Ethereum’s Founder on What Crypto Can — and Can’t — Do

When most people hear “crypto,” the first thing they think of is “currencies.” Cryptocurrencies have skyrocketed in popularity over the past few years. And they’ve given rise to an entire ecosystem of financial speculation, get rich quick schemes, and in some cases outright fraud. But there’s another side of crypto that gets less attention: the segment of the community that is interested in the way the technology that powers crypto can decentralize decision making, make institutions more transparent and transform the way organizations are governed. That’s the side I find far more interesting. There are few individuals as central to that latter segment of crypto as Vitalik Buterin. When he was still just a teenager, Buterin co-founded Ethereum, a decentralized platform whose token Ether is the second most valuable cryptocurrency today, surpassed only by Bitcoin. But the vision behind Ethereum was that the blockchain technology could be used for more than digital money; it could create a sort of digital infrastructure on top of which organizations and companies and applications could be built — ostensibly free of centralizing structures like banks and governments. Over the last decade, Buterin has become arguably the core public intellectual on the nonfinancial side of crypto. His new book, “Proof of Stake,” is a collection of long, thoughtful essays that taken together lay out a vision of crypto as a truly transformative technology — one with the potential to revolutionize everything from city governance to voting systems to online identity. I myself have dueling impulses about Buterin’s vision. On the one hand, I believe many of our governing systems and institutions are badly in need of the kind of reimagining he is engaged in. On the other hand, I’m deeply skeptical of whether the issues Buterin and his ilk are focused on are actually technological problems that blockchains can solve. So this is a conversation that sits squarely within that tension. Mentioned: Seeing Like a State by James C. Scott Book recommendations: The Rise and Fall of the Third Reich by William L. Shirer Harry Potter and The Methods of Rationality by Eliezer Yudkowsky Algorithmic Game Theory by Noam Nisan, Tim Roughgarden, Eva Tardos and Vijay V. Vazirani Thoughts? Guest suggestions? Email us at [email protected]. You can find transcripts (posted midday) and more episodes of “The Ezra Klein Show” at nytimes.com/ezra-klein-podcast, and you can find Ezra on Twitter @ezraklein. Book recommendations from all our guests are listed at https://www.nytimes.com/article/ezra-klein-show-book-recs. “The Ezra Klein Show” is produced by Annie Galvin and Rogé Karma. Fact-checking by Michelle Harris, Mary Marge Locker and Kate Sinclair. Original music by Isaac Jones. Mixing by Sonia Herrero, Isaac Jones and Carole Sabouraud. Audience strategy by Shannon Busta. Special thanks to Kristin Lin, Kristina Samulewski, Will Wilkinson, Alex Tabarrok, Glen Weyl and Nathan Schneider.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:04Ezra Klein?

  I'm Ezra Klein. This is the Ezra Klein Show.

• 0:22Ezra Klein?

  So forgive me, I'm going to take a moment to set this conversation up. It is a conversation about crypto, but it is not really about the part of crypto people are most familiar with. So I want to give some context. So there's crypto as this frothing speculative, hyper-financialized market. Crypto as an investment where the promise or the hope is that over time, the line just keeps going up. And anyone with the wealth and the courage and the foresight to hang on will become a millionaire many times over. I find that crypto pretty easy to have an opinion on. You can go see my conversation a few months ago with Dan Olson for more on that. But there's this other side too, this more idealistic side, the side of crypto that is not

• 1:06Ezra Klein?

  just interested in, but truly obsessed with. Blockchains and protocols as a way of bringing governance and community into the digital era, as a way of unlocking new forms of cooperation. There's an idea from the philosopher C.T. Nguyen, who's been a guest on the show, actually. He makes a distinction between object arts, where the value of the art lies in the object itself, how beautiful it is, how intricate it is, what we think of it, and process arts, where the value of the art lies in the audience's appreciation of what the art makes them do, of the process it gets them to engage in or allows them to engage in. C.T. Nguyen is talking there about games, but I think it's actually a very good way of thinking

• 1:49Ezra Klein?

  about this divide in crypto. Some people, they're interested in crypto because of the thing crypto creates at the end of the chain. They want their ape cartoon to be worth $5 million. They want their Bitcoin holdings to make them rich or to be really useful. They want this financial tool to make them a lot of money or to help a lot of people. Success there is judged by what the final crypto object ultimately does. And that's object crypto. And it gets most of the press and I think is more comprehensible. But there's this other side of crypto that is really important to why the ecosystem has been so vibrant for so long through so many crashes. And it's full of people interested in the processes crypto can enable.

• 2:31Ezra Klein?

  The way it can decentralize decision making or create new forms of transparency, or bind a whole community or project, or even company, to rules and protocols that can never be broken. They obsess over quadratic voting mechanisms and coins tied to city governance and decentralized autonomous organizations. And when you're looking at these things, they often seem really complicated and inefficient. And it's only when you spend a lot of time talking to people that you realize that that complexity, that inefficiency, that level of participation and creation is actually what is so obsessing about it. And these are the people who stick through crypto's low periods, like the one we're in now. These are the people who are trying to make a future for crypto

• 3:16Ezra Klein?

  that is more interesting than it's just some payment infrastructure operating in the background of the internet. The central figure on the side of crypto is Vitalik Buterin. Buterin co-founded Ethereum. He wrote the white paper for the ideas behind it when he was just a teenager. And his insight, and it was a big one, was that if you could program a currency like Bitcoin that cut out the need for a central authority, then you could use that same cryptographic technology to make almost a programming language that could then program any kind of digital agreement or contract and bind anybody who agreed to it in almost any digital way.

• 3:59Ezra Klein?

  And that meant that it could do almost anything. Bitcoin can be digital money, but Ethereum can be digital infrastructure. It can be the structure, a binding structure of how people cooperate online. And Ethereum took off. Its cryptocurrency Ether is the second most valuable, behind only Bitcoin. And Ethereum is just much more central to the grand visions of how crypto can change or rebuild society than Bitcoin is. And Buterin himself is a big part of that. The founder, or founders of Bitcoin, Satoshi, totally unknown. We have no idea who he is, she is, they are. But Buterin is here. He's Ethereum's benevolent dictator, and he's

• 4:40Ezra Klein?

  become something like the philosopher king of crypto. He writes long, thoughtful essays on new governance strategies. He co-authors white papers with economists. He's on the cover of Time Magazine's first NFT issue. And he's a compelling, unusual person. He's a digital nomad who lives out of a backpack and seems to think about really nothing but blockchains and what can be built on them. When people want to imagine crypto being more than a frothing speculative get-rich-quick market, Buterin and his vision is often the vehicle they do it through. And Buterin just guided Ethereum through a massive technological change. Like Bitcoin, Ethereum was built on a protocol called proof of work, which is just unbelievably energy-intensive. For years, he's been trying to move it to proof

• 5:23Ezra Klein?

  of stake, which is estimated to cut energy use by more than 99%. Just an insanely hard technological challenge, because you are shifting a currency, a programming structure that is currently running, to a new technological platform. And if you break something along the way, you can lose everything. But they did it. They did it. The merge worked. And I would have thought that after the merge, given that it's a much stronger platform for Ethereum to be on, the value of Ether would have gone up. But it didn't, surprisingly. It went down. But Ethereum is now on this more stable, much less energy-intensive platform. And Buterin sees this as a really important step towards building this crypto future.

• 6:08Ezra Klein?

  He's releasing a book, laying out his vision. It's called Proof of Stake: The Making of Ethereum and the Philosophy of Blockchains. And it's that philosophy of blockchains, this idea that blockchains enable new philosophies and approaches to governance, that really interests me. I find myself caught between two warring impulses when I spend much time in the world of crypto's idealists. One impulse is that governance will become digital at some point. We are not going to govern forever the way we govern now. And it's very possible that blockchains will be a key technology there. And maybe in that transition, very new worlds become open. To just be the guy sitting around telling people that nothing can change is

• 6:50Ezra Klein?

  to, in a sense, embrace stagnation. And you're probably right in any given year, but over long periods of time, eventually you're going to be profoundly wrong. But that sentence is mixed for me as somebody who covers governance and institutions with frustration at what often seems to be a real detachment from why institutions fail today, or why things don't get fixed, or whether you can properly represent society in code. And really, the core of this tension for me is the word trust. Crypto really claims to solve problems of trust. And I often think they get almost exactly backwards, what the problems of trust are, and how most people actually define trust. And so I wanted to bring Bhutan on

• 7:31Ezra Klein?

  to talk about what he imagines crypto can become, and also get some answers to the questions that keep nagging me about that vision. As always, my email, [email protected]. Vitalik Bhutan, welcome to the show.

• 7:48Speaker 01

  Thank you very much, Ezra. I'm happy to be here.

• 7:51Ezra Klein?

  Congratulations on the merge. That's a, that's a big piece of business.

• 7:55Speaker 01

  Yeah, no, it's a really big milestone for the entire Ethereum ecosystem, really. I think a lot of us are just so happy that this thing that we've been working with in a lot of cases for almost a decade is, you know, finally here.

• 8:13Ezra Klein?

  So one thing that has been striking to me about the merge is that it was really well done. I mean, this is something I've been watching to see if it would happen. It did not break the system. You all actually did deal with the huge, both actual and I think, political vulnerability for Ethereum.

