Why is the SEC Concerned about Privacy now? - Peter Van Valkenburgh. Ep625

Transcript

Coin Center's main day-to-day work, our number one mission priority is always to defend the developers of open blockchain networks from inappropriate regulations or unjust prosecutions. I think it's becoming more and more self-evident that a transparent layer one is not a neutral layer one. Not in the long run. Ultimately, transparency will destroy neutrality. And this is not me saying that like we should be building things that enable money laundering. It's me saying no, we should build global neutral communications networks for things like underlying settlement. This was true of Swift in the 80s. It was neutral. If we're going to build a better system, we should be building more like Swift used

to be and less like what Swift's becoming where it's just a fully mediated underlying settlement ledger for even [music] for messages for for settling transactions, let alone transactions themselves. We're here at Dev Connect and today we're speaking with Peter live and in color. Uh this is it's nice to see you here. >> I should have worn a more uh expressive outfit if we're >> Yeah, I think we're all a little bit drab, but I think it's okay. Kind of like the the background makes up for it. So, it's it's a little for everyone who's listening to this, it's a little trippy. Um but it trippy and kind of uh >> creepy AI's flaw. There's probably AI.

There's some tech in there that's definitely >> Chachi PT's idea of >> words. >> How how has Buenos Iris treated you so far? >> Oh, it's been great. I think I think one of the and it's become sort of widely discussed. One of the interesting things about DevCon this year or Dev Connect or whatever the big Ethereum conference this year is that privacy is actually being discussed >> more fully. And you know, Coin Center, my organization, has been trying to focus people's attention on the need for financial privacy and the need to protect the builders of financial privacy for over 11 years now. And sometimes there's periods in that history where you're like, "Yay, like

Zcash or things like that." And sometimes there's periods where it's like, "Okay, everything's going to be done with uh stable coins onchain and we'll have full transparency into all transactions." And that's what will tell law enforcement why they don't have to worry about the technology because it's a panopticon. And now we're kind of swinging back to like, oh well actually maybe that would be bad not just for the criminals. We want it to be bad for the criminals but actually bad for all legitimate uses of the technology as well cuz who wants to use a global financial system where every time you buy a can of coke a billboard pops up with your name. It's like Peter just bought a can of coke everyone. It's bad.

Yeah. This episode is brought to you by Nosis, building the open internet one block at a time. Nosis was founded in 2015 and it's grown from one of Ethereum's earliest projects into a powerful ecosystem for open userowned finance. Nosis is also the team behind products that have become core to my business and that of so many others like Safe and Cow Swap. At the center is Nosis Chain. It's a low fee layer 1 with zero downtime in 7 years and secured by over 300,000 validators. It's the foundation for real world financial applications like Nosis Pay and circles.

All this is governed by Nosis DAO, a communityrun organization where anyone with a GNO token can vote on updates, fund new projects, and even run a validator from home. So, if you're building in web 3 or you're just curious about what financial freedom can look like, start exploring at Nosis.io. How much of that do you think the interest in privacy is driven by market movements or or is it like is the demand for privacy that's driving markets or is it the markets that's driving privacy like you know >> Yeah, it's it's probably it's probably the markets driving privacy because you know as much as you as much as I would like to as an advocate as a lawyer as a person who stands up for like what we

should like normatively what we should have you can be like build privacy and People are like, "Why would I build something no one's going to use?" Like the thing that infuriates me more than anything else at all of these conferences, I meet all these great people and usually they're like, "Oh, how can we stay in touch after we leave this conference?" Um, and they're like, "Do you have Telegram?" And I say to them, "Why would I use Telegram?" Like those strange halfbaked, not encrypted for group chats and vaguely encrypted for individual chats platform where they rolled their own hash functions. Like that's a disaster. Like we've had Signal for decades now. What's wrong with you?

Why aren't you using Signal? Yeah. Um, and it's because people don't take their privacy seriously. Ordinary consumers don't take their privacy seriously, especially people who are like new to crypto or new to blockchains that we all want to onboard now. Like, they have no idea. And rightfully so. They're time constrained. They're not sophisticated. But the thing that bugs me is that sophisticated people, people who've been coming to DevCon for multiple years, people in the Ethereum community are like, "Let's let's just give all our information to to the Russian secret service, which has compromised this app, and and the the French intelligence services as well, right?" I mean, like that that's just odd. Why wouldn't we use Signal anyway?

>> I mean, I think I think there's a good reason why people didn't use Signal up until recently. And maybe >> it's bad for creating a boiler room because the group chat function is complicated. Well, the group chat the group chat function is complicated, but also up until just a couple years ago, you had to give your phone number. >> That's true. >> And the UI around that was is not great. I think that gave Telegram the opportunity to really like create a huge >> You found a more forgivable argument. >> Yeah. But also UI is not great. >> The group chat UI was bad. So like people who wanted to like pass insider information about what thing they're going to pump next like we're like, "Let's be on Telegram because we can have a good group chat."

>> Yeah. >> I don't know. So yeah, I think it's market driven because not because consumers are demanding it, but because the new potential consumers of like typical consumers, the new potential consumers of blockchains are institutions. And institutions take their seriously because if you're JP Morgan or you're a proprietary trading firm or whoever, like you're not interested in just hemorrhaging all of your uh information to the public public blockchain because that's not how the traditional financial system ever worked. You have trade secrets. you have practices that you don't want revealed and that's not because they're nefarious that's just because how business is always run. So we have a more serious sort of institutional potential user for this tech and I think they demand

privacy. So I think that's real. I think the other thing that's driving the privacy conversation is also and this is maybe more my biased view um because I come from Washington DC and I see the evolution of these things um from a legal standpoint the I think it's becoming more and more self-evident that a transparent layer one is not a neutral layer one not in the long run ultimately Transparency will destroy neutrality because if you're validating on this chain and you see all this activity, at what point is someone going to run up to you and say

