completely interwoven. And this is, you are a Nobel laureate in economics, but I think this book, What Happened to Liberal Democracy is a very political book and its ideas are absolutely relevant for our times. I wanted to ask whether it is a response to right wing populist politics, because you have Victor Orban right at the start of the book. And you're charting the fact that in 1989, he's a passionate Democrat. And by 2014, he's talking proudly about wanting an illiberal state. [00:08:24] Daron Acemoglu: Yeah. That's such a good beginning. Of course, it became a little bit more nuanced because as the book was going to the printers, he lost the election, which is great. But it did require a few sentences to be rewritten in the book just before it went to the press. But in some sense, the book is seeking an alternative to right wing ideas. And the aim is to contribute to the emergence of a new governing philosophy for liberal ideas, for liberal democracy. On the other hand, I do make an explicit choice of not discussing ethno-nationalism, alternative right, quasi-fascist ideas, or Trump is only mentioned once or twice in the book. That's because I want to put the spotlight on the center left and the core of liberal democracy itself, not its alternatives. And if you want to sort of think of an analogy, there's a debate about why Rome collapsed. And some people say, oh, well, because of the barbarians. But the barbarians weren't so mighty as Rome at its peak. At its peak, it was Rome's internal weakness that made it so vulnerable to the barbarians. So that's what I think about the citadel of liberalism, that it is its own weakness that has made it weak against these sordid movements. And that's why I want to put the spotlight and say, how can liberal ideas be updated and revived for our modern age? [00:09:53] Speaker 2: You want the center left to be better? [00:09:55] Daron Acemoglu: Much, much better. [00:09:56] Speaker 2: That's politically where you are. [00:09:57] Daron Acemoglu: Exactly. [00:09:58] Speaker ?: Yes. [00:09:58] Speaker 2: And you say that part of the reason that it has gone wrong, liberal democracy, the center left, liberalism in this context is self-inflicted because you point to identity politics and you say that the center left should never have got so caught up with issues. [00:10:14] Daron Acemoglu: Yes, but that's in some sense the upper layer. I want to start from how we ended up there. I see cultural politics or identity politics as a consequence of changes in the social fabric of post-industrial society. And it comes coupled with economic as well as political changes. And it is this ensemble that we have to understand. So post-industrial society made shared prosperity one of the key promises of liberal democracy, much harder because of digital technologies and globalization. You could now as a business person make a lot of profits, expand your market, but at the same time lay off workers because either you're automating their jobs or you're sending them to some cheap labor far away. In the last half century, we've seen college graduates, people like me, who have good education, well off social and material conditions. Living in urban centers become much more important in numbers and in social standing. And that has really changed political dynamics, especially at the center left, which used to get its majority of its votes from the working class. Now the college graduates in urban centers support the center left. The working class has shifted out of the center left to mild or sometimes hard right. [00:11:38] Speaker 2: But isn't some of that a natural form of progress? Why do we have so many college graduates, too many college graduates perhaps because people aspired for their children to go to university? And this was a sign of social mobility. This was a step away from the fields and the factories into the white collar work. I don't know what the alternative would have been. [00:11:59] Daron Acemoglu: 100% you are 100% correct. First of all, living conditions, working conditions are so much better for college graduates. Economic opportunities are so much better. And the economy needed college graduates. And college graduates also bring a number of additional benefits. But two things. First of all, I don't think it's either feasible or desirable to have everybody go to college. People have different talents, different interests. Second, while college graduates themselves became more and more better off, they also started living separate lives from the rest of society. So in the United States, 50% of the current cohorts are college graduates. That means another 50%, much less well off, are not. So who's going to speak for them? That is the problem. In the past, that question wasn't so alive. First, because the two groups were in the governing coalition together. And secondly, because they live in the same neighborhoods. They worked in the same firms. They had the same social networks. They married between where there was great cross married. And they had broadly similar values. All of these has changed. And that has created, I think, a very important part of the crisis of liberalism. And that's why I think cultural politics or identity politics, whatever you want to call it, isn't something that you have to, you should analyze in isolation. It has to be put into this context. [00:13:28] Speaker 2: So I imagine you're not saying that it's not worth investigating discrimination and trying to tackle it because that's where, you know, some of what is often dismissed as identity politics might come from. If you send out a CV with a certain kind of name in some societies, you might get