How to help people who lose their jobs to AI

Tech CEOs keep telling me that AI job loss concerns might be overblown, but one economist just told me she thinks the results could be earth-shattering. So what do we do about it? That's this week on Platformer. This podcast is brought to you by Atlassian RoVO, the AI that takes your team from AI novice to AI native. Welcome to Platformer. I'm Casey Newton. This week on the show, Brookings senior fellow Molly Kinder joins us to talk about what she calls the messy middle, the period in between the start of AI job disruption and whatever might be waiting for us on the other side of it. Molly is somebody who believes that the possibility of AI job loss is very real, unlike some of our previous guests who tried to sort of present us with the idea that jobs are going to change a lot but not go away. Molly really takes seriously the idea that they might, and she has noticed that we seem to have basically no plan for what to do if that happens. So that is a great conversation. I'm excited to share it with you. First, though, as always, we begin by checking in on the state of AI and jobs. And that means it's time to bring in Platformer fellow and Gen Z AI correspondent, Ella Marcianos. Ella, how are you this week? I'm doing great. In my personal life, I'm feeling kind of good about AI automation because I just had a Claude design some platformer laptop stickers for me that we're going to be handing out at our live event. And I love a sticker. So I think all of this job loss is probably worth it. Sure. I mean, yeah, if the sticker is good enough, I'm willing to put up with a lot of job loss. What part of the sticker process do you feel like the AI was most helpful with? Yeah, so it did two things for me. One is I basically just edited the banner image slightly. I like changed it so that it was like still just the logo and the name, but it had the accent color for the website on the logo. Yeah, so it's just like a very boring edit. And then it also upscaled the image so that it could actually be used as a sticker, which is like something I would have no idea how to do. Nor would I. And it was like so fast, which was so satisfying to me. All right. So, yeah, this is a classic case of the AI is augmenting us in our jobs and not yet replacing us. But I imagine that this week you have once again looked into questions around AI job displacement. And I am curious what you have been thinking about. Yeah, so today, this week, a little bit displacement-y. Basically, a cast of characters that's like a very fascinating squad to me, which is separately Gavin Newsom, OpenAI, and also the Pope, all have some notable things to say about AI job displacement. Gavin Newsom did an executive order. OpenAI is announcing some funding. The Pope did an encyclical. But they're all talking about how AI job loss could be a really big deal and we need to invest in retraining workers. I'm trying to decide if those are the three folks that you just named, I would consider that a nightmare blunt rotation or more of a dream blunt rotation. Maybe it would depend a little bit on what the conversation was about. So let's dive into this a little bit. I feel like just for, you know, reasons of respect, we may want to start with the Pope. What does His Holiness have to say about this? Yeah, so yeah, he basically, he said, we need to invest in retraining and also companies need to keep in mind the dignity of and like pleasantness of work in addition to keeping in mind jobs as like an AI transition happens. But the thing that was like most notable to me is he basically says, we should establish social criteria for innovation. Where, what that means is in the encyclical, he says, every introduction of automation and AI should be accompanied by verifiable measures to protect the employment, retraining, and participation of workers. And what do you take that to mean? Yeah, so there is like some ambiguity here because it's like not necessarily saying whether this is in the hands of the companies or the government. But basically what he's saying is if AI is being deployed that like really has a chance of automating people's jobs, like there needs to be a measure to mitigate this at the time the AI is deployed, like at the time a big new tech change comes. And like it would be probably bad and irresponsible to just have that innovation without it being pretty clear that you have a plan to protect people. Right. Well, look, if I know one thing about American corporations, it's that they are laser focused on the dignity of their workers. So I feel pretty good about this one coming true. Let's talk about the OpenAI Foundation. What are they looking at doing? Yeah, so they're dedicating $250 million in funding to new economic futures in the age of AI. They had one quote to me that freaked me out, which is the current pace of change means the window to get this right is shorter than we're used to. And the cost of getting it wrong is immense, which is very classic OpenAI language, but I was like, guys. Relax. Yeah. But maybe we shouldn't relax. Maybe we do need to be thinking about this right now. So, okay, so