you're kind of culpable for that activity? Like maybe you didn't directly promote it, but you facilitated it, you aided it, you abetted it. Like if you see a multi-IG or if you see a a a a moneyaundering transaction, a multi-IG moneyaundering um and you you you validate that as part of a larger block um what's your long-term argument for you're not being complicit, especially if you could have used chain analysis as a validator to identify these things and you didn't. That's called willful blindness in the United States and it's not a defense to culpability for criminal law. actual blindness is. And so this is the difference. People say like, well, how will they ever allow Zcash to be a thing or something with

base layer privacy or Aztec built on top of Ethereum? My answer is like that's the only thing they will allow because actual blindness is a defense to crime. Like if you had literally no knowledge of some criminal act that you facilitated, you didn't actually facilitate a criminal act. Not with intent to facilitate. and you can't be held culpable and you shouldn't be held culpable. >> And so that's the only way to actually have credibly neutral systems is to have systems that are actually blind to the underlying transactions in the block. I said in my talk the other day, if you really want dumb pipes, you need the pipes to be actually blind to the that flows through them because otherwise it's not a dumb pipe.

>> Yeah. I I feel like there's there's a lot to unpack here. So kind of like I think we should distinguish between kind of absolute privacy and relative privacy, right? So kind of like people people um kind of regular consumers don't want absolute uh don't want um don't want don't need absolute privacy. So kind of like or they they don't think they need this. So kind of like they have no issue with Telegram reading their stuff or their bank knowing kind of like where their money goes. They don't want everyone to know it. So they kind of need relative privacy. Um and I think kind of like that's that's also something that that kind of plays into this a lot. I really feel your discussion uh discussion point on protocol level blindness. Um because we

we recently did a thing on Nosis chain. Yeah. >> Um I don't know whether you follow the balancer hack. Um >> a little bit >> a little bit. So basically uh there was a hacker. He was clearly malicious. Um uh he uh stole 128 million across many chains. uh almost 10 million of which were on Nosis chain and there were um there were some attempts to kind of uh freeze some of it uh wi with uh the freezable stable coins and then he concentrated on mainet and kind of uh sold positions for ETH. So kind of it gave us a little bit of time to kind of think what do we do here because clearly this is someone who's stolen >> actual people's money kind of like it

was stable stable pools it wasn't kind of like some sort of DGEN uh sort of thing and it was also pretty blue chip protocol so it was balancer v2 stable pools so it's it was yeah so what we ended up doing is um first we kind of asked bridge governors to kind of reduce liquidity so that um the hacker wouldn't be able to bridge off nosis chain easily. >> Um and then um we try to get um validators on board for a soft fork meaning kind of uh and I mean just just to kind of clarify a hard fork is where kind of like you change the state uh kind of like some some time in the past and a soft fork is basically where everyone agrees to not process certain

transactions. So kind of like you don't include them in your block and but you also don't attest to them. Yeah. And I think not including things in your block. This is something that kind of has been done before and is currently done on Ethereum, but kind of saying I'm not going to attest to this if it's in there. This is this is a line that so far hasn't been crossed until now. so on until now our nosis chain. I'm not proud of this, but kind of like it it made us and kind of um it made us appreciate that kind like while we still have this power, >> it's probably the arguably the right thing to do, >> but we still want to get to a place where we do not have this power anymore.

>> Uh >> yeah, I mean so one thing I think about a lot lately is the swift interbank settlement. So that's the Society for Worldwide Interbank Telecommunications. >> Oh, is that what that means? >> Something like that. >> Okay. Yeah, I just learned that. >> It was started in um Belgium in the 1970s. And the central bankers at the time didn't even know what the heck all this computer nonsense was. And so they didn't >> they were they were like given an option to have a seat at the table amongst the the the banks that were developing this underbang settlemental. And they were like, "Ah, we don't know like computers, that's not what banking's about." Yeah.

>> And Swift end up ends up becoming the most like single most important cross crossber settlement messaging network. They are not doing the settlement themselves. They're a messaging network. Sounds familiar to a lot of stuff in our space >> in the world. They settle 150 trillion um in transactions or they don't settle, they message them. Yeah. >> So that banks can settle them every year. 150 trillion. >> And up until 2012, Swift was credibly neutral. And if you think about the early days of Swift as a messaging network, it's all plain text transactions between their bank their their member banks. Their member banks, it's a permission set. Um, but they are credibly neutral in a way because they don't have the computational ability at that point in history or really the the

wherewithal to accumulate all the messages on their proprietary network and really know who's sending what to who and unwind it all. Like, it's just not something that's feasible. >> By the time we get to the '9s, the 2000s, of course, it's totally feasible. They know everything that happens on their network and yet they continue to claim to be just a messaging provider neutral. >> And that finally breaks down when the US government and the EU parliament and others say like you need to stop relaying messages to Iranian banks who are you know facilitating the Iranian government's nuclear program. We have sanctioned them. Until then, Swift said, "We are not an obligated entity for purposes of US or EU sanctions because we are not actually involved in the

underlying economics of the transaction. We are just a messaging provider." >> No. >> And there's a good reason for that. You want a messaging provider to be globally neutral. Otherwise, we won't have a like throughput in the global economy. Every transaction will end up mediated and a whole council will have to decide whether to let it through and the whole thing collapses. And you fast forward to the Russian invasion of Ukraine and Swift removes the um the the the Russian banks from Swift and now and that was in 2022. So we're not we're not long beyond that. To me, there's this slowmoving crisis where Swift is going to become a fully mediated, fully obligated global settlement tool for messaging for settlements. and it just won't have the

throughput and and and viability that a truly neutral tool would have. It's going to collapse. This is one thing that makes me actually very bullish about open blockchain networks is because we're going to have to find alternative tools that don't become political footballs in geopolitics if we're going to have actual free and open global markets. But I I think that to to Felica's example about Nosis, there's there's a there's a distinction between coordination within a group of participants in a blockchain network and coercion. And in the case of Swift, Swift is being coerced.