• 8:30Ezra Klein?

  And one of the things that was surprising to me about the aftermath of it is that when I went to look at the value of Ether last November, at the sort of height of the crypto bull market, it was tracking around $5,000. For Ether, the other day when I looked, post-merger was around 1,400. That was unexpected to me in the aftermath of doing something that big and so obviously beneficial for the future of the platform. So why do you think it is?

• 8:56Speaker 01

  That is a good question. I think the post-merger price drops, I mean, really surprised me. They really surprised everyone. Like, there is an expression, buy the rumors, sell the news. And I've never really known like how seriously that expression should be taken, or like how often that kind of pattern actually plays out in real life, right? I guess the reason why something like that would happen psychologically is that everyone gets very excited and confident running up to an event. But then when the event actually happens, it just feels in retrospect almost less significant than what everyone had been expecting going in. And then there's this wave of disappointment in the price if it ends up dropping a bit.

• 9:41Speaker 01

  So if you look at it just from a price point of view, like ETH and even like the ETH BTC ratio, they're still considerably better off than they were a few months ago before this kind of merger cycle started. But at the same time, it does feel like maybe that kind of buy the rumors, sell the news pattern is what just for whatever reason ended up happening here. So that's one explanation. The other explanation is that crypto does, I think, also actually to a lot of people's surprise, correlate with a lot of the broader tech markets. And I believe the broader tech markets have also not been doing very well for the last couple of weeks. And, you know, obviously, there's energy markets.

• 10:24Ezra Klein?

  It's a bad time for markets in general.

• 10:25Speaker 01

  Yeah, exactly. There's just lots of energy and geopolitical and macro, whatever stuff. And that stuff affects crypto too. So.

• 10:34Ezra Klein?

  Let me ask you about that side of it. When I go back to a lot of the longtime arguments for cryptocurrencies, and I mean, I'm not talking about all the way back right at their launch. I mean, Jack Dorsey saying, we're in the age of hyperinflation, you know, time to get into crypto, you know, just, you know, over the last year. There's a real belief that a reason we needed cryptocurrencies was central banks. We're doing too much easy money, then over time, they would inflate their currencies and that as inflation took hold, you'd really need crypto, you'd want crypto as a hedge. Crypto would not be correlated with the inflation that you saw in more traditional currencies, or if it was correlated with it, it would go up during inflation because it would be safer.

• 11:16Ezra Klein?

  And we have not seen that at all, that as inflation has risen, cryptocurrencies have plummeted. They've been correlated in a, in a way that's really quite worrying. What do you think the crypto community missed there?

• 11:29Speaker 01

  I mean, I think the correlation between the cryptocurrency prices and broader tech and stock markets is definitely something that has surprised a lot of people. And you're right, like two or three years ago, many people really did believe in the anti-correlation narrative. I would say, yeah, I mean, it's hard to come up with an explanation that like I feel really confident is correct because it feels like I'm just like kind of backfilling an entire story. So like what's realistically only one or two years of evidence. But like if I had to do that.

• 12:02Ezra Klein?

  Welcome to my job.

• 12:05Speaker 01

  Of course, if I had to do that, I would probably say, one way to think about it is that the narrative of crypto is that it is this kind of like very safe and robust shelter from the storm kind of thing. But if you look at where crypto is today, just in terms of its trajectory, it's still this very early stage, untested and risky thing, right? You know, there is a case for why cryptocurrency could replace gold, right? Gold has an $11 trillion market cap. Most of that is not industrial or jewelry related. It's just investment value and the younger generation just like honestly thinks.

• 12:46Speaker 01

  Gold is pretty lame and is much more comfortable with digital things. And you know, you could easily see cryptocurrency kind of flip into that. But that's a like multi-decade long narrative that would probably take up until, you know, 2040 or 2050 to play out. And in the meantime, cryptocurrency is still like ultimately this nascent baby, where lots of crazy things that are typically associated with nascent baby concepts are playing out. And there's still huge events happening regularly that increase or decrease the chance that crypto is kind of going to make it in some big way in the world by a factor of, you know, two or five.

• 13:26Speaker 01

  And so people today are largely kind of mentally putting it into a more disruptive tech stock like bundle than a stodgy, stable, dependable gold sort of bundle. And the prices reflect. So that'll be one way of putting it. Another possible narrative, of course, is that cryptocurrency was never about providing that kind of stability. It was more about providing stability in the sense of, let's say, protection against asset seizures in places like Argentina or like third world dictatorships or Russia or and places like that, where the political situation is much less friendly to people.

• 14:11Speaker 01

  And so it's more about these sort of more political properties of it. And if that was true, then, I mean, of course, well, that narrative also doesn't really fit the facts perfectly either, right? Because we saw a lot of geopolitical instability in 2022 and crypto prices went down a lot in 2022. We saw a lot of geopolitical instability in 2016. And I think on the day of Brexit, Bitcoin went up. And, you know, it's, it's hard to fit things to this. I think realistically, different people value and believe in cryptocurrency for different reasons. And people value and believe in blockchain technology for different reasons. Sometimes people even have separate narratives for the two.

• 14:54Speaker 01

  So it's still early, but like those might be some of the, I guess, possible stories that I would guess at.

• 15:02Ezra Klein?

  I'm really interested in the political dimensions of crypto and to signpost the conversation, I think we'll probably spend most of our time there. But I do want to ask about another piece of the crypto market we saw over the past year, a couple of years. I've always been very conflicted on crypto, where a lot of the idealistic arguments for it and the experimentation in it, I find really inspiring. And then some of the VC culture around it, the speculation around it, I find really concerning. And one thing that really turned me emotionally, and actually substantively, on a lot of crypto over the past year was watching the crypto companies spend huge sums of money advertising to bring in people who they knew did not

• 15:46Ezra Klein?

  understand these currencies or these markets. I mean, I think of the Matt Damon ad telling you to invest because fortune favors the brave. I think of the Larry David ad, which was explicitly telling you to invest even if you don't understand it, because the people who didn't understand it were going to miss out on the gains. And I mean, that was clear to me as somebody who studied and reported on bubbles, like you're looking at a top of the market kind of thing there. And then I'm sure a lot of people did get in. And we know a lot of people lost a lot of money, many of them who really couldn't afford to do it. And it was all just clear, like it was being played out, I thought, with a very high level of cynicism, not by you, but by people who often talk in very starry eyed

• 16:26Ezra Klein?

  ways about what all this can do. I'm curious how you understand the effect that the financial run up, the markets run up in crypto had on the crypto culture during those boom years?

• 16:39Speaker 01

  Yeah, I think I personally would not blame anyone for being suspicious of the crypto space, because I think most sane people within the crypto space are themselves, ourselves suspicious of large parts of the crypto space, right? And you know, sometimes this is even not true with like different groups being suspicious of each other, right? Like, like, I'm going to be telling you, you know, which parts of the crypto space I think are a scam. But then you're, if you ever get a Bitcoin maximalist on the show, then they're going to tell you that I and Ethereum are a scam.

• 17:16Speaker 01

  There definitely is a large part of the space that is very money oriented and that's very growth at all costs oriented. And that side of crypto is a side that absolutely comes out during the bull markets. And a lot of people in crypto, they even welcome the bear to some extent because the bear is a kind of a return to sanity when the money types no longer have something to cheer about every week on Twitter. And so the only thing that you can cheer about is actual progress in the technology. And so people start caring about that much more again. So it's a real dynamic and there's definitely parts of the crypto space that are much more

• 18:01Speaker 01

  like that.

• 18:02Ezra Klein?

  So anytime you make a cut like this, you're going to leave out a lot of people who are on both sides of it. But as a thought, it often seems to me that there are a set of people who are very interested in the crypto technology because of the products it can create, and in particular, in this case, because of the money it can create. And then there's also this set of people who seem drawn to it very deeply for the processes the technology can support. There almost seems to me to be a set of crypto people who are primarily there for the output, and then a set of them who are primarily there because they think what is revolutionary

• 18:44Ezra Klein?

  is the input and the modes of participation and the modes of governance. And I often think of you as leaning towards that latter side. But for a lot of people for whom that would be obscure, why? What about crypto is interesting from a governance perspective? Why should I understand this technology as some kind of political enabler?

• 19:07Speaker 01

  Sure. So I think of blockchains as being like a Lego for mechanism design. Right? So what I mean by that is that blockchains create, give you these tools that let you create computer programs that directly control money.

• 19:26Speaker 01

  Computer programs that directly hold money. The ability to collect inputs from particular people and use whatever function of those inputs that you want and do things with assets or potentially other kinds of things in response to those inputs. So basically, whatever system of rules for the economic interaction or potentially even some kinds of financial and non-financial interactions that people wants to have. Blockchain are great at encoding those rules.