interesting. So they say that like job retraining might be part of the answer, but we're going to need to explore new economic futures. Do we have any details on what that would mean? They bring up some stuff. They say like in the short term, they talk about conventional policy solutions like also like things like unemployment insurance, other like sorts of welfare systems. But then they're like, these types of programs are not preparing us for the future where like AI actually does a very large portion of all work. So we need to try new stuff. One of that is those things is like sovereign wealth funds, like, for example, Norway has where like the government basically invests money that is then like distributed to citizens. They kind of have a lot of it partially because they're starting the program is really vague. They're like, we just may need to rethink how the economy and work happens completely, which like I see, like, in fact, if AI is doing like 80% of labor, you know. Sure, we might need that. I mean, I'll say like a sovereign wealth fund appeals to me in the sense that it would just be the government sending me money, which is probably one of the best things the government can do for you. All right, so that's them. And then what did California Governor Gavin Newsom have to say for himself? Yeah, I mean, I was it was remarkable to me in this case, how aligned Gavin Newsom and OpenAI were because despite Gavin Newsom being Mr. Billionaire Tax, basically Gavin Newsom made an executive order that called for new investment in and study of retraining programs, in particular for people whose jobs might be eliminated by AI, including like software developers, customer service representatives, marketing and salespeople. But then he also called for investigation into new ideas, including universal basic capital, which, yeah, would give people stakes in stuff like stocks, bonds, or wealth funds. I see. OK, so there's some interesting ideas here and it does seem like job retraining is now a notion that is appearing in more and more places. How effective do we think that sort of just training workers to do other things could be in helping us navigate that transition? Yeah, so it's kind of complicated because job retraining, it basically large government job retraining programs have worked to increase people's earnings. The like overall literature review find is like a year, the first year of the program, they're usually workers are like in school or something. And so their earnings drop. And then over the next few years, workers like do get a real gain in income. The thing is, when you look at how the evidence for this slice of people whose jobs are getting displaced by AI or like in general, how this might work out for worker displacement, the picture gets a little bit more complicated. So one of the like best pieces of evidence on this, uh, that is this basically, um, study of 20 years of the trade trade adjustment assistance program, uh, in the U S, um, which like tries to look at like over 20 years, uh, workers who have been displaced by some sort of, uh, technological change. Uh, how well does the training program work for them? And then especially compared to a baseline of like a similar demographic. Um, and what it finds is like, uh, you get tens of thousands of dollars on average in wage gains 10 years out from the training program on average workers make $50,000 more than people who had never been in the training program. But then when you go 20 years out after that, uh, the gains disappear basically like it returns to baseline. Um, so the author of this study, uh, obviously we don't know, but guess that it was partially the new skills that these workers were being taught for the first 10 years were useful skills, but like actually eventually became obsolete because work kept changing. Which is, of course, the fear of, of AI in general, right? Is that no matter how quickly you're able to, like retrain a worker by that point, the jagged frontier of AI capability will have caught up with that worker and there's sort of nothing to retrain them to do. That's obviously a very like pessimistic view of this. But like, what views do you have? And now, after you've sort of done this research on the, the job retraining question. Yeah. What views do I have? I do think like, um, one thing people try, which is promising That's what we have in store for you. Today, here is my conversation with Molly Kinder. Molly Kinder, welcome to Platformer. Thanks for having me, Casey. Well, you have spent your year looking at AI and the economy through the lens of the worker. Many of our guests on the show so far have come at it from the technology side. When you look at the economy, how do you think you see it differently than a Silicon Valley technologist might? You know, that's a great question, Casey. So my whole approach to my research is to start with the individual. So I come at this not only what sort of I'm held responsible at the end of the day for my work, is making sure I'm promoting what I think is right for society and the individual. But a lot of my approach to my scholarship is interviewing people. So