rejected on the basis of that name. [00:13:45] Daron Acemoglu: Discrimination is real. And in fact, I emphasize, I have a full chapter on this in the book, that dealing with inclusion, intolerance, and how to actually educate people, ignorance, have always been dilemmas for liberalism. So liberalism started as an elite project by aristocrats and very well of very comfortable conditions. So all of these are real. And it was inevitable and socially beneficial that we should discuss these issues and try to improve inclusion and deal with certain aspects of intolerance. But the problem was that it took a particular top-down manner, which perhaps provocatively I call social engineering because it evolved into a form in which a group of socially powerful, sometimes economically powerful groups were trying to change the values of the rest of society. And that never works. [00:14:41] Speaker 2: Do you mean DEI? You're against DEI initiatives? DEI is absolutely. [00:14:44] Daron Acemoglu: Well, I'm not against the objectives of DEI, but I'm certainly against the way that DEI came to be practiced. [00:14:50] Speaker 2: I mean, you do have a lot in common with Donald Trump in that case on this front. [00:14:53] Daron Acemoglu: Well, no, no, because it's a big difference because he doesn't sympathize with the objectives. I sympathize with the objectives. And I think we need to work towards inclusion. I'm an immigrant myself. And I think integrating immigrants into society is key. But I object to the way that it got to be practiced and I object to the way that it took, the form that it took in particular, looking down upon the values of traditional communities, working class communities. And that I think, again, you have to understand in the context of post-industrial society. College graduates, once they become 50% of the population, they're not a uniform entity at all. There is such a wide difference between people who are doing so well in, for example, in technical fields, going and working in Silicon Valley or Wall Street, and people who are burdened by a large amount of student debt. There are people who have very right-wing ideas. There are people who have very left-wing ideas among the college graduates. But one thing united them, that was cultural issues because of the way that they themselves are cosmopolitan and the university imparts those cosmopolitan values. They were much more open to immigration. They were much more attuned with issues of minority rights, et cetera. So that's the reason why they rallied around that issue and ignored economic problems that are so important. Like in the United States, the UK has similar trends, but not perhaps as jarring as the US. But in the United States from 1980 to around 2013, 2014, every year, the real wages, inflation-adjusted wages of men without a college degree fell in the midst of a period where essentially the economy expanded quite a bit. That has no equivalent in history, so as I know. And at the same time, they were also told your traditional values are no good. You have to change them. You cannot use the language that that's not right. I think you can see how discombobulating that's going to be. And that backlash is exactly what we are seeing. So even if you agree fully, as I do, with many of the aims of diversity programs, that was the wrong way to go about it. And it's more importantly to me, it's also a way that's not consistent with the emphasis on individual freedoms and self-government. [00:17:11] Speaker 2: That should be at the heart of liberalism and liberal ideas, yes. Well, that is essentially your diagnosis of the problem. So obviously, I want to turn to your prescription, because you do have ideas about what you'd like to change, and you call it working-class liberalism. The UK has a new prime minister. I think he would very much be ideologically aligned with you. What should he do? Give Andy Burnham some advice. [00:17:34] Daron Acemoglu: Well, I think many countries, the United States ahead of the UK, but also the UK need more tax revenue. Increasing marginal tax rates, the rate that you pay as you earn your last dollar, has often costs, but we have a big opportunity to expand the base of taxation in the US, and to some extent also in the UK. We don't tax capital income enough. If you are a business owner, and you call the income you get from your company labor income, you pay a much higher tax than if you call it your capital income. So if you broaden that tax base, that you increase tax revenue, but it also has another benefit, which has to do with the direction of AI, that also removes the incentive to over-automate, excessively automate, because you are essentially subsidizing firms to use capital rather than labor. [00:18:30] Speaker 2: Yeah, but let's say an employee like me, I'm going to pay national insurance. Exactly. And my employer is also going to pay national insurance. If the government was going to reduce that, that burden on the employer, that is going to give up a large chunk of tax revenue, which they're going to find very difficult to make up elsewhere. [00:18:48] Daron Acemoglu: Yeah, so that's where you need to increase taxes. Don't increase it on the working people, but increase it on the tax dollars that are not marginal, so that are already accruing to the richest people. [00:18:59] Speaker 2: And what you might then have is businesses complaining that they are being taxed too highly, but you also might have very wealthy people choosing to leave the UK. And that's just happened to the third biggest taxpayer in this country, Chris Rockos, hedge fund billionaire. [00:19:13] Daron Acemoglu: I know that is a real concern. That is