>> Yeah. By >> but any individual validator can be coerced, right? And I mean, so kind of the thing is Nosis is 300,000 validators, right? I mean not not as many individual entities but kind of like we still had to make the case to them that this is the right thing to do and kind of like they they kind of they had to press the button themselves to kind of upgrade their note or not upgrade their note and kind of like in the end um most of them decided to upgrade their note. So kind of the soft fk went went through um but kind of like if you have a threeletter agency kind of uh calling you up and saying ohas you're you're you're processing some really dodgy transactions here. This is this is the dynamically updated list of things you shouldn't be processing and please

please make sure you subscribe to this and and and uh and uh it it informs your your validator at any given time. That that's a totally different situation. Right. >> Right. But I mean I I think I think that's true. But I think that you we need to make sure that our systems are resistant to coercion so that Yeah. Right. And maybe privacy fixes this, but like are not resistant to coercion >> and and the original project of Satoshi and the project of Vitalic and the larger Ethereum community is not necessarily to um ensure that validators are incapable of all malfeasants but that there are so many validators that even if some are like regularly

censoring or blocking either because they've been coerced or because they are themselves not interested in a free and open blockchain there will be some minority participants on the network like located in far-flung areas of the world who will continue to put things in in the blockchain, right? But I think >> it's no longer true kind as long as kind of like you have this attestation thing, right? Because then you kind of need a twoth3 majority kind of like to to attest to this. So kind of like not only do you need a block builder who knows this, but you also need kind of like to lock out and have twothirds of the um attesters kind of be on the on the non uh three data agency list. I'm just learning about this here at devcon connect but this proposal fossil which

would um be a ethereum improvement proposal where in block builders would have to sort of be mandatory it'd be required to include the transactions that are validated by like a committee of some 16 or something like that and there'd be minority members in that committee who might be the people likely to put in transactions that aren't wanted by by by the majority of the staking power on the Like this is interesting cuz it's like oh let's let's rebuild censorship resistance in a world where we're starting to lose it because we we don't have those far-flung validators on chain. This is actually very bad from a regulatory standpoint if you don't get this with privacy because now an American staking individual or company is going if they want to v build valid

blocks is going to be forced to include transactions in the block that might val that might violate say OFAC sanctions law. And I don't know what their defense is going to be. If it's if it's a plain text transaction that clearly actually can be read by law enforcement or anyone with chain analysis or blockchain analysis tools that shows North Korea using the Ethereum blockchain, you had to put it in your block. They come to you as a validator say, "Why did you put this sanctioned transaction in a block?" And you say, "Well, the Ethereum protocol demands that I do it." And then they say, "Well, you don't have to be on the Ethereum protocol, right? You violated sanction law." But I think kind of like I think there's there's an interesting distinction here because kind of like the the kind of like put putting uh not putting certain

transactions in blocks. This is currently the status quo. Right. I mean this is and it's been for years but kind of not attesting to them. >> This is new. >> This is new. Okay. >> Right. So it's that hasn't been done before. Clearly it can be done >> but uh yeah and yeah and I think I mean I think it'll happen and you put kind of like it just drives home kind of like how we need to um improve systems such that you you have no way of knowing what you're validating >> be actually blind not not willfully blind and not and certainly not just like eagerly looking and being like yeah sure I'll get that one and then wait for the knock on the door. Um, the case that Coin Center has recently worked on that

really brought this into focus for me just before DevConnect here is actually the Pereira Bueno case. The Pereira Bueno brothers are the the guys who found the exploit in the MEV boost software >> and then sandwich attacked sandwich attackers. >> Mhm. >> And >> it was hilarious. >> It was hilarious. What's kind of not hilarious and rather terrifying is that you could have imagined that the victims victims because they also just lost in cutthroat competition. They didn't actually like get I don't think they're criminal victims, but the victims of that exploit, they could have brought a civil case maybe and said like, "You misuse the MEV boost software in a way that that hurt us. We want to reclaim our $20 million." That didn't happen.

That would have been bad enough. What actually happened is prosecutors in the Southern District of New York, criminal federal criminal prosecutors brought a wire fraud case against the brothers saying your behavior on the Ethereum blockchain as a validator is fraudulent, is wire fraud because you were not, and this is a quote from the jury instructions that they actually asked the judge to provide the last moment in the case a couple weeks ago, because you were not an honest validator. But this is absurd because kind of like the entire point of me boost is that kind of like you extract from from un unwitting users of kind of whatever DeFi protocol.

>> Yes. >> Like let let the robbers rob in peace. >> Yes. Well, and not even let the robbers rob in peace. It's just like you wouldn't go to a marketplace that is known for having established rules of commercial conduct that are effectively like like if you were the highest bidder, you win. It doesn't matter if if you made people think you were going to bid lower earlier or bid higher earlier. Like if this is a market where the expectations of all the participants is that like people will profit maximize up to the bare rules of the protocol, then there is no fraud or misrepresentation.