I spend a lot of time talking to people of all backgrounds, all stripes, about how AI is impacting them and what they want from the future. So I think I come at this both grounded in a sense of my scholarship is about how do we make sure this is a technology that people will benefit from and not be harmed by? And I'm ruthlessly focused on what this technology at the end of the day is going to mean to individuals, to workers, to families. So I think that's probably a slight distinction. And I spend a lot of my time in dialogue with individuals as well. And not just in Silicon Valley, although, of course, AI is impacting the tech sector, but really a much broader range of occupations and people. Yeah, I mean, as a journalist, I'm always telling people the interview is one of the most powerful technologies that we have. So I'm glad to see you deploying it. Yeah, no, for sure. It's a little bit, you know, atypical for my role at the Brookings Institution. I mean, we pride ourselves in being scholars and data and experts. But I actually think it's really critical. You miss things if you don't get out into the real world and talk to real people. Yeah, yeah. There's more to life than reading X.com. So let me get to the piece that led me to reach out to you and talk today. You wrote a piece recently about what you call the messy middle of the AI transition that you argue is about to come. What is the messy middle? Yeah, thanks, Casey. So I wrote this piece in part because I've been so frustrated by the state of the AI and jobs conversation. And I think you must be, too, because you've been doing this great series bringing perspectives. I feel like we're in this giant seesaw where, you know, one group of people or one journalist will put something out that declares an apocalypse is coming and takes a really extreme view. And then the other side will say, no, the apocalypse isn't coming, so there's nothing to see here. So we go from sort of really extreme, like all jobs are going away very soon, and that's awful, to you're overreacting, there's nothing to see here. And I'm calling out the messy middle both to stake out a more reasonable middle ground between these two extremes, but also I was talking about a time period. So I think we are in reality, this like time T0 right now, what we're in is such early days of this labor market transition that you almost can't see much of an evidence of labor market disruption yet. So everyone's fearful, but we're not yet seeing a lot of labor market impact. I would say reality 3, and credit to my colleague Pollock Shaw for coming up with this r1, r2, r3 framing, r3 is what I think a lot of Silicon Valley describes. This world of post-AGI where essentially robots and AI can do everything. And that's the jobs apocalypse narrative that sort of we're debating over and over whether that's realistic. I actually think what we're really entering is this messy middle period. It's a world in which AI gets better. It's more capable of taking on more work, but we don't overnight see a jobs apocalypse. Instead, we find something that's still painful, but it's more narrow. It's sort of a world of partial automation where AI starts to get capable enough to do certain types of jobs. And that is still very painful. A world where most jobs are intact, but there's a concentrated loss is still a world that is politically, societally, and economically explosive. So I'm trying to call attention to this messy middle period, which could last for decades. It depends on how good the technology gets. But it is still one that we have to take seriously because even a world where we don't see a full jobs apocalypse, if we're seeing a lot of pain in the knowledge sector or to early career, that is still going to be something that we're going to feel as a country and that we're not prepared for. Yeah. You argue that this middle period of AI disruption is going to hit white-collar workers much harder than blue-collar workers. Make the case for why you think that is. Yeah, and I would say this is just what is going to go first. And when you look at ChatGPT world technologies, large language models, we've got lots of data. OpenAI made public a data set that looked at every task across the economy and its task exposure. And then you can aggregate that up to individual jobs and individual sectors. And it's just basically saying, is this the kind of job where you can use a version of ChatGPT to save a bunch of time? And if you look at that, and I have a chart in the Substack that I wrote coming from some research I did at Brookings with some colleagues there. If you look at the kind of sectors that are most exposed where we're going to see the most usage, and for good or bad, in terms of job security, it is by and large the knowledge sectors. It is at a computer. My framing is if you can do your job locked in a closet with a computer, eventually you're probably going to be in trouble. So it's the really computer-based work. Most of that is kind of bachelor's degrees, law, finance, consulting, sales. There's a big component of it that's more clerical back office that's not usually a BA associated with