• 20:05Speaker 01

  They're great at implementing those rules without requiring you to trust any one single actor. So in that sense, they put everyone on the same playing field, right? It's like whether you're an elite programmer with decades of experience or if you're an anon, if you just publish a contract, then other people can go and inspect it. And if it looks correct, then you know, people can choose and they can go and start participating. Whereas in any other kind of system, you know, basically the only one who would really be in a position to like actually build new mechanisms would be someone who has

• 20:46Speaker 01

  a lot of like some kind of trust or reputation so that people would be actually willing to trust them. So this can be used for a lot of things, right? So you know, you can use this to issue assets. We can experiment with different kinds of financial infrastructure, like different ways for exchanging between assets. So Uniswap would be one example. It's this very simple decentralized exchange that launched in 2018 that just made it very easy for users to trade from one asset to another. And potentially you can do lots of like even more advanced stuff. And all of these things have this theme of allowing

• 21:30Speaker 01

  large groups of people to collaborate with each other and have rules around that.

• 21:36Speaker 01

  Collaboration without needing to trust a particular person, both to kind of not rug everyone and steal the money, but also to just like stay online and not get bored, right? Like if the Uniswap team got bored and quit tomorrow, Uniswap would keep running. And anyone who is building on top of Uniswap today doesn't have to worry that Uniswap is going to disappear tomorrow. Right? And so all of these properties make it like really easy and really powerful to just build all of these different structures that talk to each other. And I think the thing that excites me is just a question of like, well, you know, what kinds of things could you build? What kinds of things does it make sense to build? What kinds of things can we learn from people trying to build

• 22:19Speaker 01

  various projects and some of them succeeding more than others? And I guess what excites me about continuing to work on Ethereum is that it's both a decentralized open blockchain platform and it's a system that supports that's a general purpose programming language. So it actually is like general enough that people can go and build all of these things.

• 22:45Ezra Klein?

  There's sometimes a shimmering quality for me to some of this thinking where you've described in different places a possible future for blockchain technology, which you call the Linux of finance, right? It's a underlying internet utility structure that has made a certain set of things more efficient, that runs in the background of some things, not all of them. That could be a pretty big business, or at least enable a lot of quite big businesses as Linux has. People talk about digital gold. And when you give a lot of the examples you just gave, like Uniswap, that's what I think of, right? That something that is running there on decentralized networks and could potentially be part of particularly

• 23:28Ezra Klein?

  the financial infrastructure of digital life. But then it also, there's something about blockchain, separate from other technologies. People don't talk this way about APIs exactly. There's something about blockchain that leads to, enables, encourages a lot of ideas about how you would rebuild society quite fully on top of its structure. People think about, you know, coin cities and they think about, you know, new nations and they think about upending virtually every major institution from central banks to governments to, you know, big corporations like Facebook you can think of. And so tell me a little bit about that jump. There's something here about imagining this acting far beyond the walls of a computer,

• 24:13Ezra Klein?

  far beyond what you might imagine something that can execute code could do. Why? What is it about blockchain that makes it appealing as a sort of imaginarium for rebuilding society on top of it?

• 24:26Speaker 01

  I think the idea here basically is that if blockchains do become successful enough and they survive long enough, they have a good enough track record of actually being the base layer for many kinds of interactions. And you know, fast forward a couple of decades into a future where it's just considered normal for there to be, you know, trillion dollar assets that are managed on Ethereum, for people to, like say, instead of having Facebook accounts that they use to log into everything, have Ethereum accounts that they use to log into everything. And more and more stuff gets built, then you know, you get this question of if you're

• 25:07Speaker 01

  even a government and you know, you want to try to build some aspect of your society in a way that's more open and democratic in some way, possibly more accessible to the outside world in some way, more like credibly neutral in some way. So you might imagine you might be one of those governments that not a lot of people trust you, but if you go and build some thing on a blockchain, then you know, whatever rules you write are going to end up constraining you. And at the same time, whatever rules you write are going to give confidence to whoever participates in that, right?

• 25:48Speaker 01

  So if you end up issuing some kind of like central bank digital currency, or if you end up issuing assets, like whether it's a city coins or potentially tokens, you know, representing land ownership or potentially things like, identity is another example. Like I don't think, like basically, if people start having identities that are these kind of decentralized blockchain based identities, then it might start to even make sense for governments to, instead of having completely separate identity packages, but essentially start issuing what we call attestations, right? Like Cryptographically signed claims that say this existing

• 26:32Speaker 01

  identity is also a citizen of this country that has these other attributes. Right. And so instead of the government being an entire root infrastructure provider, the government has this kind of more circumscribed role where in this particular case, that's basically providing this service of verification and trust. Right. So if blockchains do manage to become like actually stable and reliable to that extent, then we might end up actually seeing a future where it just makes sense to build stuff on top of them because you just get a lot of interoperability from all of the other applications that are based on blockchains. You get this kind of automatic

• 27:14Speaker 01

  auditability. You get a lot of this, you know, automatic verifiability and trust properties. So that's kind of the more wild eyed dream. And that corresponds to the, I mean, I forget what, what, what name I used in the triad. It was like, the, the triad was either blockchains are Esperanto, or blockchains are Linux, or blockchains are the internet. That was the, that was the third one. Right. Like people use the term the internet of value, like this global neutral network, except instead of just being for transmitting packets of data, it's for keeping track of value or these kind of longer term records. Then like, that's what blockchains as the internet of value world would start looking more like.

• 28:01Ezra Klein?

  I want to ground this in a couple of use cases, but, but before I do, I want to pull out a word used a couple of times there that I think is really important, which is trust. And as I understand it, a lot of the argument about what problem is being solved in terms of transactions, in terms of financial markets, in terms of how institutions might be built, even in terms of how governments might be run, is trust. There's a view that crypto, for a variety of reasons, is able to create trustworthiness, even in contexts where trust would, would normally be hard to come by. And I want you to poke holes in whether or not I understand this correctly, because it often seems to me that what is being said here about trust, if you unpack it, is that what creates trust is transparency.

• 28:49Ezra Klein?

  Rule boundedness and a kind of predictability that you know what's going on. You know that the rules will be followed. There's not that much room for discretion where people could kind of pull one over on you. You know that the protocols are kind of going to automatically execute. And the idea is if you add those features to a lot of institutions in our lives, they might become more trustworthy. And if they became more trustworthy, the ways we could cooperate would magnify. Is that a kind of fair description of the vision here?

• 29:19Speaker 01

  Yeah, I would say so. I mean, I would say the like the main pushback that like I would anticipate and that I think pretty much everyone who's trying to actually do things in this space already agrees with is that obviously there are lots of things that have a lot of room for discretion. And there's a lot of things where if you just try to automate everything and replace things with open code as much as possible, then there's cases where that's like just not good enough. And so in those cases, blockchains definitely cannot like do or be everything. But you know, there are a lot of cases where if you want to give people some kind of guarantee

• 30:03Speaker 01

  that things will be done according to the rules that you specified, then the blockchains can be a good way to do that.

• 30:10Ezra Klein?

  Okay, I just want people to keep that in mind because I do think that's an important, both the pushback there, which which I share and will probably give voice to, but but also the the vision there because I think as we go through some of the cases here, that's a that's useful to keep in mind. So I want to start with one that I think is a pretty base layer case, which is remittances. How do remittances work right now? What are they? How do they work? And then how could crypto technology make them work better?

• 30:35Speaker 01

  Sure. So the concept of remittances, I mean, it generally refers to the pattern of people usually in wealthier countries, working and earning money and sending money back to their families in usually poorer countries. And this is a significant portion of GDP in a lot of places. It's something that a lot of families in, you know, places like Southeast Asia or like Africa or Latin America ends up relying on. And it's something that is notably inefficient with traditional finance in far from all cases, but definitely in at least some cases. So. Remittance system through the traditional banking system, they often either go through

• 31:20Speaker 01

  complicated intermediaries and it takes some time for the money to get through. There's various kinds of fees. Sometimes places are just not well-connected to global financial networks. And cryptocurrency, I think even today already, is offering at least some people a more efficient channel for doing that, right? So even today, there's a bunch of people in a lot of these places where the person who's working in like, let's say for example, the United States or Europe, they buy some crypto. If they're an enthusiast, then they would just like buy the Bitcoin or ETH or whatever else and they would hold it.

• 32:02Speaker 01

  But if they're just trying to minimize risk, then they would probably convert it to what's called a stablecoin, like either DAI or USDC are some popular ones. And then they would send that cryptocurrency to their family members in the poorer country. And in the poorer country, there would be some local cryptocurrency exchange. And the recipients would take those stablecoins, deposit them in the exchange and convert that into whatever the local fiat is and they would withdraw it. And in a lot of cases, the fees on this are lower. In a lot of cases, it's faster.

• 32:43Speaker 01

  And like, this is something that, like, there aren't huge remittance focused startups that are like making huge amounts of money on this, right? Like, this is a thing that I think people tried to do much more of back in the day. They're like, oh, you know, crypto remittance is a thing and so we'll just try to make crypto remittance startups. Like, I think instead, it's much more of a kind of DIY thing where, you know, okay, you get cryptocurrency, move cryptocurrency over, you sell cryptocurrency, and there's just like a whole bunch of exchanges in, on both sides of that process that are facilitating the transaction.

• 33:19Ezra Klein?