Everyone knows that in the world of minor extractable value, people are going to compete to extract as much maximum extractable value as much as they can get out of the protocol. And that might be not desirable, but we intend to fix that using technology, not layering a whole system of duties and honor that would then be enforced by criminal prosecutors on top of all validators because if that's what happened, we'd be in a very bad situation. And there are specific claims in in that prosecution because they did equivocate according to the Ethereum consensus rules. they presented two, you know, um, blocks that can't be reconciled with each other simultaneously, for which they got slashed. And so, in a in a highly technical, like going all the way back

to Satoshi's white paper, what is an honest anonest chain, they did something that is not honest validation. They equivocated. But this is not dishonest commercial conduct in the wire fraud sense. Yeah. where they made a representation to someone that someone relied on in a fiduciary or or contractual relationship and then they were able to profit from that. And so this notion that like honest validation and dishonest validation being something that you could actually criminally prosecute just further indicates that we're going to have a huge problem with transparent um ledgers wherein the government's going to see validator behavior and figure out all kinds of ways to get jurisdiction over it, including when they're being

self-interested in a way that maybe some people in the community don't want them to be self-interested, but everyone knows that they're self-interested. and it's it's just it's not going to work. So, we filed an emergency amikas brief in this case because that jury instruction was something the defense flagged and we had been in touch with the defense and they were like, we can't have the jury decide that they committed wire fraud because they were a dishonest validator. That that doesn't make sense. And so we wrote an emergency brief um basically explaining the history of that term honest validation within these networks and how the Ethereum blockchain and the community at large has an expected penalty.

So a slashing fine or slashing fee for exactly that behavior and that's carefully calibrated to be actually quite modest just to prevent constant equivocation on chain. And to think that you'd have now a slashing penalty for that behavior and possibly like multiple years in jail from a felony criminal charge completely unseats the the calibrated uh norms and established rules within the protocol and it just would make Ethereum not work. >> Yeah. Yeah. >> This is a lot but uh >> Yeah. No, I >> fascinating me lately and >> yeah, I mean it's Yeah, I Yeah, it kind of it breaks kind of like this closed system game theory of kind of of of the protocol itself. Y >> um

>> what what kind of what struck me while you were talking um is I mean obviously kind of like we have this entire situation way worse on a way bigger silver platter with L2s, right? kind of like and kind of like the multisig L2s where kind of like you you you can you can upgrade from hopefully a multisig and not a single address. I mean this is there's also some of the EOA whatever whatever but kind of like um and uh where where you can where you typically have a single sequencer a very small set of sequences. So basically effect in effect the entire chain is built by a single entity. Um h how how do do you

think we'll see that play out in L2 first or do you think they'll go for for the base layer? >> I think no that's going to be a very appetizing target. >> Yeah >> because you you'll have an identifi identifiable um criminal defendant. They have very clear control over a lot of relevant aspects of the transaction flow if we're talking about some illicit transaction in the bundle that they that they sequenced and they're probably well- moneyed defendants too. So you could extract large penalties from them because they're all like these companies that raise tons of money to build L2s, right? So it's a very dangerous state of affairs. And I mean, I've said for a long time that we need truly

decentralized sequencing on L2s, and there's efforts to build that. I do feel like some of those efforts got maybe stymied or put on the back burner when you know like American politics changed and everyone was like oh the SEC is maybe not going to care as much anymore which is not necessarily true by the way but also still not the right way to build because you should be building decentralization for decentralization's sake not just because you're afraid of a wells notice from the SEC and then the other thing I would say is like I'm much more um for obvious like this was not would not be a surprise to anyone right now I'm much more of a fan of like the the let's find ways to to have fully private blind sequencers like a like an Aztec type model so that you know even

if we have a certain amount of centralization with the sequencer they are not they are not like looking into what they're actually sequencing in a way that would make them responsible for mediating disputes whether that's filtering out you know transactions that might be related to moneyaundering or whether it's just like favoring their own transactions on chain or any number of other things for which they could be held culpable. And this is not me saying that like we should be building things that enable money. It's me saying no, we should build global neutral communications networks for things like underlying settlement because that's the only way we're going to have a global economy that actually functions at the scale and efficiency that we want the global economy to function. This was

true of Swift in the 80s. It was neutral. They were just moving money for Iran all the time. And that was a necessary cost of having a liquid and free and open global economy. And you say that like this is ultimately going to be good for human freedom because totalitarian regimes will suffer in that world even if they get occasionally the benefit of the underlying transaction rails because they're free and open. You lock them down, you actually empower totalitarians. And so like if we're going to build a better system, we should be building more like Swift used to be and less like what Swift's becoming where it's just a fully mediated underlying settlement ledger for even for messages for for settling transactions, let alone transactions themselves. You know, >> I'd love to talk about KYC, AML, and compliance

>> as it relates to privacy. I have a hunch that a lot of the reasons why we have so many >> Let me let me put this another way. Yeah. the the the compliance cost to companies to funds not even just in crypto just broadly individuals >> is absolutely massive. Yes. >> I don't know that there are any real estimations of what this costs to the economy but I I'm sure it's in the tens or perhaps hundreds of billions of dollars. >> Yeah. I think I think a reliable estimate was some firms in the US spending upwards of 20 billion um just individually and that the total cost was something like 300.

>> Yeah. >> Um I mean >> there's some number by a guy named Robert Paul uh who's a big a very good critic of anti-moneyaundering. We've also published a report recently so I hope I'm not misstating any of these um I'm sometimes bad with decimal points but I think we're actually on target here. We published a report called Tear Down This Walt Garden. Uh it's a Coin Center report. It was authored by myself and Ian Meyers. And the first half of it or maybe the first quarter just lays out how expensive AMLKYC is. And that would actually be fine if it had bang for buck. But there's also good there's actually more reliable estimates as to like how much illicit finance is actually deterred and how many how much illicit money is actually frozen or

blocked in the economy. These are estimates from they're not estimates from weird crypto libertarian orgs. It's the United Nations and the Financial Action Task Force. Their estimate of the amount of elicit funds actually um seized within the global economy thanks to AMLKYC is sub 1%. In fact, it's about 0.2% of all elicit funds. >> And this is coming from the FATF, the very organization that is implementing or pushing to implement >> the financial action task force. Yes, they have a stock take that actually basically puts that number up and the UN as well. So these systems, now a counter-argument to that is okay, yeah, the seizure and blockage rate is actually extraordinarily low compared to how much elicit money is out there, but