that. And I can talk more about why that's something we need to focus on. What you see as not really that exposed, and it's reflected in the data of usage, are blue-collar jobs, physical jobs, jobs, service sector jobs, jobs that you have to show up in a workplace, whether it's a restaurant or a haircut salon or repair shop. Those are jobs where ChatGPT is not really going to help you do very much. Eventually, we're certainly going to see robotics enter the picture. I don't have a crystal ball. We don't have good data on that. I think we're seeing much faster advancements in computer-based knowledge work. So that's what I see as the frontier. That's where we're already seeing. And I think that's going to hit and be disruptive first to those computer-based knowledge jobs and that clerical role. And that's pretty intuitive to me, right? Like I now spend a fair amount of time just gently testing these models to see which parts of my job they can do. And they can do more of it today than they could a year ago, right? Same, Casey. Same. I do a lot of my work at a computer. And it's the stuff at a computer that it's getting remarkably good at. Yeah. So, I mean, I think we have to keep our eye on that trend line. Can you sketch out like some of the implications of maybe white-collar work falling to automation first and like how that might contribute to messiness? Yeah. So let me first give a little bit of a historical perspective to show why this is such a change if this does happen. And again, I want to caveat, I don't have a crystal ball. Nobody can fully predict the future. There's lots of ways, you know, the impacts on white-collar may be softer than I imagine. I'm also not saying it's all happening tomorrow. I think we're at the precipice of a transformation. The possibility of commoditizing cognition is something I talk about in the paper. It's not overnight. This is the where I see the technology going. If you look at the last, so I'm in my mid-40s. So I was born in 1979. In the paper, I show a chart, a figure that goes back 150 years and it looks at what types of jobs dominated the labor market in the U.S. over time. And it shows the sort of ups and downs and changes. And since I was born in around 1980, you see this huge surge of knowledge and professional jobs, you know, the kind of white-collar jobs we think of that dominate now the economy in business and finance and accounting and law. And what we saw before was first there was an agricultural revolution and sort of that was mechanized. And then we had tons of blue-collar employment. I mean, upwards of a third of all employment was in manufacturing blue-collar. Then we saw this big decline. We've had this huge increase in knowledge jobs since basically the invention of the computer. And we've called this skill-biased technological change. Up until the moment ChatGPT was launched, I've always considered computers boosting the knowledge worker. It made them more productive. It made them more in demand. It allowed them to be, you know, here I am typing away on my computer. I'm going faster. I'm more productive. But the computer didn't do my job. My brain did my job. In comparison, a lot of jobs that have gone away were substituted by computers. So if you think about a lawyer, in the 1980s, you probably had one lawyer and one legal secretary. The legal secretary did all the typing, did all the scheduling, did the dictation. Computers came along and replaced a lot of that work. And it made the Potentially true. There's also some demand issues. Maybe if lawyers get, you know, very productive and it all becomes cheaper, we demand more. All of those things, I think, are true. I think what, when I listen to Aaron and James, part of what I think was, though, missing a little bit is project ourselves out a few years. We are still talking about the early days, the doctor that was the example Aaron used. Of course, every doctor wants to save time on note-taking. That's not why doctors are fearful of their job, and I've talked to many in the medical profession. What strikes fear in doctors or nurses is, is AI going to get good at the part of my job that I think of as the crux of my job, where I really add value? There hasn't been an academic study yet where the human bests AI in diagnosing a disease. You know, there's lots of examples in the medical profession of, of AI being better at doing the hardest parts, the stuff that doctors enjoy the most. A lot of doctors have said they enjoy the diagnosing, and that, that might, that's not the same thing as saying, let's just get rid of the sort of routine part of our job and do more of the good. When I talk to creatives, they're worried about the thing they love about their job being substituted. And even for me, you know, now I'm 46. I'm very senior in my job. I have a lot of, you know, interpersonal parts of my job. I didn't for the first 15 years of my career. It was mostly sitting at a computer and typing. And so I think what a lot of people fear is, not where we are today, in this trajectory, if AI gets better and better, I can easily see it. When I'll have a two-hour conversation with Claude going