Чем лучше становится ИИ, тем меньше его доля в экономике – Alex Imas и Phil Trammell

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. 今天我和 Alex Imas 聊天，他是 Google DeepMind 的 AGI 经济学主任，也是芝加哥大学的经济学教授；还有 Phil Trammell，他是 Epoch 的经济学负责人，也是斯坦福大学的研究学者。 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 and more advanced AI. 这次访谈里，我想搞清楚的是：经济学告诉我们，在自动化程度不断提高、AI 越来越先进的世界里，我们能期待什么。 what that tells us about what will happen to wages to labor share. 这对工资和劳动份额会意味着什么。 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. 还有，如何给 AGI 带来的财富征税并重新分配，以及什么东西会变得稀缺，因为稀缺性在某种程度上决定了价值流向哪里。 I want to start there. 我想从这里开始。 What are some plausible candidates of what will be scarce? 哪些东西有可能变得稀缺？ 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. 类似于关系性部门这样的东西，我把它定义为：人类参与其中本身就是产品价值的一部分的服务和商品。 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. 因为人类天然是稀缺的，如果自动化使很多其他东西不再稀缺，那么凡是人类在场、人类参与其中的东西仍然会保持稀缺。 I'm curious to understand whether humans doing services for other humans can ever be a big part of the economy. 我想搞清楚：人类为彼此提供服务，这在经济中能否成为重要的组成部分？ And here's maybe one intuition pump. 有一个直觉上的论据。 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. 在一个 AI 可以做任何人类所能做的事情的世界里，有一套庞大的机器经济体系，它们建造工厂、从事研究、产生新想法；而人类在这些物质生产过程中可能参与，也可能不参与，但如果机器人技术得到解决，在极限情况下大概率是不参与的。 If you don't care about humans being involved in that process, why would humans be involved in that process? 如果你不在乎人类是否参与这个过程，为什么要让人类参与？ But then there are these other things you point out where well we actually maybe in some cases do want the ballerina or the barista or whatever to be a human. 但你也提到了另一些东西，在某些情况下，我们实际上可能希望芭蕾舞演员、咖啡师之类的是人类。 That's part of the value of going to a cafe or a performance but only humans have that preference. 那是去咖啡馆或看演出的价值所在，但只有人类才有这种偏好。 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 will want some of the automated goods this machine-only economy is creating and so part of that wealth is flowing out. 所以存在这样一个人类经济：人类互相提供服务，他们的一部分财富流向其他人类，但另一部分财富也会流向机器经济，他们也想要那些纯机器经济创造出的自动化商品。 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. 如果你把这个想成：这不是一个封闭循环，但纯机器经济中有很多东西是封闭循环的，因为机器不在乎让人类咖啡师给它们冲咖啡。 And so within that model, isn't isn't it intrinsic that like the human-only economy will become a smaller and smaller share? 那么在这个模型里，人类经济成为越来越小份额这件事，不是天然内嵌在里面的吗？ I would like to pitch 我想提出 kind of 一种 a rephrasing of that question. 对那个问题的重新表述。 My view is that kind of forecasts 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. 我的看法是：像我和 Phil 这样的经济学家做出的个人预测，不一定很有价值。 The reason I think that so there 我觉得原因是这样， was 有 this 这么 blog post by 一篇博文， Andre uh Fredkin, Andre 啊，Fredkin， Brian Jabarian, and Andrew Brian Jabarian 和 Andrew Co Co that came out yesterday 昨天发出来的， actually uh that looked 实际上研究的就是 at 这个 like kind of people's forecasts economist 人们的预测，经济学家 forecasts about the labor 对劳动 market and what they found is that there's a ton of disagreement like in every single direction. 市场的预测，结果发现分歧极大，各个方向都有。 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 going to do rather 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 200 years old. 他们倡导的做法，我也赞同，是不要做个人预测，而是通过预测市场获取聚合预测，利用群体智慧。我之所以这么想，是因为我们在预测方面一向糟糕透顶。回到 1820 年，我们现在这场争论其实已经持续了 200 年。 David Ricardo is one of the classic economists not neocclassical classical economists and he when industrial revolution started happening, he was wrote a bunch of stuff saying like look this is be going to be great for everybody prices are going to come down.” David Ricardo 是古典经济学的代表人物之一，工业革命开始时，他写了很多东西，说这对所有人都是好事，价格会下降。 