  So it's actually, in a way, the exact process you named there right at the end that makes me wonder about it a little bit. And let me offer two questions here, or two observations, maybe. One, we can make the remittances process much more technologically efficient before you ever get to crypto. I mean, it is, like a lot of the financial plumbing in the economy, quite bad and has not been updated in most cases for a very long time. But you can imagine a million ways to update it. And so I think you need a very clear theory of why none of them have happened. I have heard theories. You're dealing with pretty poor people here who are very scattered. So on any given transaction, there's not much money to be made in it.

• 34:02Ezra Klein?

  You're dealing with so many different countries and so many different approaches and people who might be refugees and all these different, very, very complex problems that it's actually hard to harmonize a system in a real way. And I've noticed a little bit of the same thing happening at the end of the game on crypto there, that in part for the same reason, there actually isn't a ton of money to be made by it. You end up with a quite complicated DIY system, which, you know, people could do, but they don't do in very large numbers now. And DIY systems don't tend to work out that broadly. So I guess one question here, and this is one that's going to come up for me a few

• 34:39Ezra Klein?

  times, is is the remittances problem a technical problem that is amenable to a technical solution? Or is it a problem of incentives, power, political structures, that the same obstacles that have made it very difficult to create something better, despite the fact that many people would like something better, also afflict crypto, where nobody really has the incentive to spend enough money to make something that will scale big enough that people who are not highly tech savvy are going to be using it as their main approach?

• 35:15Speaker 01

  I would say that it's definitely more about power and incentives and just like various path-dependent issues than it is about which technology is better in some kind of like platonic absolute sense. Like, I think it's often the case that trying to reform a system into something that works more effectively using new technology, in this case by new technology I just mean the internet, is more difficult than creating a new system that's just internet-native. And in this case, cryptocurrency is very internet-native, like it follows the kind of email and website-like pattern of like

• 35:59Speaker 01

  caring less, much less about which country you're in, just having like one global standard for addresses, for how to send funds, and I mean, you just have one global thing that everyone is plugged into. And in this case, it's just easier to get that started than to start from the remittance thing and move it into something optimal, especially once you take into account the political difficulties that happen whenever you end up reforming or trying to reform any kind of existing thing. So, that would be one argument that you could make. I mean, another argument, of course, cryptocurrency does have this

• 36:42Speaker 01

  kind of speculative aspect to it, and it could even be the case that like the fixed costs of building the infrastructure for the exchanges on both sides are basically paid for by speculators, and that creates enough volume and kind of kickstarts it. And that's what allows the fees to get low enough so that the remittance users and international businesses and other sort of more boring users like that are just able to piggyback off of that infrastructure and essentially get at very low rates. I mean, a third option, of course, is that sometimes, you know, financial transfers become difficult, not just because of efficiency, but more because of this various kinds of

• 37:27Speaker 01

  political factors, right? So, sometimes it's a particular country that gets repressive for whatever reason, sometimes it's like different countries that end up not trusting each other, then... You know, of course, this gets into the philosophical debate of basically, you know, on the one hand, you're improving efficiency and you're really improving a lot of people's lives by helping them just like have savings in US dollars, which is something that's not possible in a lot of cases versus, you know, on the other hand, there are times where there are restrictions are put into place by people and those people have various goals for implementing and maintaining those restrictions and cryptocurrency does end up interfering with that sometimes.

• 38:14Ezra Klein?

  There's an interesting tension here, I think, that I sometimes think of as almost like a Mott and Bailey dimension of crypto, where you'll talk about an argument being made that crypto can become the system, right? It can replace the dollar, which is something very, very big crypto people have said, like, I'm not making that one up. Or it will, you know, overturn the banks or overturn Facebook or revolutionize institutions. And then you'll press, or I will press. And there'll be a sort of, I'm not accusing you of this, but I want to note this dynamic and get your thoughts on it. There'd be a kind of a treat to, well, look, I'm not saying crypto is going to replace a system in countries that have strong institutions, because actually, that's a pretty high bar to clear. But when you're talking about Venezuela, you're talking about North Korea, maybe you're talking about

• 38:59Ezra Klein?

  people trying to flee Russia, that there, there's a lot of room for it there. And I think that's probably true. But there's just a really big difference between something that is a technology for getting around systems on the margin for people who can figure that out, and a technology for becoming the system. And I find that people sort of shift between which one they seem to be telling me crypto is likely to become fairly often.

• 39:29Speaker 01

  This is true. I think that the story that I would tell that kind of puts those two pictures into harmony with each other, it would be that this is an instance of leapfrogging, right? This kind of pattern where often less developed countries end up getting certain kinds of new cool things even faster than the wealthier and more established economies do, basically because for

• 39:55Speaker 01

  A newer technology to like really beat the older technologies in the network effect game. Like it can't just be one step better. It has to be at least 10 steps better. And so in places where the old thing is crappier, sometimes is the case that only there is there enough activation energy for the new thing to like really properly come in. Because in other places, like the new thing might be one step better, but like whatever one step isn't really enough to really radically inspire and motivate people to get over a network effect hubs. And so in the places where the old stuff works so worse, those are the places where the new stuff ends up working better.

• 40:34Speaker 01

  So that would be one argument for how both of those could end up being kind of in harmony with each other. Another way to put it, I think, is that I do think there is a very real dynamic where cryptocurrency and blockchain technology appeals to people in different parts of the world in different ways. I've actually gone to, you know, a lot of these places, whether it's like Africa or Argentina and Southeast Asia, you know, Ukraine. I was just there a few weeks ago. And in those kinds of places, the vibe does tends to be more pragmatic in the sense that

• 41:18Speaker 01

  people really understands that they have a specific need of, you know, I want to move money between the United States and my country, or I want to create a company with 10 people and I don't want to trust the legal system or any one of the people to manage the funds. Or I want to create a decentralized insurance project because, I mean, people actually do need insurance for like lots of various small scale things and then they go from there. But in richer countries, I mean, this is obviously a stereotype and a kind of like very broad kind of smothering of a complicated dynamic. It's more often the case that in rich countries, the vibe is more.

• 42:04Speaker 01

  You know, I want to use cryptocurrency to express my political values, or I want to, you know, enjoy gambling with three million dollar monkeys, or I enjoy being in one blockchain tribe and, you know, trolling on Twitter about how my blockchain tribe is the great one and the other ones are the scammers. And these kinds of, you know, higher level self-expressive motives. And it does seem sometimes the case that if you want to look for these kind of providing lots of genuine value for people today, uh, kinds of cases, then either poor countries or sometimes marginalized people within wealthy countries are the places that you want to look at.

• 43:20Ezra Klein?

  Let me ask you about another use case, which I've heard you talk about, which is intergovernmental digital currencies that could be used to track transactions or publicize transactions between government agencies and therefore reduce corruption. Can you talk about that idea and the way you think it would provide benefits?

• 43:40Speaker 01

  Sure. So here the core idea is basically you'd have a digital currency. So basically, digital cash issued by the central bank, like legally it could be a structure to be equivalent to a paper cash or some other format, except this would be limited to use within individual government agencies. So Federal government, local government, probably one level up and down the supply chain to employees and suppliers. And the idea would just be that all of the transactions that happen within the governments would just be done in the public and they would be traced. And people would be able to see what's going on.

• 44:23Speaker 01

  This would achieve the idealistic dream that a lot of people have of, you know, being able to like actually see where their individual tax dollars are going and what kinds of projects that they are contributing to. Idea there would basically be to also make it easier to just inspect the graph and make it more difficult for various kinds of corruption to happen. This is something that I have talked about to government officials in a bunch of non-Western countries, and there's definitely some where it gets some interest and positive attention. I think realistically, the challenge is it is a deep systemic reform of existing stuff, and deep systemic reforms of existing stuff are hard.

• 45:06Speaker 01

  And of course, that the kinds of people who benefit from corruption would definitely do their utmost to put the brakes on this kind of thing. But, you know, if you were starting, say, a new charter city in one of these countries, or you do have some kind of blank slate opportunity, I think it's definitely something that's worth always looking into.

• 45:27Ezra Klein?

  I think this is a really interesting case, because I think it gets at that earlier set of questions about what creates trust in an institution. So I want to interrogate it a little bit. So here's an observation, just as somebody who covers politics in America primarily, which is that over the past, it's called 70 years, our institutions here have just gotten way more transparent. We have C-SPAN, we have FOIA requests, we have much more public budgeting, we have GAO reports. We have, I mean, particularly for a journalist, my ability to track what is going on in government is far, far, far beyond what it would have been ever before. You know, we have visitor logs at the White House. It's actually quite a bit.

• 46:10Ezra Klein?

  It's not to this point of this currency where you can track every dollar, but it's further than it was in 1950. Over that same period of time, trust in those institutions has completely plummeted. And something I've often noted and written about is that when you look at the US government, it's often the less transparent institutions that are more trusted. So the military, police, the Federal Reserve, the Supreme Court, things and institutions that do not have nearly as much oversight tend to maintain a higher level of public trust. And even those institutions are suffering today as they become more transparent, as their internal workings become more known. And so it looks to me like there's an inverse relationship here, that when things get more

• 46:53Ezra Klein?

  transparent, or at least pass a certain level of transparency, the ability of the institution to operate internally begins to break down, or the ability of the institution's enemies or critics to publicize what is going wrong rises up. And the outcomes of that are not things people like as much as they think they'll like it in advance. The outcome of that tends to be less trust in institutions, not more. How do you think about that?