really deep on some policy areas, and I sometimes walk away and think, that is like talking to a scholar at Brookings. And this is still early days. You know, it is humbling. It is genuinely humbling to see how smart it is getting. And so I think that is what has been missing in some of the previous discussions, is an acknowledgement of what happens if AI gets better at what makes you special in your job and what you love. Does it mean all jobs are going away? But I think in certain realms, it's going to be very automating. This is really important. And I think this accounts for like 85% of the uncertainty around this conversation, is lots of folks I speak to believe, and I think I believe as well, we are living through a period of exponential improvement in large language models, and it is just basically impossible to predict the future when you are living through exponential change. The ground keeps shifting from underneath you. You mentioned talking to Claude and how humbling it can be to talk to it already. Anthropic has a model called Mythos that they have not released to the public yet that they say is sort of a step change. How might this conversation be different if we had it in six months once people have tried to automate their tasks with a model of that caliber? So it just gets really, really tricky. And one reason why I wanted to have you on the show is I think if you work at a technology company, your incentive is to tell me, no, no, no, Casey, don't worry. It's going to be fine. Like, yes, there's going to be a lot of change, but like, you are going to keep your job, so you can calm down. You don't have to try to, you know, destroy the data center that we're building in your community. You don't have to boo us at graduation, right? But I think what's important about what you're saying is that if the technology improves rapidly enough, that sort of jagged frontier is going to catch up to a staggering amount of tasks, at least, and that at some point, that probably tips over into a full job. Yeah. I mean, I also, you know, I think you, I think that your listeners should understand there are so many different types of jobs out there. And the way AI is going to impact jobs is going to be very different in different sectors. And it doesn't, I think another thing that I worry about is this idea of de-skilling, that if the cognitive part of your job, the really hard cognitive part, can be deferred to Claude, you might not need as much education experience, and that becomes a lower-paying job. And that's something I'm already hearing in the healthcare space. I've heard examples from doctors telling me a completely untrained person off the street being guided by, you know, generative AI to perform an ultrasound, that's an $85,000-a-year job that requires a year of post-secondary. If you can just de-skill jobs, that's another issue. There's a lot of different ways this could play out. I don't think we are heading into a jobs apocalypse. Right now, my use of AI makes me better in my job. I think there's lots of upsides. I think, though, the notion that there's nothing to see here just because we're not heading into a jobs apocalypse, it won't take that much if a handful of really coveted sectors start seeing a displacement of some talent. I think what the whole public is on pins and needles. Everyone's worried about this. They're looking for the first real proof that what they fear in their gut is coming, is going to happen. My point is, if policymakers and politicians are not ready with a response, they don't, and if they see that this is going to be a repeat of social media where, you know, kids were harmed for 10 years with nothing, a repeat of deindustrialization where good jobs in the Heartland were lost because of trade and automation and we did basically nothing to stop that or respond to it, well, I think people are going to lose faith and they're going to turn even more against AI. I think it's imperative that, you said it, I've heard you say it many times, Casey, you know, what's the plan? If jobs are lost, one of the things I wanted to highlight in my piece is if you do lose your job and you were making $200,000 a year and you have a graduate degree or an undergraduate degree and you're very specialized and you can't just slip into something else, your potential fall from that livelihood and your home mortgage and, you know, what you're able to support in your family could fall pretty precipitously. And we have a really threadbare safety net and there is no unwritten rule that says you get to get another job at the same pay. Imagine if that repeats over enough people. It doesn't have to be a jobs apocalypse for people to feel that what they hold dear is slipping away. Absolutely. Our previous episode with the labor economist, Catherine Ann Edwards, we talk quite a bit about how threadbare that safety net is and what some policy solutions might be. And I want to get to your own policy thoughts later. I thought before we move on, though, some folks might be curious, you know, I feel like almost everyone I know has at least one eye on this question of how good is the AI at my job? And is