an absolutely real concern. [00:19:15] Speaker 2: So what do you do if you have the UK government? [00:19:17] Daron Acemoglu: First of all, this is why absolutely you should not do this at the local level. You should not increase, say, corporate taxation at the local level because businesses are very mobile. But at the national level, most businesses are sticky. And I think there's a very simple thing we can do, which actually also gets rid of a lot of accounting effort that we spend. Just tax all income at source without any regards to whether we're going to call it capital or labor income. It will really simplify things. Now, your concern still remains because businesses might flee to Bermuda or the Bahamas. But I think at the moment, that's got both social, economic, and political costs. And you know what? Here is the very important thing. A new governing philosophy has to come with aspirations that unite people and based on shared moral values. We've lost those. If we have those shared moral values, I think it would be a horrible thing for people to say, because I don't want to pay like three more percent of my vast wealth, I'm going to go to Bermuda. That would be just like the most sort of selfish act. Today, we accept those kinds of selfish acts. But if we can bring people together around shared values, that becomes much harder. So some of that is also internal, endogenous to the kind of governing philosophy we choose. [00:20:36] Speaker 2: What about wealth taxes that are not just about the billionaires, but about a much broader base of people who say live in properties worth $5 million in the US or the equivalent in the UK? I'm saying $5 million because that's the property tax that's coming in in New York. Are you in favor of that kind of tax? What about people who might be retired and living in a property like that? [00:21:00] Daron Acemoglu: Yeah, I mean, I think those are much more gray areas. And one of the things I emphasize in the book is that liberal ideas don't come with a rulebook that provide clear answers to every question. For example, you can ask me even a harder question. And you can say, is the level of social welfare spending in Sweden or the United States more consistent with liberal ideas? And I can't give you an answer because liberal ideas create a big tent. We have to reach compromises and we have to make local decisions on local issues. [00:21:34] Speaker 2: But I'm still not sure whether you're in favor of that kind of tax on assets, on which would be a wealth tax on a broader base of people. [00:21:42] Daron Acemoglu: Precisely because it depends on how it's presented and how it's perceived. So if that kind of tax is imposed by one group without the consent of anybody else, I think definitely no, that's not going to work. It's going to create backlash and it's going to create much bigger exodus of people from local areas because now they can really avoid those taxes by moving. But if people come together, just like during wartime, imagine during wartime, people save so much more taxes. They work so much harder. Why? Because they understood that they were in a boat together and they had a sense of fairness and duty. And so if we can have a debate that supports higher taxes on people's property, even for those less than a billionaire, that's a very different affair than one charismatic politician with a small group of supporters says, I think the line should be at 5 million and not at 4 million. I think that's much more problematic. That's why liberalism really calls for us to rise up to the responsibility of self-government. That's what self-government is about. [00:22:47] Speaker 2: What do you think of Zoran Mamdani and what he's doing in New York? [00:22:50] Daron Acemoglu: I think he has amazing charisma. I think he also is talking about absolutely the right issues and he's talking about them in a very clear and compelling way. Affordability has become a household name because of him. So I think that's all to be uploaded. But he also is a creation of a New York's milieu of politics. And in that sense, he's catering to that very left-leaning activist college graduate base. And his brand of politics, emphasizing socialism or other issues that are really so foreign to working-class communities' daily concerns, I think, is not going to generalize. So sometimes one needs to combine Mamdani's energy, charisma, with somebody who's got a much better way of talking to these communities and creating a bigger tent in terms of value. [00:23:48] Speaker 2: Is that a warning to the Democrats then, not to be pushed as a party or as a future presidential candidate too far towards the Mamdani approach? Absolutely. [00:23:57] Daron Acemoglu: But you know, many of the discussions of that issue in the United States is misleading because they create the impression that there is a single line that divides moderates versus progressives. And we have to choose a point along that line. No, I think we need to be radical in the way that we even imagine liberalism. And we need to be radical in the way that we need to think about economic solutions, in particular about AI. I think this is not a time to be timid and say, oh, well, let's wait for another five years and see what Elon Musk and Peter Thiel and Mark Zuckerberg and Sam Altman come up with about AI. So we need radicalism there, but we need to combine that with much greater moderation when it comes to issues that can create a bigger tent. Progressives have made the mistake of, yes, they've thought radically, but at the same time, they've also narrowed the space, which is permitted. That's a cardinal sin in politics. And it's a cardinal sin when