the deterrence is what makes it work. When a criminal goes to use one of these tools, they go, "Ah, no. I have to put in my mother's maiden name. They're going to catch me now." You know, and I'm making this counter-argument sound too weak. Maybe there probably is some deterrence, especially for like low sophistication criminals. There's no deterrence for high sophistication criminals. In fact, the very fact that financial institutions have been collecting all of these documents and keeping them in insecure uh facilities means they constantly get hacked and you can buy on the black market any number of like identity credentials that you can use to open bogus accounts if you're a sophisticated criminal. So sophisticated criminals I don't think are effectively deterred at all by the

system. If anything, they might be empowered in a way because they know they can find readily available fake documents that have been hacked from an institution that was obligated under law to collect all of them in the first place from their users. So that system is not working. And that's the point that we make in the first quarter of this paper. The rest of this paper is okay, we're not just here to be mean to financial crime authorities. Um, we all need to recognize that there's a deficiency here and we all do want to stop terrorism and terrorist usage of financial networks as best as we can without destroying the global economy of course. So, what can we do better? This approach clearly isn't working.

>> Not working. It's not working for law enforcement. It's not working for ordinary persons whose privacy is invalidated. Oh, I would also add whenever people factor the costs of KYC, they they factor the costs to financial institutions. You rarely ever see anyone actually measure the the costs of say the the global identity theft losses from people whose privacy has been compromised or just the psychic losses of knowing that your privacy is fully compromised which would be hard to monetize or or account for but are are real like if you believe in in a right to privacy. Setting all that aside, we can do better. And a lot of the technology that's been pioneered in this space, zero knowledge proofs, multi-party compute, verifiable credentials, is actually like exactly what you need to sort of brew up in

order to build a system that works better. And we're seeing some of these as like early proofs of concept prototypes. Like in the in the Aztec token sale that's happening right now, I think you can use their own >> can use their own passport >> DK passport which means you create a proof that you your passport you you are in possession of a passport at least in this moment that says you're an American and this is your name and then they can cross check that against the sanctions list and things like that. These are still not enough though and this is the paper the paper we're writing says look isn't this cool? These are already here. But what we really need is global composable user sovereign digital identity wherein I know I have a passport what I can create zero knowledge proofs of that. I

have a bank account. They will allow me to attest to my balance. Um I have um a Gmail account that's had this message history for the last 11 years and I can actually make an attestation to that effect. And I can compose these myself into a risk score that might be recognized by some particular authority as being credibly not a criminal or credibly of a low risk. And there's some important features there. It should be multiffactor. It should be attribute- based. Maybe they don't learn my identity when they when when I provide them this risk score. They just learn that I had the ability to construct this risk score out of credentials that I myself possessed. And it it should be far better I think than than just you

know oh you want to open account send us a scanned version of your passport which is a garbage version of actually putting deterrence in. So I think you can build deterrence into these systems but you should be building them with as much as possible um attribute proofs from zero knowledge proofs rather than identity proofs like I am this is everything about me >> right >> and they need to be modular so that we can like assemble the right series of credentials that would be difficult for a bad actor to fake but actually quite easy for a normal person who's just a good person to accumulate but I mean I hate sorry I have to be conspiratorial But my feeling is that there will be and probably is a lot of resistance to these

systems >> because of the amount of control and information that financial institutions indeed governments are able to collect. Yeah. >> From the blanket uh capture of everyone's financial data, companies financial data, documents, etc. Uh and then of course all the service providers and businesses that sort of live off of this uh off of this model. I guess my my what what I wanted to get to here. What what do you think within United States Congress and and policy makers is the appetite for uh moving towards a world where we have more sort of ZK at testationbased KYC and AML uh compliance rules.

>> Um because you know coming from from Europe I feel like the consensus view in Europe is that we should just collect everyone's information and know what everyone's up to all the time. I I don't know if that's the same in the US and in fact at the end of the day it's the US policy that will dictate what the FATF does in the future. >> That's usually what happens. Usually usually the US says to FAT if this should be the global policy uh implement it bullies the world into doing it and then actually often doesn't implement it for its own citizens because there are constitutional right concerns. So it's it's a bad it's a bad I call it policy laundry actually like we'd love to impose these on ordinary Americans but we can't. So we'll go to we'll go to Brussels and we'll tell people to do it

globally and then we'll come back to the US and be like see it's now the global standard and then Americans will be like actually you can't you can't do that with our anyway that's a digression. So to answer your question directly um like what's the appetite and you framed it very well in like maybe it's not conspiratorial but maybe the reason people really like AML KYC is not actually deterring criminals it's mass surveillance and there may be some truth to that. I think that the less conspiratorial reason why there isn't the appetite we would hope is simply that if you're a regulated financial institution today, what incentive do you have to stick your neck out and try alternative modes of of

anti-moneyaundering compliance? like are you really going to be the bank that talks to your probably like mid-level bank examiner and says this year we decided to stop doing KYC and instead we're relying on these risk scores which are actually amalgams of much more credible information about the likelihood of criminal behavior from our users but we don't know their names anymore. No one is going to do that. And no no bank compliance department is going to support their chief innovation officer going to Treasury and being like, "Hey, let's do this." Right? Like there's there's no money in it for them.

There's no reason for them to do it. It's just downside risk. And so it's a collective action problem at the end of the day. We might be able to get to a better world where financial institutions and technologists could build these tools that actually make a bigger dent in illicit finance while still protecting our privacy. But who's going to be the first to do it? Government's not going to design that system. They don't have the technical competency and it's really maybe not their place to even develop the technical competency. The private sector as far as any individual bank is not going to lead by sticking their neck out because there's no incentive for that.