that starting to show up in the productivity statistics? You published research about how the data suggests that there is no jobs apocalypse yet, but I'm curious, like what signals are you looking at to serve as a kind of early warning system? Like when it shows up, like where where will it appear first? Yeah. So just to credit my colleagues at the Yale Budget Lab, Martha Gimbel and Maddie and Joshua did an incredible job of taking aggregate labor market data and basically gut-checking, is the house on fire? And the answer is, it is not. We are not seeing economy-wide labor market disruption. And I don't, and our starting point is we shouldn't expect to three years in. This is really early days. I don't, I think it's going to be some time before, at a macro scale, when you zoom out, you're really going to see this. I say it's almost like, that methodology would tell if the whole house was on fire, but not if there was like a stove fire in one individual room that was just like an early flame. I'm looking at individual sectors where adoption is very high. So for instance, I want to know the early moving. Right now we have very wide variance in how quickly certain sectors are truly adopting AI versus others. You know, I'm looking at customer service and some of these back-office functions where it's not so much that individuals in those sectors are using cloud all the time, it's that employers are automating from the backend a lot of that work. So that's a barometer to me. I'm looking at sectors like, obviously, software engineering and the tech sector, but also marketing, market research, finance, to some extent law, though I don't expect to see much impact there. The challenges are labor market data when you start getting kind of granular is not very recent and not very good. But we're trying to get, that's where I'm looking to see, we should, we should look at the sectors where AI is moving the fastest and ask the question, what's happening there? And that's where I'd like to see, you know, more evidence in the coming year. I expect to start seeing, particularly in early career, more impacts and sort of over the next one to three years, more impact beyond that. Yeah. Interestingly, it seems like the number of postings for software engineers are up in 2026. Yeah. Which, you know, I don't know how much we should make of that because again, like we have the models that we have now, they can do what they do. It's really impressive. They're automating tasks. They haven't automated the job. And yeah, it's like, well, if we were about to see the beginning of something really, really disruptive, And one of the arguments that you make is that we're heading into a kind of inverse of COVID situation that might disrupt the way that you might think UBI could go. So tell us a little bit about why you don't like the UBI solution. Well, what I don't, what I was reacting to was kind of the San Francisco consensus I hear when I'm engaged in a conversation with someone who's very AGI-pilled, very sort of in the AI space and thinks we're going right to that post-AGI reality. When I say, oh, you know, my job is to think about policy solutions in this interim period, for instance, like, what do we do about young people who are, you know, need to get on the career ladder, or what about these displaced knowledge workers? They say, well, no, I mean, we just go straight to UBI. You know, we get checks big enough for everybody that replaces essentially their entire full income. And it's universal. Everyone gets it. And so that's partly my motivation of writing this piece is if we're not skipping right to a world where nobody works and the answer is everyone gets a check, if instead you're in the messy middle, which is some jobs are disrupted and you have this concentrated pain, and I give the historical parallel. It's just like when in the 1980s through 2000s, when we lost in a concentrated way jobs in the heartland and manufacturing. Overall, the economy was fine. And overall, jobs were fine. You just had this concentrated loss in the heartland. If we have this partial automation, and I use the example of COVID because I've been thinking a lot about how the doing your job locked in a closet in a computer or the folks who could stay home during the pandemic, they didn't have to show up into a workplace and risk the virus. I actually, what I researched at Brookings during the pandemic was essential workers. I spent all my time with grocery workers and care workers and really arguing that they deserved a living wage and safety and whatnot. And the folks who had to show up into the office or into a restaurant or a hospital or you name the sort of essential workplace, almost by definition, computers can't do that job because you had to be in person. And the folks who were the most safe from the virus, who could take all their meetings at home virtually or were mostly typing on a spreadsheet or whatnot, those are the things that AI is best at. So what we have just learned in the last decade was the economy literally cannot function without all of those tens of millions of people showing up in hospitals and telephone repair