• 47:18Speaker 01

  Yeah, I mean, this gets into these challenging questions of governance. Even the attempts to create various forms of governance within crypto, like all of these DAO experiments, they often end up running into, you know, similar kinds of problems. Like where if you try to kind of put, you know, maximum global democracy in charge of everything, it often ends up making worse decisions than something that can still have some kind of separation of concerns. So it's a good point. Obviously, it's not an argument to kind of throw away transparency in its entirety, right?

• 48:00Speaker 01

  In this, yes, there are many forms of corruption that people do get away with, basically, because they benefit from obscurity. So I suspect the answer lies in the middle, and we can probably figure out more detailed pictures of what kinds of things it makes sense to have more private and what kinds of things it makes sense to have less private. And this is a Also one of those places where I think Ethereum style programmability has like really shows its power, right? Because it actually makes it possible to very easily try to create those different structures. And basically say, you know, okay, we're going to do this thing where this component runs according to these rules.

• 48:44Speaker 01

  And this other component runs according to these other rules. And then we're going to see how that works. And then someone might create a different system where you kind of flip those and you make the first component less transparent. And the second component more transparent and you get to see what happens. So actually, one very specific case where non-transparency in governance is a very good thing, right? And this is one that I've talked about in my writing a lot is secret ballots and voting, right? Because if you don't have a secret ballot, then you can like either threaten or bribe people to vote in a particular way. And so the theory is that to prevent that, not only do votes have to be private,

• 49:25Speaker 01

  but you need this very strong level of privacy where you can't prove to other people how you voted, even if you want to make that kind of a proof. And the hope is that, you know, by kind of creating this field of experimentation, we might actually have more of an opportunity to see what combinations of privacy in particular places and transparency in particular places actually end up making the most sense, right? So I think, yeah, in general, I'm, you know, definitely with you on the, just the core idea that more privacy good applied naively and especially cranked up to infinity eventually starts leading to all kinds of problems.

• 50:10Speaker 01

  And I think the thing that I'm more in favor of is basically that like, hey, you know, we have tools so that if we want more privacy, we can have a lot more privacy. We also have more tools so that if we want more transparency, we can have a lot more transparency. And you can have privacy of what individual decisions are plus transparency that some like rules are being followed. So whether it's money out equals money in, or whether it's this is actually a correct vote counting and all of the votes are being counted correctly and everyone who wanted to vote was given a chance to vote. And what we would actually be able to see better over time, which combinations of these kinds of things are able to work.

• 50:54Ezra Klein?

  It's very much my favorite thing about the crypto community is the amount of governance experimentation. But that's a little bit why, as a governance person, I want to press on this question of whether or not blockchains are the tool that one would want to use. I mean, I do think, I'm sure you know the old James C. Scott book, Seeing Like a State. Sometimes it seems to me to be like a seeing like a blockchain dynamic that emerges, where things that if you're going to work with a blockchain, you have to work with things that would be visible to a blockchain. And this idea of trust, I mean, I've said many times publicly that I think the crypto community has evolved a very unusual definition of trust, which is not the way most people use or experience the term or the sentiment, where it often seems to me that in crypto

• 51:38Ezra Klein?

  world, trust is about rules being transparent and rules being followed and having full confidence in that. And for most people, I think trust is about trusting that the outcomes are going to be fairly just towards you, you know, that if you get spoofed on your Visa, they're going to return the money to you or something. And so the reason I think this kind of question of would it be better or worse if we could track every dollar in the government is interesting, is that I think it's worth thinking about that from the perspective of, well, would that make government more efficient? And if it didn't, would it still make it more trustworthy? And I'll give another example. I was having a conversation the other day with somebody who is working fairly high

• 52:18Ezra Klein?

  up in the US government on, let's call it, scientific response to ongoing challenges. And something they're telling me they're doing, because Congress has not been appropriating money for some things Congress should very much be appropriating money for, is they were trying very hard to move money around within the boundaries of literal legality. So things that are more outdated in terms of the challenge they posed to us are getting a little bit less, and as much as possible is going towards the threats we actually face now. And it's not for me to blow up their spot here and go into too many details, but I hear stuff like that a lot. The difficulty with bureaucracies is that they are very rigid. And the people who we end up seeing as heroes later on are the people who are able to make

• 53:02Ezra Klein?

  them, by hook or by crook, sometimes by outright defiance, more flexible. And the crypto community seems to me to have a slightly dual consciousness on this, where on the one hand, there's a very, very widespread view, correctly, I think, that a lot of modern institutions have become rigid, bureaucratic, inflexible, hard to move, we need new things. And then a view that you're going to get that by making things that are in code more rigid, more trackable, more transparent, harder to make flexible, easier to punish somebody if they try to do something against the rules, or harder for somebody to do something against the rules. And that doesn't just seem to me to be a tension of programming. That actually seems to me to be a real question about these tools

• 53:45Ezra Klein?

  and their applicability to some of the challenges people want to have them solve.

• 53:49Speaker 01

  If you try to make every single decision maximally accountable to the entire community, that just often leads to worse results. And you need room for iconoclasts. You need room for crazy geeks that have weird ideas about how the biggest problem in the world might actually be AIs turning us into paperclips. Regardless of how we think about the issue, it's good that there's somebody putting a few cycles into checking it just in case that's a big deal. We need groups of very smart people who have space to advance particular visions about what kinds of public goods are really worth funding. We need even people who are willing to take risks, what I call entrepreneurial public goods,

• 54:32Speaker 01

  the intersection of goods that don't have a business model, so they're not going to get funded by traditional venture capital, and where their value isn't immediately obvious to a global public until long after they're done. And so they're not going to get funded by a perfect democracy. So I'm very sympathetic to the need for some kind of pluralist vision for funding those kinds of things. My view on that is pragmatic, right? Like, I'm sort of not left or right on that issue in the sense that if billionaires can help solve the problem, that's great. If government-funded DARPs can help solve the problem, that's also great. But From a kind of zoomed out architecture point of view, like it shows that, you know,

• 55:17Speaker 01

  you want a structure that isn't just this kind of like perfect circle where everyone's voting on everything. You want to have a structure that has layers. You want to have a structure that creates room for different sub-bubbles to emerge. You want to have a structure that opens up room for people who have some hypothesis that they believe in, but nobody else believes in, to still have time to flesh out that hypothesis and see if they can bring it to life, at least far along enough that people do start to believe in them. And, I mean, these are also issues that I've thought a lot about because they're very relevant to public goods funding within the cryptocurrency community, to the

• 55:57Speaker 01

  design of DAOs, decentralized autonomous organizations within the crypto space. When I was visiting Ukraine a couple of weeks ago, I was talking to one of the co-founders of Ukraine DAO, which is a decentralized autonomous organization on Ethereum that is supporting various Ukraine-related projects. And Ukraine DAO got a lot of flak for not having a governance token, right? Like, a lot of people in the crypto space deeply believe in this idea that in order to really be decentralized, you have to have a token, where if you have the token, you can vote and you can contribute to making decisions. And anyone should be able to go and buy that token. And the Ukraine DAO co-founder, this

• 56:39Speaker 01

  like steadfastly refused this. And the Ukraine DAO model of decentralization is more holocracy inspired. There is this concept of pods, where each individual pod operates very autonomously, and it gets some resources from the core. And the core is governed in a somewhat more distributed way, but still not using a token, because they really want the governance to be aligned, right? And they definitely don't want, you know, 30% of the governance token to just get bought up by Russian governments or whatever, right? And these are... Definitely ideas that blockchain communities are starting to play with more and more.

• 57:21Speaker 01

  Like, I think the conception of how governance should happen, like, it's already changing to some extent, right? Like, I think if you looked at crypto governance thought five years ago versus crypto governance thought today. Crypto governance thought today is already more nuanced. It's already learned from some of its early successes and some of its early failures. It, you know, understands the limits of financialization more. And, you know, it understands the need for some of these more complicated designs. And these designs where you do sometimes intentionally have levels of indirection a bit more. So it is starting to come a little bit more into balance in that sense.

• 58:07Ezra Klein?

  I want to stand the core idea here a little bit more before I move on to another use case. Because I think what you just said actually helps me sharpen a point I've been trying to circle here, which is, I think there are two very different definitions of trust and how one would achieve it on the table. And one way of thinking about trust, which I think is more dominant in crypto, even if only kind of implicitly in how it works, is how do I make it so I don't need to trust you? People used to use the term "trustless" about crypto a lot. And I have some disagreements with whether or not you can truly get that. But but there's a version of this where the question is, how do I not need to trust you in order to still work with you? And what then if we can achieve that, can we unlock in terms of human cooperation?

• 58:48Ezra Klein?

  And then there's this other question of trust, which I think people in kind of my job think about a lot, which is, how do I make it possible to trust you enough so I don't have to constrain you? Right? How do I make it possible such that I have enough trust in you that I give you the discretion to do what you need to do? Because that is typically a more flexible way for human beings and institutions to work. And I think that's the thing I'm trying to untangle here. Whether particularly when we think about political institutions and governance institutions separately from, say, a financial transaction, whether we're dealing with a question where the problem is, how do we make it so I don't need to trust you because your actions are well constrained?