But then he turned around and he's like wait 但后来他又转回头说，等等， I can actually see all of these jobs that are creating value they're going to be automated by these machines this is going to be really bad. 我实际上能看到所有这些有价值的工作岗位，它们会被机器自动化，这将会很糟糕。 Everybody's going to become unemployed, and there's going to be political unrest and things like that and if you look at Ricardo's predictions, they're actually right. 每个人都会失业，会有政治动荡之类的事情。如果你看 Ricardo 的预测，他其实是对的。 If you look at all those jobs that made money in Ricardo's time, they got automated. 如果你看 Ricardo 时代那些赚钱的工作，它们确实都被自动化了。 So, if I was David Ricardo and I woke up and somebody told me all those jobs did get automated, and you asked me, David Ricardo, like, what do you think the uh prime age employment rate is in 2026? 所以，如果我是 David Ricardo，现在醒来有人告诉我那些工作确实都被自动化了，然后问我：David Ricardo，你觉得 2026 年的黄金年龄就业率会是多少？ I think he would be surprised if you told him it was the highest it's ever been other than 2000. 如果你告诉他是历史上最高，除了 2000 年之外，我想他会很惊讶。 We have the highest number of employed people that could potentially be employed since 2000. 我们现在拥有自 2000 年以来最高的潜在就业人口就业数量。 That was like the peak and now it's like the second peak basically. 那是一个峰值，而现在基本上是第二个峰值。 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. Ricardo 遗漏的是结构性变革的经济学：所有被自动化的东西都变便宜了， People had more money to spend on things and then they started spending money on services. 人们有了更多的钱，然后开始把钱花在服务上。 And you know this is kind of like the lump-of-labor fallacy. 这就是所谓的固定劳动量谬误。 That's what they call it. 他们就是这么叫的。 Dave Ricardo didn't think, hey, I should have, you know, considered the fact that new jobs would be created. David Ricardo 没有想到，嘿，我应该考虑到新工作会被创造出来这个事实。 But it's kind of not obvious that like money would go to services. 但钱会流向服务这件事，其实也不是显而易见的。 Like why wouldn't they go to more automated goods or and something like that? 为什么钱不会流向更多的自动化商品之类的东西？ And I'm not saying that like I'm not using this anecdote to say like this is what's going to happen now and that we're going to have full employment. 我不是用这个故事来说这次也会发生同样的事情，或者我们会实现充分就业。 I'm using that anecdote 我用这个故事 as to say it's really hard to make predictions. 是想说预测真的很难。 And what I think maybe a a really useful tool that econ economists have is instead start with a premise like maybe we'll start it today. 我认为经济学家真正有用的工具，是从一个前提出发，比方说，我们今天从这里开始。 Look, labor share is zero. 好，劳动份额是零。 Like labor share has gone down. 劳动份额已经下降了。 What could possibly explain this? 这可能怎么解释？ Let's write down an economic model of what happened. 我们来建立一个经济模型来描述发生了什么。 Phil will talk about this later today. 这个 Phil 稍后会讲。 Or you can start write down a model to say, hey, what if labor share just stays the same? 或者你可以从另一个角度建立一个模型：假设劳动份额保持不变，需要什么条件才能做到？ What can make that happen?” 是什么能做到这一点？ 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. 我最核心的观点，如果你从这场对话里只记住一件事：我们没有数据。 I've been kind of saying we need a Manhattan Project for data. 我一直在说，我们需要一个数据曼哈顿计划。 We don't have data on basically consumer demand elasticities. 我们基本上没有消费者需求弹性的数据。 We don't know what they are. 我们根本不知道那些数字是多少。 We don't know um we're not really tracking what jobs are getting created or destroyed like the Onet database with all of the tasks and different jobs that's been rarely updated and is super low quality. 我们甚至不知道哪些工作在被创造或消失，追踪工作任务的 O*NET 数据库更新极少，质量极差。 