and food manufacturing and farming and public safety and schooling, all these critical infrastructures, by and large, are not the ones that are on the front lines of AI displacement in the early days. I can't predict if robots are gonna suddenly take over everything. It's the laptop class that was not declared essential that are the most at risk. So what happens if you have, and again, I can't predict, is it mild displacement in knowledge work? Is it, you know, more extreme? But if you see a lot more examples of the people with the $150,000 to $200,000 incomes who lose their livelihood and can't connect to another job, the idea that everyone in society gets a check large enough to replace their lost wages means why in the world would anybody show up to still do the policing, the construction of the house, the hospital jobs for way less money than that? And we live in a country, I don't agree with this value, but we live in a country where distributing checks for people not to work tends not to be very favorable in America. So how politically sustainable is it for all the folks with essential jobs earning quite modest wages to keep working for $40,000 a year or $60,000 a year and can we in perpetuity send checks to the $200,000, you know, software engineer network? We end up in this very messy situation. So it's not that I don't think we need to have some kind of compensation, it's the idea that we can just give checks large enough to replace everyone's income in a world that's the messy middle, you've just destroyed the labor market. So I think what we need are more targeted interventions that support those who might be in the midst of the messy middle if we do get to a reality-free where literally no one works, of course we need some kind of full income replacement, but I think we're facing something much trickier in the messy middle. Well, let's talk about another solution that has been proposed recently. Senator Bernie Sanders said this month that he will soon be unveiling a proposal to take equity stakes in the major U.S. AI labs and create a sovereign wealth fund that would share dividends with the U.S. people. What do you make of that idea? I would separate two really pressing policy challenges that we face right now. One is how do ordinary people across society benefit from the gains of AI? There is going to be a phenomenal amount of wealth that's going to be generated post-IPO in San Francisco. I mean, Casey, you live amongst it, you know where this is going, biggest IPOs in history, huge amounts of wealth. I think we face a society-wide question of how do we make sure that doesn't just benefit a few thousand people in San Francisco? And I think just generally, how do we make sure this technology is one that's benefiting all of us? It's not just going to shareholders. Those are a set of questions about how do we share the surplus from AI? I would separate that from how do we handle the concentrated pain that will come from people on the losing end of job disruption? Those are not the same thing. For instance, if you think about something like a sovereign wealth fund or some kind of dividend fund, which I, again, I'm supportive of this notion of how do we make sure this wealth benefits everybody? Everyone's going to get some kind of amount. It's not going to be enough to replace your income. It's going to just be something we all benefit from. I think we face a really big question in society, which is, how do we make, there are going to be some people really on the losing end of this, and their pain is going to be acute and real, and it's going to create societal, economic, and political challenges. We have to figure out how to capture some of the gains to invest in solutions that help them. That's not going to be the same thing as a small check to everybody. So I would just separate those things. I care deeply about this question of making sure ordinary Americans benefit. Lots of policy ideas on the table. I think we can debate which one is best. I think there's a separate question, which is, how do we make sure we help people navigate who might be on the losing end, knowing we messed up so badly deindustrialization and it was a wreck? It was a total mess. We're bad at this. And so we have to figure out in the messy middle, how do we not leave people behind? That's young people who feel like they can't start a future. It's any knowledge worker who could lose their job and not find another good one. It's these high school educated women across the country as bookkeepers and medical coders. Those are the best jobs for high school educated women. What's going to replace them? So I would separate those things out. Got it. Okay. Well, so then let's now, now that there's been an appropriate amount of buildup, what do you think are some good ideas to handle the job disruptions that may come? So I have a few categories that I'm playing with. One is, I think I've been really impressed with some of David Shore's polling. I'm not sure if you've seen some of his polling that shows kind of where the American people are. When I talk to the American public, I hear all the time anxiety that comes in