This is why we think there's this opportunity in our space in the more novel fintech and blockchain space for a group to get together and say like actually we can do a lot better. share these ideas with regulators, regulators who are genu genuinely interested in deterring illicit finance, whose day job is watching terrorist financing and just knowing that there needs to be a way to stop it. We're more concerned with that than they are with mass surveillance, like knowing whether Peter is buying the wrong books today. Domestic terrorism might kind of conflate some of these things together and maybe that gets questionable, but in a free society, we wouldn't we wouldn't want to collect surveillance data on what books people are reading. So we take these ideas and we socialize them with those regulators

and ideally and this is Coin Center's goal for work over the next few years actually we're we're doing it under a internal project we call the John Hancock project because it's the most famous signature on the Declaration of Independence. The goal is to see if we can actually get regulatory buyin once we present like truly robust and viable alternatives to KYC that are privacy preserving. And by regulator buyin, I don't mean like a strong like, yes, if you build this, it will be 100% compliant. I simply mean a credible statement from the administration, from Treasury, or maybe from the SEC if they're starting to talk about tokenizing equities or things like that, that we are open to doing pilot programs for regulated financial institutions to do alternative customer onboarding where

a risk score is maybe preferable to a full full identity documentation because I think if you got that signal from regulators, that starts to break the collective action problem wherein who wants to be the first to stick their head up with a new a new system for deterring elicit finance when everyone seems to be fine with the bad system that's working today and there's no no incentive or or money for you to build a better system. Right. >> Well, I'm I kind of I I really commend you for this. I think it's it it's a terrific initiative. Are you worried about the fact that most people just don't care? Because it seems like it seems like in principle kind of like this this will be kind of like a political fight and kind of if if

you look at >> how much people willingly divulge of themselves, how much they use extractive services despite the fact that there are feature equivalent private privacy preserving ones. >> Yeah. >> Um do do you think it's kind of like a culture battle that kind of we have to win first? So, if we were just talking about changing consumer behavior, yeah, this would be fairly hopeless. But I'm talking about something >> Okay, let's wrap it up. [laughter] >> I'm I'm talking about something far more narrow, which all down today, even if you have a consumer who cares deeply about their privacy. When they go to open an account at a regulated financial

institution, they have to submit something very specific. They have to it in most cases it will be a photo of your passport and your face right now and that will be stored in a in a probably in an unencrypted database at some point in the in the flow of data and that's just bad, right? >> It's really bad. >> And so just that is all a regulatory function. That's not consumer behavior being, you know, not not conscientious enough about your own online safety. That's literally just a rule that is an interpretation of the bank secrecy act in the US that says institutions need to collect this information for their customer due diligence. And that rule could be changed if you had people who cared not in the consumer base but just

in the regulatory and compliance areas. And I think I think there are more people that care now just from my meetings with people in Treasury and the SEC over the last 5 years. there is a growing awareness of the how how bad the current AML efficacy is and you know it's kind of like finding fun common ground with people. I hate it because it's the system that destroys my privacy and I have certain civil libertarian um ideological underpinnings. You hate it not necessarily for exactly the same reason. you you might believe in in certain like rights to privacy and freedom, but you hate it because it's just not even stopping criminals anymore. It's in fact stopping um disenfranchised people, poorer people,

people who are easily dissuaded from using financial services when you put up barriers to them. And it's not dissuading, you know, hardened criminals who can easily amass a whole dossier of of fake passports or passports that are not their own. Right. Yeah. I I mean I I think the there there's sort of two parts to this. I mean there there is the consumer part. I I I feel like it's probably fintech apps that would implement this sort of riskbased KYC uh using attestations. You know, I could see like a Revolute doing this right for consumers.

>> Um >> but it gets so much more complicated for companies. I mean we run a small fund and the cost of compliance and the amount of compliance work we have to do is just like I I don't even want to imagine like a you know a fund twice or three times our size or like a company like Nosis the amount of compliance work goes into just day-to-day operations and the cost that has to the company and if you extrapolate that to the entire economy it's a huge number like you you flowed some numbers earlier and and I think it gets much harder then because it's not just you know your passport and what your salary is why it's you know it's the entire uh organizational structure, you know, all of your ubriven today, which is wild and part of the

reason the system doesn't work well. Like if you if you had more composable proofs that are difficult to um that are difficult to forge and you had some ability probably from a third party like a watchdog organ dog organization or even the government to recalibrate when one particular aspect of a proof say like your um say say a particular financial institution gets hacked and they were one of the ones making attestations about like past bank transaction history for civil prevention just proof of human or maybe that they'd KYC these people. Then you need to be able to on the fly devalue that as a bundle of a risk score. Maybe it doesn't go to zero within the risk score, but suddenly that which was providing some

robustness to this risk score gets unweighted to down to like half of what it was before because we think it's compromised. We don't think all the identity data they have is bad and all the attit in in the in the in the ultimate cumulative calculation of risk. >> And like that's what you want. You want a system that's got the ability to observe threats and rapidly recalibrate. It's sometimes referred to as an UDA loop. Uh it's like observe. I don't know. I don't know what it stands for. It's it's in like military. It's like fighter pilots. They're supposed to like survey their realm and then make decisions based on new information and rapidly iterate. That's what you want in a in an environment with revolving threats, right? But that's not what we have right now. But right now we have

like people who actually manually look at passports still and like ah when you go to the airport all the security theory is like yeah it looks like you. It's like you have facial tracking now like why are you still doing this? I don't know. Anyway, how much [clears throat] I think this is kind of like part of part of the equation kind of like making sure that actual bad guys are more easily catchable, right? But I think kind of like you can also see it from the other from the other side. um of you actually owning your identity because kind of like if you if you look at um the way how we use identity services um mostly we're renting right kind of like if if I kind of like log in with Google kind of

like what do I have I mean they have they have all my stuff they have my data they know who I am and kind of like I I am beholden to them that they don't they don't offboard me >> and same for all other kind of it's there's there there's they're kind of like in web two there's nothing >> that's truly yours. It's kind of like you are the renter for everything and it's a huge power imbalance kind of like and kind of puts every single person at at an enormous citiz disadvantage. And if you could just say this is your self-s sovereign identity and you can kind of you you can you can use this to validate your identity in whichever way you choose against whatever. You don't

need you don't need someone else to kind of stand up for you and say, "Yeah, this is really Peter." Yeah. Um I think I think this this it's super powerful. >> I've started saying that like when you are you online, you have a different name. Like going back to like the feudal medieval period like where your name was like Peter Blacksmith because you were the blacksmith, right? When you're online you're Peter Instagram or your your Nirage X.com like you're not yourself. You are you are a vassel of a feudal lord. That's a problem. >> Yeah. >> This has long been discussed in crypto circles as being part of the problem.