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 tight what dimension of scarcity will generate that scenario. 所以我认为真正有用的是：想清楚有哪些可能的情景，一一梳理，然后说清楚哪种维度的稀缺性会产生哪种情景。 If there's full employment, we could talk about the relational sector or something like that. 如果是充分就业，我们可以讨论关系性部门之类的东西。 If there's 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. 如果劳动份额崩溃，我们可以讨论其他情景，然后弄清楚我们应该收集哪些数据。 It's probably worth 可能值得 the 先 defining labor share and capital share real quick. 快速定义一下劳动份额和资本份额。 The whole economy like the the total sum of goods and services sold um is either paid out to people in wages. 整个经济体，也就是所有出售的商品和服务的总和，要么以工资的形式支付给人们， Yeah. 嗯。 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 get paid out. 要么支付给资本，也就是有建筑物租金，还有公司的股东也会得到回报。 For many hundreds of years in the economy uh 60 something% 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 claims on companies and whatever. 几百年来，经济中60多%的东西，也就是某一年里所有售出商品的60多%，以工资形式支付给了人类，另外30%-40%则支付给了拥有机器、土地和公司股份等的人。 Um and the question is well right now our 60% is going to wages does that shrink as automation um or as AIS get smarter and smarter and better and better 问题是，现在60%流向了工资，随着自动化推进、AI 越来越强，这个比例会缩小吗？ and it's like it 它真的 really 是 this is a Kaldor fact like right so it's incredibly we should stress this. 一个卡尔多事实，这一点非常值得强调。 It's incredibly surprising 这令人难以置信地惊讶， that it's over 60% after the industrial revolution after all of the automation we've ever seen the fact that it's almost like some people are worried it's an accounting error or something like that that it's kept being so constant and uh the fact that it's like been over 60%. 工业革命之后，经历了所有我们见过的自动化，劳动份额还是超过60%。有些人甚至担心这是不是会计误差，它居然能保持如此稳定，一直在60%以上。 And you know there's there's even a controversy right now. 而且你知道，现在还有一场争议。 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 there's been a lot of accounting changes in the last 30 40 years. 有人说劳动份额在过去20-30年里有所下降，但具体怎么衡量，过去30-40年里有很多会计准则变化。 So for example Andy uh Atkinson have has this paper showing that actually if you keep the accounting constant over the years labor share hasn't even fallen ever. 比如 Andy Atkinson 有一篇论文表明，如果保持会计准则一致，劳动份额其实从未下降过。 But it's not 但这并不是 it's not 说 that surprising, right? 这不令人惊讶，对吗？ Phil, you made this point that if labor and capital are compliments, you need both to do anything. Phil，你提过一个观点：如果劳动和资本是互补的，你需要两者才能做成任何事情。 It would kind of make sense that you'd kind of need to pay both of them to get something done. 从某种意义上说，两者都需要，才能完成任何事情，这也说得通。 You have had stuff 有些东西 can 可以 be completely automated. 被完全自动化。 Although, you had the post where you were pointing out that actually. 虽然，你写的那篇文章里提到实际上…… Oh, yeah. 啊，对。 Well, I was going to say there's a sense in which some nothing's yet been completely automated. 我想说的是，从某种意义上说，目前还没有什么东西被完全自动化。 Look at the network-adjusted factor shares of a good which is to say you look down the supply chain and say 看看某种商品的网络调整后生产要素份额，也就是沿着供应链追溯， not just like the final step how much of that is done by capital and labor, but what went into the machines that can automate that final step. 不只看最终那一步有多少是资本、多少是劳动，还要看能将最后一步自动化的机器本身是由什么构成的。 You'll find that labor is adding a lot of belly down the supply chain so like you know computer and electronic products in the US have a very stable capital share network adjustable capital share of around 50%. 你会发现劳动在整个供应链中都在大量加值，比如美国的计算机和电子产品，其网络调整后的资本份额非常稳定，大约在50%左右。