Sam Bankman-Fried Revisited

Published Nov 16, 2022, 9:46 PM

In May of this year, we interviewed Sam Bankman-Fried, the young billionaire philanthropist who started the crypto exchange FTX.

Last week, in a matter of a few days, FTX collapsed and filed for bankruptcy, and Sam resigned. It's unclear if customers or investors will ever get their money back.

In light of the news, we are replaying the episode -- and trying to figure out what to make of everything Sam told us earlier this year.

If you’d like to keep up with the most recent news from this and other Pushkin podcasts be sure to subscribe to our email list.

Pushkin. In May of this year, I interviewed Sam Bankman Freed for this show. Sam was a very young, very rich entrepreneur and philanthropist who ran a crypto exchange called FTX. And then last week, in a matter of just a few days, FTX collapsed and filed for bankruptcy. And you know, even to say that, even to say it collapsed and filed for bankruptcy in a few days, somehow amazingly kind of understates how fast and dramatic the fall of the company and of Sam Bankman Freed was. Until last week, Sam Bankman Freed had this persona of being like a good guy. He wasn't chilling some ridiculous future where crypto was going to solve all our problems. He was running a real business, an exchange for crypto t and all along. He said. His plan was to make as much money as possible and then give it all away, and it seemed to be working. He was only thirty and already he'd given away hundreds of millions of dollars. Now the money is gone, customers and investors may never get paid back, and federal prosecutors are reportedly investigating both Sam and FTX. This week as all this was happening, I listened back to my interview with Sam from May, and there are parts that I definitely find embarrassing in light of the new news, moments when I should have challenged him more, been more skeptical, and for that matter, my whole framing of the show seems absurd, over the top. I mean, I called the show Sam Bankman Freed wants to save the world spoiler alert. He did not. And yet, and yet, I did find the interview really interesting to listen to again in light of the news. Some parts of it still seemed true, but in a really different way, and at least one thing Sam said seems very very prescient. I'm Jacob Goldstein, and this is what's your problem? But today, let's be honest, it's more of a what's my problem situation. My problem is this, I interviewed a guy who was not at all what he seemed to be, and I didn't quite see it. So what we're gonna do for today's show is we're gonna replay the original interview, and at a few points in the interview, I'm going to pop in now with a quick update based on everything that has happened since now here is my conversation with Sam Bankman Freed as it aired back in May of this year. Before we get to the interview, I just want to take a minute here and set up this one big idea, this really useful intellectual framework that drives almost everything Sam does. It's called expected value. I try to use it a lot because I think it's started is the default correct way in some senses to calculate something. It's like if you're just trying to do a generic calculation, I think it's usually the right thing to use. You can understand expected value by understanding how Sam decided to start his company, the crypto exchange FTX. He was working as a trader making millions of dollars, and when he thought about starting FTX, he knew there was a really good chance it might fail. It might ultimately be worth zero. But on the other hand, if it succeeded, it could be worth tens of billions of dollars. So here is a slightly oversimplified version of how you would use expected value in this case. Say, Sam thought there was only a one percent chance that his exchange would be really successful, but that if it were really successful, it would make him twenty billion dollars. The expected value of starting the exchange is the probability of it being successful one percent times the value if that success happens twenty billion dollars, which comes out to two hundred million dollars a lot. So in twenty nineteen, Sam started FTX, and Sam told me there is a really important lesson here about expected value. One of the sort of takeaways that often ends up coming from really thinking hard and critically about expected values is that you should go for it way more than is generally understood. Go big. You should really go really big, even if you probably will fail and wind up with zero. That's absolutely right. And I think one of the intuitions for why that's the case, for why I think going big is often the right thing to do well if you think about it, like you know, you've got obviously a number of options available to you. Somewhere on the far right hand side of distributions, is like the best possible thing, meaning the really good outcomes. That's right, the best possible thing you could imagine happening, And the best possible thing is probably really good. You know, it's probably orders of magnitude bigger than whatever you're sort of expecting to do. Right, it's not a little better, it's wildly better. It's almost unimaginably better, that's right. And if you're thinking about, well, if I found a company, how is it going to go? You know, you're probably thinking this might be a million dollar company, right, But the right hand distribution, the right hand tail of that is going to be a billion dollar company, and that's a thousand times bigger. And so in order for it to be justified to choose that decision, if you really do care linearly about money, if you really do think that getting that marginal dollars worth a lot, you know, even once you already have a lot of money, then it should lead you to think that that, you know, the best outcomes might be outcomes that have a ninety nine percent chance of failure, right, because in ninety nine percent chance of failure and a one percent chance of that billion is still that's ten million, and that's a lot. And so any time that light there is some nonzero and non negligible