And a big a big the the the goal is the end goal is to be self- sovereign over your identity over your persona online. And this does factor into things like KYC because the simplest way to do better KYC would be to have one big corporation that collects all of this information and generates a risk score. that is far more computationally efficient than trying to do it using disparate credentials uh multi-party compute um and maybe some like anchoring on open blockchains which require proof of stake or proof of work or any number of other inefficient non-paralizable processes.

It would be much better if it was just one company. But then of course we all understand why it shouldn't be one company because this is not just one company that collects all the passports. Now it collects your entire trail of breadcrumbs that you leave across your entire life that you want to amass into a viable proof. It should instead be the user using commonly available transparent open source tools and amazing new privacy preserving technologies can amass these credentials and voluntarily themselves generate probably variously differently calibrated risk scores that might be calibrated by third parties but third parties don't hold the underlying data.

the underlying data. All the att testations that the person wants to make with their credentials are things that they control in a wallet that they have. So very much like the alternative to Google Sign on being like sign on with MetaMask or sign on with some sort of like key pair that you control. But that would need to be that there's a lot of building left to do there obviously. I mean, you say that that entity shouldn't exist, but kind of like if you look at what Google plausibly knows about you, Peter, I mean, that that entity exists, right? And multiple of them actually exist. I mean, they shouldn't exist. I think this is uh >> Yeah, >> but right now they're not identified in law as the as the entity that should be doing anti-moneyaundering uh verifications. They'd probably be

really good at it. >> They would be excellent at it. >> Yeah. If you started if you passed a law that said banks don't need to do KYC anymore. They just need to ask Google if somebody's evil or not. >> Yeah. >> And Google can't be evil. So that would that would probably be much more efficient as far as like stopping a lot of money laundering and crime, but it would also be a disaster from a civil liberty standpoint. So ideally, we need to identify in law something else that could be equivalently superior to the existing AML regime, but doesn't hand all the power to one big tech corporation or big bank. Yeah. And the stuff in our space is the stuff that can build those systems. Though it will be a very difficult thing because we we're competing against much more efficient technology companies that don't have the

costs of civil liberties which are spreading data out, decentralization, strong encryption, things like that. >> We've we've talked a lot about privacy and uh identity. >> Yeah, this kind of just became the privacy identity. I hope you don't mind discussion. >> This is perfect. Are those the things that you think kind of we need to um we need to pay the most attention to right now or are there are there lesser um publicized things where where you think kind of like this is something that is currently overlooked that we absolutely need to get right. I'm glad you asked the question because because people who don't know Coin Center might now be mistaking me for like like the digital identity center and I'm like no that's that's an initiative that we're just beginning because we've realized there's

appetite from regulators to try alternative AML methods that could better preserve people's privacy. Coin's main day-to-day work, our number one mission priority is always to defend the developers of open blockchain networks from inappropriate regulations or unjust prosecutions. And so this goes back to our earlier discussion about say the Pereira Bueno case where you know you've got people who are validating on chain obeying the rules of the protocol and yet the Southern District of New York wants to accuse them of being involved in wire fraud or committing wire fraud. That's not a good legal precedent to set and it's it's it's going to discourage people from participating in these free and open networks as validators. The other big things, the things that we

were definitely discussing more say like last year are the tornado cash case. Um first the sanctions of tornado cash which a big win actually in US policy was um Coin Center and others challenged um the legality of those sanctions. And ultimately just this past spring um those court challenges we won and that means that the the administration delisted tornado cash. Um so Americans are now free to use tornado cash again. This is actually like a very real >> Wait, only Americans or >> Well, let me yeah, let me be clear. The sanctions only ever applied to Americans. That sanctions law says these are foreign sanctioned entities.

Americans are are not allowed to transact with them. foreign persons could have maybe secondary liability even though they're not Americans for violations of sanctions law, but that's less clear. Um, but Americans were just banned from using tornado cash after the sanctions came out. That was unequivocal. But just a small tangent here speaking, does that mean that people who had funded funds in tornado cash can now >> take them out and, you know, carefully >> send them to Binance and trade them back for Bitcoin or whatever? >> Absolutely. >> Really under the law. under the law, there's the the the sanctions are now null and it is as if they never existed.

Got it. Um so you you are not liable for sanctions violations if you now reclaim your property that was left in the tornado cache pool. you should be very careful doing it. And I'll just say this because many of the user interfaces to tornado cache the underlying pools are actually compromised now from what I understand because they kind of fell into disuse and people started setting up like frontends that actually had back doors that would steal your steal your funds instead of allowing you to take them. So be careful if you are trying to reclaim property that was left in tornado cash from a long time ago. But legally you're you're safe.