• 59:30Ezra Klein?

  Versus, how do I make it possible to trust you so that your actions don't need to be so constrained? And a lot of blockchain-based solutions seem to me to really circle that first question when it often seems to me that the problem of our society is more built around the second one.

• 59:45Speaker 01

  I would say it's not a binary so much as it is a spectrum, right? Like, I think in both cases, what you're trying to do is that you're trying to give people more freedom and more empowerment to do things that you would want them to do, but at the same time, take away their ability to completely break things, right? And I think that's true in politics as well, right? Like, the reason why a lot of people are against monarchy or dictatorship as a form of government, for example, is that you don't want to trust one person to be a good guy. And I think, you know, in that context, that's something that's very easy to relate to.

• 1:00:25Speaker 01

  And the goal of not being a dictatorship or monarchy there is, like, it is trustlessness in the sense of wanting to avoid trusting one particular person, but at the same time, you're explicitly saying, well, you might be okay trusting a couple of hundred people where those hundred people themselves have some kind of, you know, regular democratic elections as a feedback mechanism. So it's still sort of moving up on the ladder from like a huge amount of trust to a somewhat lower level of trust. And then something like Uniswap is going all the way and saying, well, you know, can we try to get all the way to a zero level of trust, right? And

• 1:01:07Speaker 01

  Well, Uniswap in particular is one of the few things in crypto that I think actually does get you to the zero level, though with the caveat that practically using Uniswap, like generally, involves trading between different assets and those assets themselves almost always have some kind of trust component in them, even if it's a very distributed form of trust, right? And I think in crypto as well, there are real benefits that you do get from the trust reductions. And often, the worst disasters that happen in crypto happen in the parts of the systems that haven't been a kind of trust minimized in any crypto-y way, right?

• 1:01:51Speaker 01

  So, back in 2014, the Mt. Gox collapse, where Mt. Gox got hacked for about $400 million, at a time when $400 million was a huge chunk of the entire crypto market cap, that was a centralized exchange that got hacked. And then, you know, two years later, Bitfinex got hacked. And then, I think that series of hacks was part of the motivation for the excitement to move toward this decentralized exchange model, where instead of there being so much attention and money on the centralized exchanges, we would have these stablecoins. And stablecoins would still be trusted to hold dollars, but in order to do the actual trading, you would just do it on-chain, and so you don't need centralized intermediaries to hold things on the cryptocurrency side.

• 1:02:36Speaker 01

  And I think that actually has improved things. The times when it hasn't improved things, probably the biggest one was the Terra-Luna collapse of a few months ago, right? Which was interesting, because I think there is a combination of two causes. One of them was that the mechanism behind Terra and Luna was just fundamentally bad economics. Lots of smart people were saying, like, hey, this is fundamentally bad, and it's basically a Ponzi, and it's going to break at some point. But then the other part of it was, there was this reserve in Bitcoin. This was in Bitcoin, right? And Bitcoin, the blockchain, does not have this whole smart contract stuff, and so... If you're holding Bitcoin the asset,

• 1:03:18Speaker 01

  that means there is like an actual person or institution that's holding Bitcoin the asset. And nobody knows what the Luna and Terra team were doing with Bitcoin, the asset. And they made a lot of promises. And like they really tried hard at making these kind of very centralized efforts to manipulate the market and prop up their coin, but it ended up eventually failing, right? So, so I think that story is instructive because it shows like to some extent, decentralization by itself doesn't solve every problem because if the algorithm is bad, then even a fully open and transparent implementation of a bad algorithm is gonna break. But at the same time, it does still show the difference between

• 1:03:59Speaker 01

  the decentralized and trustless part of the ecosystem where lots of people were able to see ahead of time what was going on. And lots of people were able to warn about what could happen versus the totally opaque and black boxy part where you ended up not even having that, right? And like a lot of the disasters in the group of space that have happened recently, the disaster report has happened on this kind of more completely untransparent and more centralized side of the problem, right? So I don't think of trustless versus trustful as being a binary. I think of it as, you know, you can make a system that relies on you trusting one person.

• 1:04:43Speaker 01

  You can make a system that relies on you trusting forty people with checks and balances, or you can try really hard to get as close to zero trust required as possible. The right points to choose in different applications is always going to be very different. And this is one of those things where, you know, again, I think the the space has gotten better at thinking and having nuanced perspectives than it was something like five years ago where basically there were just the two extremes of like, you know, hey, yeah, let's say zero trust everything versus like, hey, yeah, you know, we're humans and any attempts to try to reduce the need for trust is sociopathic in some way. And, you know, now we're seeing the entire spectrum, which I think is great.

• 1:05:58Ezra Klein?

  Tell me about the idea for Soulbound Tokens, which is definitely my favorite name of any idea in the crypto space.

• 1:06:04Speaker 01

  So the concept of a Soulbound Token, the name comes from World of Warcraft, where there was a type of item that you can get called a Soulbound Item. And the idea is that, once you pick the item up, you cannot trade it. So you cannot give it to someone else. You cannot sell it to someone else. There is no way that you can transfer it to someone else. And often, you would get Soulbound Items as a reward for completing difficult quests and/or killing very difficult monsters in dungeons, right? So it's like the item actually becomes a badge that says something more interesting about you than like, "I have a lot of spare time," or "I have a lot of money." And what I think is interesting about that concept

• 1:06:49Speaker 01

  is that there's a lot of applications, even just in the real world, that could really benefit from that, right? So all of these different, what we call attestations, that people make about each other, right? I attest that, you know, you attended this conference. You attest that I participated on the Ezra Klein Show. Or Bob attests that Alice is a unique human, and Bob attests that he trusts her. And all of these different claims about things that we've done, relationships that we have, these are all things that can be valuable inputs into a lot of different things, right?

• 1:07:33Speaker 01

  So governance is just one example. You generally want the people who govern a project to be people who care about that project, to be people who have experience in understanding that project and its ecosystem, right? So, you know, going back to the case of Ukraine Dow, right? Like, you want the governance of Ukraine Dow to be run by Ukrainians. You don't want it to be run by whoever controls the crypto stash of the Kremlin. That's probably the most extreme possible example that I could come up with. But, you know, there's a lot of... Kind of milder versions of that kind of misalignment pretty much everywhere in projects that people are making today in the crypto space.

• 1:08:16Speaker 01

  And so if instead of having governance tokens be tradable tokens, what if we have governance that's somehow gated by soulbound tokens, by tokens that are not transferable and that actually do reflect the facts that you actually participated in some community, that you actually went to some number of events, that you actually completed some training course, that you have some particular piece of knowledge, that you personally did something good for the ecosystem at some point in the past, right? Like attributes that are about yourself. Another example might even be like, who do you accept as a member into a particular kind of club? Who are you going to be more willing to give loans to?

• 1:09:00Speaker 01

  If you want to issue a new currency, then who do you give more units of that currency to? So in a lot of these cases, like, you know, you want to have more visibility into what kinds of relationships someone has, what kinds of things that someone has done, and you want to really avoid this effect where all of this gets collapsed down into a single dimension, and we call that dimension money, right? Because if we make all of these tokens perfectly tradable, then what happens is that they basically just become isomorphic to money. And so the only thing that anything you build can be aware of is basically what level of resources each individual person has access to. So the question is, well, can we try to move away from that, right?

• 1:09:44Speaker 01

  And if we can have this internet where people can make proofs and claims about these more detailed things, about what communities they have more participation in, according to whom, you know, are they more trusted, various kinds of proofs that they are unique person, proofs that they've participated in some things, proofs that they've learned some things, then there's just all kinds of applications that could really make use of that kind of information. And these are things that happens in an informal way all the time already, right? And the challenge is basically, well, how do we formalize it and get the benefits of

• 1:10:26Speaker 01

  being able to formalize it and, you know, being able to have these applications run across bigger social distances? But Without that formalism, like basically collapsing into something that completely destroys the information.

• 1:10:42Ezra Klein?

  I think it's a really interesting problem for the internet. I think that we've all had the experience to some degree of beginning to upload markers of our identity online, and then feeling that either we got flattened by that, flattened by, you know, our Facebook profile, and, you know, I no longer like the bands I liked when I was 19, or weirdly pushed more into an identity that we only held weekly, right? That, you know, I'm a, I don't know, I'm somebody who doesn't like dogs, and then all I get is, I do like dogs, but somebody who doesn't like dogs, and all I get is a lot of hilarious anti-dog content, and I got into a fight about dogs on Twitter, and now soon I'm like the anti-dog guy. And

• 1:11:25Ezra Klein?

  I wonder how you think about what can be seen here and what is lost in trying to see it. Because I, on the one hand, I think the idea of using much more of, you know, the fuzzy kind of recommendations of each other, I guess LinkedIn tries to do this a little bit with recommendations, but that's for more professional context, makes a lot of sense. And then on the other, you know, I could imagine a lot of people who, if an idea like this took off, people for whom the internet was a respite from a life that socially didn't go the way they had wanted it to, where they didn't to some degree have the networks and connections, and not succeeded in the place they were, and the internet

• 1:12:08Ezra Klein?

  was a place to sort of abandon that, and now it's become a replicant of that, but in a thinner way, where sort of all you can see is it is what you don't have, rather than the sort of story behind it. I guess this is just circling again the question of what it is like to see as a blockchain, but I'm curious if you have reflections on that as somebody who I think is probably very attuned to the way, you know, it's sort of a mixture of the ability to socialize and the ability to not be your offline social identity that has made digital worlds great for so many.