chance of a really really good outcome are times when you're gonna be incentivized more than seems natural probably to choose extreme outcomes. Okay, that was Sam back in May, and listening back now, I have to say this part kind of holds up, but in a really different light. It's basically saying you should make massive, long shot bets with a high probability of failure if the potential payoff is big enough. And you know, that's clearly what Sam was doing for the first part of his career when the price of crypto was going up, and he kept making these big bets and having huge successes. Now we don't know all the details of what happened at Sam's crypto exchange FTX before it collapsed, but remember one of the key things that prosecutors are apparently looking into is FTX allegedly loan customer funds to a trading firm also controlled by Sam making that loan. Deciding to make the loan, that is a kind of bet, right. It's a bet that the training firm will make profitable trades or investments with the borrowed money, a bet that the market will go up, the money will be returned, and nobody will get in trouble. It's possible that Sam decided that was a bet that had positive expected value, and he went ahead and did it. Okay, now let's get back to the original interview. At this point, we're talking about how expected value shapes Sam's thinking about philanthropy, and what he says is when he's thinking about giving away money, he doesn't just think about the world today. He thinks about like the whole future of humanity. How many people will live if we play our cards right, as a world, like in the ninetieth percentile outcome, how many people will live in the future. The answer is trillions, probably maybe hundreds of trillions. Right, It's thousands of times more than the number of people who have ever lived. And so that's just that's a huge factor. Right, Anything that we do that actually has packed on the whole future of the world is massively important. It's kind of a ridiculous way to think at some level, right, it just gets so big then, like you're just some guy with a lot of money at some level, right, talking about like trillions of people in the whole future of humanity. Like it gets weird, Right, it does get really weird. You should really stress uses and and think like, Okay, do I really believe this? Like do I really actually think that there's compelling evidence that like, you know, these numbers that I'm looking at are as big as I'm claiming they are, or am I kind of bullshitting myself on this? Like, you know, you should absolutely have some humility around that, but but it's not totally implausible. And there are examples of people people who we've heard of were very famous, and people who no one has ever heard of, who have had massive, massive impact on the world, who have had that massive multiplier, and so it's not totally implausible. And so you know, I think that, like, while we should absolutely have you know, ay dosa humility towards extreme outcomes, you know, we should also acknowledge that they can be real and that often the highest expected value things are in fact pushing directly towards them. Well, and in fact, you did hit the extreme right tail of the distribution in work, right, you did just get implausibly rich in a ridiculously short period of time. So at least on that one it worked. That's right, and I think it's that's certainly it's been a big update towards me in the direction of life. This stuff is plausible. Uh huh. So the fact that you got so rich so fast in crypto, does it make does it push your altruism toward like Welsh shit? If I could make twenty billion dollars in three years, everybody on Earth could in fact die from a pandemic or from some out of control AI and I should spend some of the money to try and reduce the probabilities of that. I mean, is it like that? Yeah, it absolutely does. I think it absolutely does make me think, you know what, like these you know, really extreme outcomes are probably plausible, and they're probably plausible enough that I should be taking them really seriously, you know, and that has pretty profound implications, I think for what we should be doing. So listening back to that section now, I do feel like there was a moment in there when I was starting to push in the right direction. It was when Sam started talking about the whole future of humanity thing and helping trillions of people, and I was like, come on, trillions of people, the whole future of humanity. That seems like a bit much. But Sam had a reasonable answer there. He said, yes, we should have some humility and thinking about things this big. But you know, look, he said, there are people who have had really really profound effects on humanity, and why shouldn't I, Sam Bankman freed try to be one of those people, and that made sense to me. It did seem reasonable, and I think that is part of what was so appealing and convincing about Sam. I think that is part of why I and so many other people really believed him and wanted to believe him. He seemed guided by this sort of straightforward logic. He seemed like a very smart, very reasonable person. In a minute, we'll have more from the May interview about Sam's role as one of the biggest donors to democratic political candidates. Now, let's go back to the show as it originally aired in May of this year. Sam told me he's given away about two hundred million dollars so far, which obviously is a lot, but it is also somewhere around one percent of what he plans to give away eventually. His giving has been broad anti poverty, animal welfare, healthcare, but he has started to focus on a few areas. One of the biggest is pandemic preparedness. That is a category that fits right into that expected value framework. You know, a low probability but super deadly pandemic is worth spending a lot to prevent. Another place, where he's been giving is politics. Sam was one of the biggest donors to President Biden's twenty twenty campaign. More recently, he donated over ten million dollars to support a candidate in a Democratic primary for a congressional