you want to build a liberal coalition. [00:24:58] Speaker 2: AI is part of the critique that The Economist recently published about your work, which I think you said at the time feels like a hit piece. But I am curious about how you see AI because you're talking about it now. I've heard you describe it as this amazingly powerful technology, but also say that you think the rollout is going to be slow and it doesn't feel like the rollout is slow right now. It's changing so many professions around us every day. [00:25:24] Daron Acemoglu: Well, I actually, I mean, certainly the rollout accelerated since I wrote, for example, about three years ago about how large the productivity impacts of AI is going to be. You know, I didn't at the time see how quickly agentic AI would come, but the rollout is still not that fast. We sometimes confuse the adoption of ChatGPT or other chatbots in people's personal lives, like answering questions, making, helping making decisions versus AI being used in the production process. That's very, very slow. If you look at which occupations have been completely transformed by AI, there's only one that's coding. [00:26:12] Speaker 2: I'm not sure that's true. I think you've already acknowledged you've understood AI before, so you could be underestimating it again now. [00:26:18] Daron Acemoglu: Of course, anybody could make mistakes of every direction. There's so much uncertainty about AI. But let me tell you one other thing. What I really object to about AI, and I think what The Economist loves is the hubristic direction for AI set as artificial general intelligence and then artificial super intelligence led by a few charismatic people and a number of companies and frontier labs. And I think that is not really the only direction. And it's not even the best direction. And that focus in the United States has actually slowed down the spread of AI. Look at China. Okay, fine. We're told China is behind the United States in AI. That's right. If you look at large language models, they're behind something like six to 16 months or something like that. But the Chinese, despite all of their faults, and I've been a very long time critic of the Chinese regime, its repression, its surveillance, its economic structure, but they are much more active in trying to integrate AI into the production process. They are investing so much more in robotics. They have the engineers needed to do it. They're doing the same with AI. And that's where we're really not that fast in the United States or the UK. Part of it is because we don't have much of a manufacturing base, which is where a lot of benefits are going to come from. And the even bigger part, and that's where my focus is in the book, is we don't envisage, we don't invest, we don't try pro-worker AI, which is the one that's going to really make workers more [00:27:49] Speaker 2: productive. So the Chinese are better essentially at doing AI galvanizing our productivity rather than thinking about AGI. I think you write in the book that belief in AGI is almost dangerous, that once you become too obsessed with it, you don't explore alternative directions. Absolutely. I don't even [00:28:07] Daron Acemoglu: remember using the word almost there. I actually believe that. Belief in AGI can be self-fulfilling. Yes. Yeah. Even if it's not self-fulfilling, it's actually dangerous. Because if it's self-fulfilling, it will take us to a dystopian future. If it's not self-fulfilling, we'll try to go to that future and do a lot of damage without even any of the benefits. [00:28:24] Speaker 2: But I don't know how you stop it. This technology is learning and it's becoming more [00:28:29] Daron Acemoglu: intelligent all the time as it learns. No, no. That's not the right way to think about it. There is no will to the technology. Technology is not doing anything. It's the companies that are pursuing the technology and the amount of investment that's going into this technology is tremendous. And all of it is going, or most of it is going towards this AGI objective. We can change that. And we should. We should absolutely change it. But let me actually say even something slightly more audacious. If we woke up tomorrow morning and 60, even 50% of the tremendously talented young people who are going into AI changed their mind and they said, no, no, we're not interested in AGI. We want to use AI to make workers more productive, citizens more empowered, democratic spaces broader. That's what we're going to get. It's that simple. It's really where the talent and energy is going. And the question is, can we change policies? Can we change norms? And can we have democratic debate that convince these amazingly talented and generally very, very well-meaning people to change direction? Many of them go into AI because they think working for AGI is the best thing you can do. And why do they think that? A heavy dover of science fiction. And that's what all of the media outlets in the United States have been telling them for the last 15 years. Plus, very charismatic people who are making a lot of money are telling them that. So that's actually not that they're being selfish. It's like, I think they're being in their own minds, okay, this is the best way to do it. This is the only way we can actually improve [00:30:00] Speaker 2: our society. But that's not the case. And by very charismatic people, you mean the AI bosses, you mean Sam Altman, you mean Dario Amadea, you mean Elon Musk, I assume. So what are you saying to all of us media organizations? Like cover them less or don't lionize them? Yeah, cover them less. These are incredibly [00:30:16] Daron Acemoglu: valuable companies. They