• 1:12:39Speaker 01

  Yeah, so this, I think, can all be viewed as a subset of gridhards law, right? When a measure becomes a target, then it ceases to be a good measure. So when you come up with a way of measuring something and then you make other things actually depend on that way of measuring things, that gives people the incentive to try to optimize for it. And once you start optimizing, then all of the divergences between the measure and the thing that the measurer's trying to get at, like really suddenly come into the foreground. And sometimes the harder you twist the knob on optimizing, the greater the divergence between the two becomes, right?

• 1:13:19Speaker 01

  And this is a pattern that occurs everywhere in like any kind of formalized measurement of just human traits that other people care about, right? Like, you know, university admissions is one of those politically charged topics lately. And, you know, there are standardized tests, and standardized tests end up having various misalignments, and people end up teaching to the test, which is just a completely tragic waste of a lot of people's time, because you don't, if, when you learn how to do the test, you don't actually learn anything that's useful to solving, you know, real-life problems that humans care about. But then if you scrap that and you rely on sort of more fuzzy and informal measurements,

• 1:14:02Speaker 01

  then there's a lot of other, different ways in which that's unfair, and a lot of different and other ways in which people can optimize for that. And, you know, people get the idea that they should join like some, you know, BS, nonprofit, humanitarian thing, not because they actually care about the cause, but just because it makes them look good on a resume. And these kinds of patterns are patterns that occur pretty much in any context when you try to measure things about people. And I think it's a good reason to be careful about that sort of thing. But at the same time, it's also not a good reason to shy away from it entirely, right?

• 1:14:45Speaker 01

  Because, like, there's no such thing as not measuring people, because if you don't measure people formally, then, you know, the fundamental need to find people who are more appropriate to some particular task, or position, or relationship is still there, and you're gonna end up... I try to fit people to it informally and the informal ways of fitting people have, you know, all kinds of problems of their own and have had those problems for 10,000 years. So the general patterns that make sense in like at least trying to reduce

• 1:15:25Speaker 01

  the level of misalignment, right? I mean, one is don't rely on one measurement, rely on many measurements, right? And ideally rely on many measurements that have been developed independently. This is a pattern that I think generally comes up again and again. And, you know, if you look at how just modern AI works versus how people tried to create AI using rules based systems in the 1980s, right? Like rules based systems in 1980s were more like, oh, you know, check if it has ears and check if it has whiskers and if it does, then it's a cat. And that sort of stuff is just like incredibly hard to code and it just breaks horribly. But the more modern AIs are trying to do is

• 1:16:07Speaker 01

  they're basically extracting millions of different traits, millions of different sub measurements and millions of sub measurements on top of other sub measurements and trying to create some kind of composite score out of all of those. And, you know, even there, you sometimes get what are called adversarial examples, things that, you know, you can create an image that's a human looks like a dog, but to a particular classifier looks like a cat with, you know, a 99.999% score and everyone's like WTF mate, what happened here? But like, that's still harder than if you just rely on one thing. And so I suspect that soulbound token based stuff or like any kind of stuff is going to depend a lot

• 1:16:51Speaker 01

  on combining multiple measurements. The other one is to be willing to adapt, right? Like if you have a relatively rapid adaptation loop, then

• 1:17:02Speaker 01

  You might be able to kind of catch the misalignments pretty early and hopefully you're able to fix at least some of them. And that's interesting because like the ability to rapidly adapt does a kind of conflict with the, I guess, crypto-y idea of you set the rules once forever and then you forget about them. And so, you know, there's definitely a challenge of like, well, you know, what is the thing that we want to preserve in order to make sure that the thing is still secure versus what is the thing that actually does have to adapt in order to make sure that it's still secure? And, you know, it's a difficult balance and there's definitely even philosophical questions in there.

• 1:17:46Speaker 01

  But the hope is that in this kind of more decentralized context where it's decentralized in the sense of you just naturally have lots of different metrics and you get these attestations coming in from huge numbers of people. And so it is going to be more difficult for people to kind of optimize for any one thing in a pathological way. And, you know, you're going to get something that's more secure. So that's the hope, right?

• 1:18:11Ezra Klein?

  Well, let me ask about one of the possible uses of it that you all lay out in the paper, which is very close to my heart, which is around the question of deep fakes and what we'll be able to trust on the internet in the coming years or even right now. How do you imagine having this sort of more bound layer of online identity could help with that?

• 1:18:31Speaker 01

  Yeah, so I think in the age of deep fakes, it definitely becomes more and more important to start, like basically splattering any images that we make with as many kinds of this different cryptographic claims as possible, right? So one very simple thing that you can do is when you take a picture, you can immediately ship it off to a service that then bundles up a million hashes and, you know, makes a tree of those hashes and sticks that in the blockchain. And what you have then is you have this cryptographic proof with a, you know, maybe 15 second window that says this picture was made before this time, right?

• 1:19:12Speaker 01

  And then the hope is that at least sometimes you'd be able to take that picture and then combine it with evidence inside of that picture that tells you when the picture was supposed to be taken and you have this cryptographic evidence of when it was produced before. And that might reduce the chance that it could have been manipulated, at least with some techniques, right? So that's one thing that you can do, but in a lot of cases, that by itself might not be enough. And so then you might wanna also add different other cryptographic tags. Like one simple cryptographic tag you would add is, you know, who took the picture? You might add other tags of just other people who saw that picture being taken and happens to have been there at the same time, right?

• 1:19:55Speaker 01

  You might want to have some kind of, you know, trusted hardware solution where, you know, the camera as a device would add some kind of attestation that it took the picture. And then you also have attestations about the particular person that has uploaded that picture. And you could even look up and see, you know, hopefully, whether or not there have been serious claims of whether they have published a deepfake in the past, right? So the more stuff that you cryptographically sign, the more stuff that you commit to, the more stuff that there just is information floating about, the more of these pieces of information you can combine together

• 1:20:36Speaker 01

  to improve your confidence about whether or not some particular picture is real. I guess one very concrete analogy might be if there is some, you know, human rights abuse happening in, you know, say Tehran right now. And there's both three Iranian vocals and a BBC journalist and an Indian journalist who all, you know, take pictures and they have this software that immediately, you know, blockchainifies it. And they cryptographically sign it, then you have attestations about them. Then all of that information taken together should be able to get you an automated, pretty good impression that that event actually did happen

• 1:21:20Speaker 01

  and it wasn't just invented by really good deepfake software. So that's the hope.

• 1:21:27Ezra Klein?

  I think there's an interesting question here, which I think is a nice kind of wrap on a lot of what we've been talking about, which is we've been talking in a lot of the different ways about how to trust each other and how to work with situations where I think we want to trust each other, but for one reason or another, it's hard to do. We don't know each other. We've had bad experiences in the past. It's complicated. And I often think of the deepfakes problem and a lot of misinformation generally as being about cases where people actively don't want to trust each other. You want to believe something that makes somebody else look bad. I don't think we verify things as well as what you're talking about, sort of adding that amount of cryptographic proof to it. But, you know, I work for the New York Times. If people wanted to like look up information or try to see if things were true

• 1:22:11Ezra Klein?

  or find established purveyors of information, they can. And the really hard thing that nobody really has a good answer to is often they don't. And that's true for people who agree with me, not just people who don't agree with me. And so there's a kind of interesting thing here. And then the other piece of it is that particularly around leakable information, there's so much cover for pseudonymity, right? You know, if you were whoever leaked the Mitt Romney 47% video, you had a good reason not to do so from an account that in any way could have helped people triangulate who you were at that fundraiser. And that's true for a lot of things. And as such, you can imagine a world where it's very normalized. You know, at least it's a plausible argument that, oh, like, you know,

• 1:22:55Ezra Klein?

  the reason this doesn't have a kind of soulbound token attached to it or social authentication attached to it is because it would be dangerous for the person doing it, which is all to say, how do you think about the situations where we'd want to be building a society where people have attached markers of trust to them? But as we've seen in a bunch of different sort of anti-establishment or motivated reasoning cases in our own world, people often, they're motivated by the desire not to trust and to validate their reasons for not trusting. And it's sort of those more cultural forces that pull us apart. How amenable is that to sort of better technology?

• 1:23:34Speaker 01

  Yeah. So. Sometimes it is and you know, sometimes it's totally not. So the thing that technology sometimes can do is it can just at least lay the problem really bare in front of everyone's faces. So you know, you see this, you know, some piece of information comes out. And then some people believe one thing about it and some people believe a different thing about it. And then 12 hours later, you get like some really strong actual confirmation or actual denial. And you know, one of those two camps looks ridiculous to some extent, right? And this is one of those cues that I think, you know,

• 1:24:14Speaker 01

  probably both of us and a lot of people use to try to check like, well, you know, which political tribe should be like actually trust in the first place? So sometimes, you know, you can do things like that. But then, you know, sometimes there's definitely an extent to which what there are lots of people where what they're looking for is not the truth, but a sufficiently compelling a kind of justification that lets them feel good about coordinating around the narrative that they wanted to coordinate around anyway. And I mean, in some cases, like the wrong kind of technology being developed first without other technologies to counterbalance it could even make the problem worse, right?