seat in Oregon, largely because that candidate wanted to focus on pandemic preparedness. The primary was held just last week, and Sam's candidate lost by a lot. I think that there are a lot of takeaways from it, and you know, I think that might do it again. I would do it a bit differently than last time. But you know, fundamentally, I think it was a well fought race. I think that you know, he had a real shot. And you know, going back to this discussion of expected values, right like, if you're donating blagal races, that's that you think your candidates are ninety nine percent to weigh, you're almost certainly doing something right because that person doesn't need your money exactly. You should be donating such that you think that you have a pretty substantial chance of losing. And you know, I first stand by that. Do you expect you'll give a lot of money in the twenty twenty four election cycle? I would guess. So, I don't know for sure, it's going to depend on who's running, but you know I would guess so, well, let's say Donald Trump runs for president. Would that cause you to probably give a lot of money to the person who's running against him. That's that's a pretty decent guess. And you know, I think that I'm going to be looking a lot less at like political party um from that perspective, and a lot more about you know, uh same governance Like that is you know at it's for the thing that I think I care the most governing governance. I think the United States has both a big opportunity, big responsibility to the world to shepherd the West in a powerful but responsible manner, and that everything that we do there has massive, massive ripple effects on what the future looks like. You've talked before about being surprised at how little money is in politics. It is quite small, the amount of money that is donated relative to how much money the government spends. Right, does that lead you to want to donate a lot? But does it follow from that that, like, you'll probably donate a lot of money? It follows that I might in the end. That's basically what I think, um, I think that like there are in some ways, there's you know, in some ways surprisingly little money in politics and given sort of the skipe of its impact. Note that doesn't necessarily mean they're good things to do, um donating politics right, Like it might be that sure, but like how are you actually going to do anything? You full? You know, maybe that's how it turns out, but but maybe not, you know, I mean, it's not necessarily the case that more money can have a meaningful effect on the outcome. That's right, It's not necessarily the case, but it certainly gestures a little bit in that direction. I mean, I imagine you have some probability distribution in your mind of how much money you might give in the next election cycle, Like give me some number I would guess north of one hundred million UM. And you know, as for how much north of that if I don't know, you know, it really does depend on what happens, Like it's really dependent on exactly who's running where for what like like these these are are a super contingent thing. But but yeah, I think that gives maybe some sense of what the what the sort of like scale might be here more than one hundred million sort of spread across many organizations, but towards the twenty twenty four election. So if that's a floor, what's the ceiling? Like a billion? Might you give a billion? Yeah? I think that's a decent life thing to look at as a as a sort of like I mean, I would hate to say like hard ceiling, because whom knows what was going to happen between now and then, But as like at least sort of a soft ceiling, I would say, yeah, okay, So so the ballpark is like one hundred million ish to a billion ish, with again a lot of caveats on this, and you know, there's a world WHI should end up being close to zero if they're you know, if things just work out such that there isn't Is there much I'm excited about? Like that seems like a very low probability to me, based on what I know about you in the world. Yeah, it's I think it's I think it's very low that it's actually gonna end up being zero. That does seem pretty unlikely. Yeah, a billion seems way more likely than zero to me. I think it's really right. Um, So I think the most anybody gave last time, if I have the right numbers is two hundred and fifteen million. That's for the twenty twenty cycle, the last presidential cycle. It seems like you'll probably give more than that. Based only on what you've told me, I think it is eminently possible that I think that that would not surprise me. Just to pop in here again in the present in November. Back in May, when this interview was published, this part of the interview actually made a little bit of news. NBC News and Politico both reported based on the interview that Sam might give a billion dollars in the twenty twenty four campaign. So that happened, and then some time passed, and then in another interview just last month in October, Sam walked this part of the interview back. He said it was a dumb quote to give on his part, And at the time when when I saw this news last month, I thought it was just, you know, that his philanthropic priorities had changed. But now, in light of what we know now, it seems like maybe Sam knew by October that he just wasn't going to have a billion dollars to give in the next presidential cycle. One other thing pops out at me. From this part of the interview, and that is the moment when Sam says his giving for the twenty twenty four election might be close to zero, close to nothing. And I just dismissed that idea out of hand. But of course, given what we know now, the situation now, it might be the most clearly true thing he told me in the whole interview. Okay, we're gonna go back to the interview now. We're going to talk more about philanthropy and also a little but in retrospect, not enough about the relationship between his political donations and government regulation of the crypto industry. I've heard you say that you know, at some points over the next few years, you hope to find opportunities