are. No, no, but cover them less blindly. First of all, remember Trump. Part of the reason why Trump rose to prominence in 2015 was because every word he said was covered. And some of it critically, some of it uncritically. That's what we're doing to Elon Musk. That's what we're doing to Sam Altman. The Economist magazine, bless them. They are absolutely in their right to attack me, although they should have done a little better. I welcome a hit piece, but probably a better one could have been at least more intellectually engaging. But it was one or two weeks before then, they gave their own entire issue to Elon Musk. Elon Musk's ideas about politics and how Europe should be organized. Why is that? Why are we taking Elon Musk's ideas rather than talking about taxing his enormous [00:31:08] Speaker 2: wealth? He faced some quite tough questioning. Oh, no, not really. Oh, no, absolutely. I disagree. On the film that he shared, the anti-Muslim film that he shared. I haven't seen him questioned [00:31:18] Daron Acemoglu: about that anywhere else. Mild, mild criticisms after you platform these people. Yeah, you ask them a few hard questions, but you're giving them an enormous platform. [00:31:26] Speaker 2: I just don't think an interview is the same as a platform. I interviewed Elon Musk last year. I think I asked him some pretty tough questions. I think it was good to see him questioned because generally, he's on X and he's just spouting forth with whatever he says. I know. It's great to push him, [00:31:39] Daron Acemoglu: but if you provide him the authority as if he is somebody who knows how Europeans should organize their politics just because he's been successful in investing in a few companies, I think there is [00:31:52] Speaker 2: something wrong with that. Which takes me to social media. Elon Musk obviously owns X as it is now. He bought what was Twitter. And in your book, you talk about your worries about algorithmic social media feeds that essentially you would like to stop. Ideally, you'd like to stop algorithmic social [00:32:10] Daron Acemoglu: media feeds. Why do you think they're so problematic? So let's go back to the late 1990s or early 2000s. Many people, many, many smart people, very knowledgeable about tech, thought that online communication, social media, social media wikis, would democratize communication, would democratize journalism, would create new communities for us to build trust. Exactly the opposite has happened. Social media has replaced real world communities. People spend less than half of what they did with friends. But the great thing about communities is that they build trust. There is nobody who builds trust on social media. Communities share information. Social media doesn't share information. It shares misinformation. It shares emotional outrage. How did we get here? Were the people completely silly when they thought 25 years ago that something very different could have emerged? No, they weren't silly. They were perhaps naive that that would be sort of the direction at the end. But what they outlined was feasible. Social media could have had a very different path. And the reason why it went this way is very much because algorithmically manipulating people's attention has been a big business. And if you get rid of the algorithm's role in structuring people's emotion, I think we have a much better basis. And there's a very important reason for why targeting algorithms is very important. Regulating social media in a liberal society is a delicate balance. You do not want in any way to clamp down on freedom of speech. I am all for people saying whatever they want on social media, as long as it's not a call to violently attack other groups or anything like that. But algorithms don't have a freedom of speech. It's the problem that we have people saying outrageous things. And then the algorithm finds the most outrageous ones or the most envy generating ones, and it feeds them to you so that you become even more polarized or more triggered. That's the problem. That's why we need to target algorithms, not people's freedom of speech. [00:34:17] Speaker 2: If we took X and Facebook back to the way they were originally, where you would follow certain people and in your feeds, you would only ever see what those people face. [00:34:28] Daron Acemoglu: So for example, it could be Tommy Robinson, who could say outrageous things, but there is no way you would see any of them unless you specifically went and followed Tommy Robinson. [00:34:37] Speaker 2: I get social media is a more recent phase of the internet than the original phase. But if we go back to the beginning in the 90s, the treatment of these companies as platforms rather than publishers, that is really what allowed the flourishing of the internet. That was US law. That is what created this domain which has connected people all over the world. I'm sure your work has benefited from it like [00:34:59] Daron Acemoglu: so many others. Of course. I have also suggested an alternative path. If we cannot be radical, but also simple enough to ban algorithmic feeds, you could also get rid of the most extreme form of section 230, which is this platform rather than publishers. So we can apply some of the same criteria to these companies. But that's harder. It would have been easier 25 years ago. It's harder to reverse some of these things. It's another feasible path. But I think banning algorithmic feed is simpler and more [00:35:30] Speaker 2: powerful. Okay. I also want to ask you about your ideas on money and politics, because that's again something you'd like to see. You'd like to see less money flowing from corporations into politics. Absolutely. Right. So