• 1:24:58Speaker 01

  There's definitely, you know, certain kinds of evidence and statistics and, you know, logical arguments where in the wrong hands, like that just gives people more ammunition with which to create arguments that are very convincing to them. And to some extent, you just can't solve that problem. And sometimes what you're really hoping for is, or the best that you can hope for is that if there is something that's incorrect, but a lot of people really are motivated to believe it, like can you get the percentage of people that still stick with it down from 35% to 25%?

• 1:25:37Speaker 01

  Sometimes technology can also, especially some of the cryptographic technology that we work on, it can give some interesting options that can change the social dynamics. So like, for example, You could give people the ability to say something anonymously, but at the same time prove something about who they are, right? So you could, you know, using something called ring signatures. This has been actually around for years, but I think it's only going to be really practical starting around now just because of infrastructure. You can make a week and you can sign it with a signature that proves that you are a senator of the United States without revealing which senator.

• 1:26:20Speaker 01

  And that lets you reveal truthful information without kind of compromising your own social situation. Another use case of this is rather than revealing truthful information, you might just want to like speak an opinion, right? Like if there is a political opinion that you deeply believe in, but let's say there's like some rich and powerful, you know, individual, even an authoritarian regime that has a habit of canceling people who loudly speak that opinion, then you as a celebrity or a politician or even just like a mass of regular individuals, you know, you can go and prove some things about who you are

• 1:27:03Speaker 01

  to give weight to your words without revealing everything about who you are. So those kinds of tools are going to give some interesting new options in the social landscape of, you know, truth telling and signaling of a lot of different things. And there's definitely many kinds of equilibrium that that sort of stuff will break. But, you know, the kinds of equilibrium where it's just that people just don't care about the truth and they believe what they want to believe. I mean, yeah, there's a limit to the extents to which we can solve any of that. I mean, I do think that, you know, as time progresses, ultimately the beliefs do have to get somewhat less crazy

• 1:27:46Speaker 01

  because there's just more and more evidence about reality and people just have to try harder to come up with something that people are going to be willing to accept. But, you know, at the same time, you could argue that the crazies are getting more sophisticated at the same rate as the anti-crazies. And I definitely don't claim that either blockchains or cryptography or even social technology is able to solve every problem. I think, you know, every era is going to have its own challenges and its own issues and we'll see.

• 1:28:17Ezra Klein?

  Let me end on this question before we do book recommendations, which is... In 10 years, what is the problem that you most think is unsolved now that blockchain and crypto will solve, that we will feel in our lives? Like, what is the solution you're most excited about, but that most people don't realize is coming yet?

• 1:28:36Speaker 01

  Hmm. So I think right now, the biggest impact that blockchains have had has been in the area of international money transfers. I think there is a lot of good reasons for that, like basically just because the existing infrastructure for finance is just crap in a way that the existing infrastructure for email, for example, is not crap, right? Like, centralized email is centralized, but at least you can send an email from Kyrgyzstan to Guatemala within one second. In the future, I expect, you know, in addition to, like, obviously, international payments continuing to get more effective, I expect at least one or a couple of some of these

• 1:29:20Speaker 01

  other sectors that people are trying to build out with blockchain. So some form of what we call DAOs, I do expect that that concept is going to get stretched to the limit. Like, I do expect that a lot of things that people call DAOs are just going to be, like, basically, groups of five people where they agree that you need three of those people to agree on transferring money to someone. And I think that's just like a form of organization that'll just be like very easy to deploy and pragmatic and useful in at least some applications. So that could start getting used in a lot of contexts. Some of this blockchain for identity stuff, using identity in a very broad sense to cover,

• 1:30:03Speaker 01

  you know, ENS and like people having Ethereum accounts instead of Facebook accounts, the concept of attestations, the concept of soulbound tokens, and the big, long list of stuff surrounding that. So far, ENS has been by far the most successful non-financial blockchain application, and I expect that to continue to be true for a while, though I expect a kind of other, particularly identity-related, non-financial blockchain applications to do really well.

• 1:30:31Speaker 01

  In the blockchain gaming space, I expect that like probably medium amount of success. But I expect that the blockchain games that succeed are going to be games that are actually fun, even not taking into account the blockchain aspect. And with things like the Dark Forest game, like we have seen some examples, but then we've seen things like Axol Infinity that just have not done a very good job of like that aspect of being fun without the finance. And that's part of why they haven't really been able to recover well from the challenges that they've had recently.

• 1:31:06Speaker 01

  In the space of like governance mechanisms, I expect there to be a lot of intersections of blockchains and zero knowledge proofs. And I also expect there to be a lot of like use of blockchain-inspired, zero-knowledge-proof stuff make its way into various kinds of government civic applications. I mean, I expect that some of that will be blockchain-based and some of that will be based on things like zero-knowledge proofs. Some of that will be even based on ideas that kind of got nurtured in the blockchain space and then sort of works their way up in various ways.

• 1:31:46Speaker 01

  So I think in each of those spaces, I probably have some medium level of confidence that it'll turn into something very big. But I think if any one of them does really well, then the world could look very different in a lot of ways 10 years from now.

• 1:32:06Ezra Klein?

  And then always our final question. What are three books that have influenced you that you'd recommend to the audience?

• 1:32:11Speaker 01

  Oh, this is a good question. I guess technically not a book, though parts of it, I think, have been retroactively made into books in various forms. The whole kind of early rationalist canon, the kind of Slate's Codeex, so Wiesrude Yudkowsky sort of blog posts. There were a lot of very fascinating insights in some of those earlier things on questions like what are political tribes? You know, what kind of irrationality do we have and should be watch out for? The fact that just being aware of irrationality doesn't by itself magically make you rational. And just all of these kind of deep issues that...

• 1:32:55Speaker 01

  Have to do with like basically how to think better like a lot of my thinking of on various kinds of issues ended up being really influenced by that. In a somewhat totally different space, history and like very deep and intense history, the rise and fall of the Third Reich, I read this one last year and it was just a fascinating inside view tome into like basically how this country really descended into one of the scariest totalitarian regimes that humanity has seen in quite a long time and on a large scale.

• 1:33:37Speaker 01

  And I mean, they're seeing what happened on the inside, why were people in Germany leading up to that situation, even welcoming the changes that were happening at the beginning and you know, how those different actors felt, who resisted, who ended up not resisting. I mean, I guess it was good that I read that less than a year before the whole war in Ukraine started. I think it even, it helps me be less surprised by some of the things that I've unfortunately been disappointed by that happens in that whole situation. So that gets to two. Fiction, I don't know, I mean, Harry Potter and the Methods of Rationality, I mean, I don't know, it's kind of cheating because it's also in rationalist lands, but it's just like

• 1:34:22Speaker 01

  such a lovely tome that's just.

• 1:34:25Ezra Klein?

  That's also an Eliezer Yudkowsky joint.

• 1:34:28Speaker 01

  It is, yeah, yeah. It's a great mix of just like great fun and total ridiculousness, but at the same time teaching you some like fascinating ideas around things like logic and game theory. And then other ones might be, there's definitely been.

• 1:34:47Speaker 01

  Specific textbooks that have really taught me a huge amount. Um, there was one textbook on game theory that, um, I forget the name of it now. I think it's like something like algorithmic game theory. But that taught me, uh, just some of the finer points of game theory. Some of the finer points of like particularly how things break down when you have some, you start to bring in assumptions of where people can coordinate with each other. Right? Like people talk a lot about how economics is unrealistic because it assumes things like perfect information and perfect rationality.

• 1:35:28Speaker 01

  But like one of my contrarian takes is that actually those aren't even the big problems. The big problem is that economics often assumes what's called the individual choice assumption, that people don't talk to each other and don't coordinate with each other outside of whatever the game is that you're making. And how things break down once that assumption breaks. I mean, the topic of what's called cooperative game theory. Uh, so unfortunately can't remember the, the exact name, but I feel like all of the textbooks that I've seen on cooperative game theory are very enlightening on that particular topic.

• 1:36:04Ezra Klein?

  I think that's a great, great set. Um, Vitalik Buterin, thank you very much.

• 1:36:08Speaker 01

  Thank you too, Azren.

• 1:36:13Ezra Klein?

  That's the show. If you want to get at some other crypto episodes we've done, we had one with Dan Olson, which was a very, very critical take on crypto and one with the investor Katie Hawn, which was a more optimistic take on, on NFTs. And so if you want to hear some other perspectives, I recommend going back to those. If you want to support the show, you can leave us a rating in whatever podcast app you're using or send the show to a friend or if you didn't like it to an enemy. The Ezra Klein Show is produced by Annie Galvin and Roji Karma, fact checking by Michelle Harris and Mary Marge Locker and Kate Sinclair, original music by Isaac Jones, mixing by Sonia Herrero, Isaac Jones and Carol Sabaro, audience strategy by Shannon Busta. Special thanks to Kristen Lynn, Christina Samulewski, Will Wilkinson, Alex Tabarrok, Ben Wile and Nathan Schneider.