where you can spend give away like a billion dollars really quickly. What are some of the places you think that might happen. I mean, the election, the twenty twenty four election is clearly one. What are a few others? I think pandemic prevention is potentially one of them. I think you look at like how much would it cost to you know, really definitely prevent the next pandemic or you can never definitely prevent it, but to have you know, a really good shot at it. I think you're probably talking tens of billions of dollars, which is crazy that that governments aren't spending that money, right, that's right, that really should be government spending it. And part of this might be working with governments that because it's not that much like if truly if they could reduce the risk of a pandemic by half, say for thirty billion dollars a year, like, do you actually think that is that true? Or fifty billion? I think I think something like that's probably true. I think something that's like not too far off from that order of magnitude, and that probably you know, by the time you're talking about you know, many tens of million billions of dollars um. You know, that's something that you're probably going to need to have government stepping in on. But you know, I would be happy to throw in a fair bit to health facilitate that. So I should actually I should have asked this earlier. But to what extent are your political donations a pandemic prevention strategy? So I think most of them have been so far, and you know, going forward, like there may become other policy, you know, things like AI policy, that that that that you end up being really important. And so it's not to say that like pandemics are the only things that are ever going to matter to me policywise, but that has been the big one so far, and it's the idea there, like tens of billions of dollars a year to significantly reduce the risk of another pandemic is not that much for the government, but it's more than you have, right, So you need a lever You can't you can't actually spend all of your money and meaningfully reduce the chances of another pandemic. And so if you can use political donations to elect candidates who want to spend money to prevent a pandemic, that works, that's right. So you're giving lots of money to political candidates. You're also doing a lot of work to shape regulation of crypto in the US. Tell me about the overlap between those two things. So most of the giving has not been done with crypto in mind, and I have been doing a ton of policy engagement on that, but that's mostly going to DC and talking with policymakers. I mean, here's the narrow version of the question. Is part of what you want from your political donations, some particular outcome in cryptoregulation that is not a big part of it. That is where we left the main part of the interview back in May, and when we zoom out now and look at what happened at FTX, you know, what Sam did is still really hazy. The details will probably emerge over time, but for now, one big question that is really interesting to me is was the basic altruism story Sam was telling true. Was he really trying to earn as much as he could to give it all away and help humanity or was that whole story a lie. I don't think we know the answer to that for sure yet, but it is possible that in spite of everything, that is really what he was after, That was really his intent. And if that is true, if he did mean to do great things in the world, and if he did do some of the bad things that have been alleged in order to do those good things, well that is a very complex, very human story about good intentions and bad decisions and ultimately, maybe Hugres He'll be back in a minute with the Lightning Round. Here is the Lightning Round with Sam Bankman Freed as it originally ran back in May, several of the answers played pretty differently now in light of the collapse of FTX. Let me just let's just do a lightning round a few quick questions and you can answer them fast. What's the least rational thing you do? Least rational thing? And I spend way too much time like aimlessly browsing my Facebook fee. Is it true you still sleep on a bean bag chair? And if so, why I did last night? I do do many nights. Um, it's I find it. I kind of I don't know, It's what I'm used to. Is honestly just part of the answer there. It's like it's what feels natural for me. If everything goes well, what problem will you be trying to solve in five years? I would say, the details of how to of what to prioritize for pandemic prevention funding with you know institutes that have been set up and are you know online and get a ton of capital into it and you know really great eight teams who are are devoting themselves to building it out. So the dream is you'll be like deep in the weeds figuring out how to prevent a pandemic. I've seen in other interviews You're doing lots of different things during the interview. I couldn't actually tell if you were doing other things during this interview, But were you and if so, what were you doing? As playing game of Storybook Brawl. I say the name of the game again, Storybook Brawl. How did they? I took second place out of eight, could have been worse, and I apologize. I do have to haul off. Okay, last one. What's one piece of advice you'd give to somebody trying to solve a hard problem. One piece of advice I would say, I just keep going, Just keep going, stuff by stuff, you know, try and solve it, but by a bit, and you know, eventually, hopefully you'll get there. That was my interview from May with Sam banksman Free. At the time, he was the founder and CEO of the Crypto Exchange FTX. Last week, f filed for bankruptcy and Sam resigned from the front. Today's show was edited by Robert Smith, produced by Edith Ruslo, and engineered by Amanda ka Wong. I'm Jacob Goldstein, and we'll be back next week with another episode of What's Your Problem.

What's Your Problem?

Every week on What’s Your Problem, entrepreneurs and engineers talk about the future they’re trying  
Social links
Follow podcast
Recent clips
Browse 143 clip(s)