threefold. One is people are already in their working lives, like you and me, Casey, who worry about a game of Russian roulette. Am I going to be the person that one day wakes up and some version of Claude can do my job? There's a real anxiety of, you know, my fragility, am I going to end up as the Uber driver who just, my income just plunged? Second thing is young people who are super fearful that the American dream is just being pulled. They're the collateral damage of AI progress. Shareholders could benefit if you replace young people with Claude, but what happens to those young people? And then I would say as a parent of three kids, 11 and under, a lot of parents are bewildered. How do we even prepare our kids for a future? What even is the point of school? What do we do to set our kids up? So there's a lot of uncertainty and a lot of fear. And what's riding through this is people are not asking for a check and they're not asking for retraining. They're asking to have some security in a world where things are really expensive. They want to keep a job. So the first bucket I would say are what are really creative policy interventions that slow down and manage the pace of disruption before it even happens? I think we are really bad as a country of once the displacement happens, figuring out what to do. You know, I don't believe in just putting up protectionist walls like we shouldn't just have certain jobs that just because they're powerful, you know, they say no one can use AI for any legal advice. I don't believe in those kind of solutions, but I do believe there are ways that we can better manage, have higher expectations for employers in terms of retention or how much they have to do to try to repurpose someone before they lose their job. I think the same holds for young people. I am passionate about ideas that make sure young people are just not the collateral damage. How can we hold employers accountable to not just cut learning opportunities just because they can? So I have some Chatting jobs will impact jobs. I've been able to get out and talk to people. I've been to the Vatican twice. I've convened economists. I've talked with policymakers. And it's been amazing, but my anxiety levels have just keep going up as I watch the meter charts and as I see this is going so much faster than even I thought three years ago. And what I observe is angst everywhere. The public is anxious. Politicians are suddenly realizing they need a plan. They don't know what to do. We're drowning in analysis. Everyone's debating. People are just putting out what is happening now, but we're really short on answers. What is the plan? And, you know, when I'm in a think tank capacity, I feel I'm in the conversation, but somewhat on the sidelines. And I feel we've just gotten to a point. I look at my three kids and I think, this is too important. I have to give everything I have to trying to not study this, but solve it. So I have to, you know, I can't announce all the details, but I will say that the spirit of what my next chapter is, is to be dogged and urgent and put everything into figuring out, what do we do about this? How do we make sure this transition, especially this next messy middle, we learn from our past mistakes and we actually grip this head on. So Casey, have me back on the pod in a few months and happy to formally tell you our name and exactly our plan. But you're my first, you're the first tell, and this is actually the last day of my job and I take the plunge tomorrow. So thanks for being here to celebrate with me. It is a joy to celebrate with you. Congratulations on the new thing. Cannot wait to learn more about it. I think it is super important. We need lots and lots of smart people trying to help us figure this out. So- Well, and I saw too, Casey, before we started that you and Kevin have been the sort of voices in my head. You have so many times called attention over and over to how much this matters to people and yet how we don't have a plan. Where's the plan? What are we going to do? And I heard that every Friday. So, you know, you've been, you and Kevin are a big reason of why I'm taking this gutsy move, but I appreciate you putting all this focus on it and really fun to talk to you. That's very nice of you to say. You know, as a pundit, one of the dream scenarios is that you just sort of, every week you just say, when is someone gonna do something about this problem? And then somebody actually does. So, and it almost never happens. So it's huge for us. So, so thank you. It means a lot to us that you're, you're taking a swing. So yeah, Molly, thank you again. And yeah, I can't wait to talk to you as you get started with the new thing. Wonderful. Thanks, Casey. And congrats on the series. I'm glad you're going in this direction and it's really exciting. Thank you very much. Platformer is produced by Lindsay Chu and edited by Fitz Harris at Story and Sound. You can watch this whole episode on YouTube at YouTube.com slash Casey Newton. My email is casey@platformer.news, and we'll see you next week. Take your team from AI novice to AI native with Atlassian Rovo. Go to rovo.com to learn more today.