you can't make a corporate political donation. [00:35:45] Daron Acemoglu: I would, again, like freedom of speech, freedom of individuals to contribute to candidates work for [00:35:51] Speaker 2: them. That is sacrosanct. So a billionaire can donate just out of their own pocket. [00:35:56] Daron Acemoglu: Out of their own pocket, but not through corporations. I mean, what's the difference, [00:35:59] Speaker 2: really? You make a lot of money from a company, you donate in your own right as a billionaire? [00:36:04] Daron Acemoglu: Well, look, we have to do this in the context of US reality. A better system would be all parties and candidates are publicly funded. But we're not going to go there. The bigger principle here is individuals are free. They have key rights that are central to liberalism. Algorithms are not individuals. Corporations are not individuals. But people make money out of corporations and then can spend it as they wish. They can spend it as they wish. Again, we have to tax that. And that's why progressive taxation is very important. You know, again, the problem here is we have created this tax structure and fiction that corporations have their own special status. Plus individuals can leave all of their money in corporations and not get the tax burden, not pay their fair share. Plus corporations on behalf of them can spend on both goods and services and all political agendas. That is [00:37:02] Speaker 2: really the most pernicious part of it. I can see that there is an undercurrent of worry that runs Absolutely. You say there's a spectre of violence or actual violence that has been directed today against minorities and other groups. But as we come to a close, I want to ask where you see the points of light. Where do you see the leaders or the governments from whatever end of the political spectrum who are doing things that you approve of and you want to amplify? Look, I am not going to [00:37:30] Daron Acemoglu: look for a knight in white armor. The thing that gives me hope is that A, humans are very adaptable. It takes time, but we adapt to different circumstances. And two, people even in this polarized world have a deep sense of fairness. And around that sense of fairness, we can come together. There are so many shared moral values between the different sides of the aisle in the UK, in the US. It's just polarization, bad leadership, social media have obscured them. And we have made mistakes as liberal democracy's most active supporters in dividing the population further rather than uniting it. Once we take those steps, I think there is a lot more that can come. And that is the context in which leaders will emerge, that build on those unities rather than divisions. Is there really no one that you'd point to as that knight in shining armor? No, not a knight in shining armor. I mean, in the US, I can point to better models as opposed to less good models. But not the democratic socialists, I think, [00:38:38] Speaker 2: not the Mamdani, Alexandria Ocasio-Cortez group. Well, look, again, they divide. I think they have some [00:38:43] Daron Acemoglu: good ideas, but they divide. But if you look at somebody like Ro Khanna or John Ossoff, they are advocating more deep thinking about economic issues, but at the same time, emphasizing unity. And I think that's [00:38:55] Speaker 2: very important. And if we don't seize the moment, the way that you would like is, is the model that gains more currency, not just the Trump model, perhaps it's the China model? Or look, I don't [00:39:05] Daron Acemoglu: want to even think about that. I don't want to even think about that. I believe, and that's the reason why I wrote the book. And I wrote the book, in fact, in a rush relative to what I've done in the past. I think liberal democracy is one of our greatest, if not the greatest achievement. We live in complex, huge, divided societies peacefully. The state is often at our service, rather than repressing us. We reach compromise, all of these, and the individual freedoms that we enjoy, and the opportunities we enjoy are all fruits of liberal democracy. And we are at the cusp of losing it. I don't want to even think of a future where liberal democracy is decimated. We have to do our best to avoid that future. [00:39:50] Speaker 2: Do you think you've become more political as a result of your research or as a result of our times? [00:39:56] Daron Acemoglu: I don't know what political would mean in this context. I've always been interested in politics, and I always think, as you said, politics and economics are inseparable. And that's part of the reason why I didn't want to answer the question, which politician you support. I don't want to make this about this political faction versus that political faction. I think we all have to think about politics. I think we all have to participate in politics. And politics has to be more multidimensional, especially AI. Look, let me make one final comment. We are told every day in the United States that AI is going to transform every aspect of our lives. And I think 90%, that's correct. But next sentence we are told. And you better stay out of the path of AI because otherwise you'll slow it down. And just let it, let Sam Altman, Elon Musk, Dario Amadei, Mark Zuckerberg lead the charge. Those two statements cannot both be correct. In a democratic society, if something is going to be even half as transformative as AI, it is our democratic responsibility to be involved. That's what I'm hoping for. And you're willing to play a role? This is my role. Daron Asimoglu, [00:41:08] Speaker 2: thank you so much. Thank you. This was truly my pleasure to think about. Thank you.