>> But will you ever have a bank account again? Because kind of like even if kind of like you're you're it's no longer on the sanctions list, will it still be flagged? Will you be deplatformed from whatever >> Yeah. >> traditional you have concern? Yeah. Not I mean not that I had ever used to cash but I know people who have and are >> I I have used to cash. I have used it liberally and for good reason. So kind of it's uh >> this is a tough one because what you all what you what you need to do to address that is to socialize the fact that there there is no legal risk to a financial institution in now accepting Ethereum that came from a tornado cashache address especially if you know the customer who's bringing it to you like you know that they're just an American

who is like getting paid their salary on on Ethereum and wanted privacy over their salary like one of our co- plaintiffs in our lawsuit. But how can we socialize that amongst all the compliance departments of all the banks? So if there's if there's a stigma that's still attached, that's something else. My point is that the law has changed. It is now technically legal to reclaim those funds and to use them. Maybe you should run them through rail gun first before the taking somewhere else to take off the tornado cash uh funk on them. They might get new funk from other from other privacy tools, which goes back to our discussion about base layer neutrality. like base layer neutrality.

Like no one's going to ultimately end up using these privacy tools if they're still just islands of privacy within a larger transparent chain because you'll always be like, you know, the scarlet letter of having gone through this little privacy tool even if you were doing it for the right reasons. Yeah. >> And we can also watch in real time as bad people use these tools and that creates headline risk and it creates, you know, risk to the whole technology. >> So the sanctions are invalid. The other big thing that happened this year, and it's not it's bad news instead of good news, is the developers of Tornado Cash, Roman Storm, uh, specifically, um, was found guilty of unlicensed money transmission in his court in his court case in the Southern District of New York. Um, that is a very bad outcome.

It's maybe not the worst outcome for Roman Storm because he was also charged with money laundering and sanctions evasion, but the jury couldn't reach a verdict there. So, he could get reprosecuted for that or may not, but he was found guilty of unlicensed money transmission. And this Coin Center has done tons of work on this is just ridiculous because Tornado Cash never had control over user funds and certainly not the developers of Tornado Cache and also not the torn token holders. No one did. they were in immutable pools on chain and Fininsen the regulator for who needs to register as a as a money services business in the US who is doing money transmission at the federal level has said you're not doing money transmission unless you have total independent control over customer

funds. So, you might ask, why is he then guilty in New York of unlicensed money transmission if the regulator said he wasn't ever doing money transmission? And it's because the Southern District of New York has this sort of wild theory of criminal liability for unlicensed money transmission. That transmission is something broader than what the regulator requires a license for. And they did it and therefore we can charge them with this. And what I think so what's really happening is it's hard to charge someone with money laundering and sanctions evasion because you have to prove criminal intent, some specific intent to actually hide some specific bad funds. And like yes, North Korea used to cash, but that was after they published the software. And so what did they have intent to do? They had intent

to publish a privacy tool and then somebody used it for moneyaundering after. That's not intent to launder money. That's intent to publish a tool that people will use to do lots of things. Um, it's much easier to charge someone with unlicensed money transmission because there's no intent requirement. All you have to ask the jury is were they doing money transmission? And we will tell you jury that we legally define money transmission broadly and did they knowingly transport criminal funds while doing money transmission? And I guess you could argue like they watched it as it happened. Their software did it. So it's a it's a much lower bar.

said there's no there's no need to show the jury that they intended for this to happen or wanted this to happen or directly facilitated it happening. You just say, "Did you have a license or not?" So, but saying that they watched them watched their software do it and did nothing about it is like insane because >> it's like watch like it's like having a gun store and then someone buying the gun and then watching that guy mow down a crowd of people as he walks out your store. >> Right. And and that's why we we would not usually subject people to a strict liability regime for things that are not ongoing conduct. We subject money transmitters to a strict liability regime because they're always in control of the conduct on their platform. And so maybe the reasonable liability rule is if you're always in control, you could

always stop it. So if anything bad happens, you're automatically responsible. But that only makes sense in the context of a money transmitter. If you're a software developer, having strict liability for all the things that people do with your software for failure to license it is just gonna make it impossible to publish software because it's everyone knows most people will do good things with the software, some people do bad things. If you're responsible for all the bad things people do with your software, you won't publish software anymore. So, it's as I said, Roman could have had a worse outcome as far as like jail time if they'd been convicted of money laundering and sanctions evasion. But the fact that he was found guilty of unlicensed money transmission is a very bad precedent for this space because that charge could be brought up against

almost anyone in this space cuz their theory of what is money transmission that requires a license is any facilitation of the movement of tokens on chain which is something that like everybody who's involved with cryptocurrency has done in some way and most of them unless they're like coins Coinbase have not gotten a license to do it. So this is a very bad precedent. it needs to be challenged at the appeal level. So, we're hoping that Roman will appeal his guilty verdict. Um, Coin Center is also supporting a um civil litigant, Michael Llewellyn, in Fort Worth, Texas, who is a developer who wants to release a privacy preserving crowdfunding tool. He calls it Feros and is willing to sue the Department of Justice for declarative judgment before

he publishes that tool to get clarity from a judge that publishing that tool without licensing with Vincent will not be a felony will not be unlicensed money transmission and Coin Center is supporting his lawsuit. So we have sort of multiple paths to address this particular threat legal threat to the permission the the the freedom to build these permissionless systems. We also have a legislative effort. There's a bill moving through Congress called the Blockchain Regulatory Certainty Act, which would create a safe harbor for non-custodial developers. And I'm actually optimistic. It already passed the House. It might pass the Senate.

There's like a good chance we'll actually get a legislative solution to this problem as well. This is the day-to-day work of Coin Center. Um and now we are all increasingly interested in privacy preserving alternatives to KYC but um most of our work is in the trenches in the courts in the legislature trying to protect developers from inappropriate liability for other people using the software to do bad things. So >> if you had to make a prediction one thing that we don't yet see coming kind of like from the regulatory sphere for the next 12 months um what what would it be? You know, this is actually an interesting one. People might be expecting me to say like something scary and bad. The thing that's surprised me most recently that I think a lot of

people haven't groed yet is that the SEC, which until recently was a very aggressive regulator in this space interested in going after even people who were potentially good actors trying to do their best to comply. The SEC has done a 180 and actually in a very good way. um Hester Pur uh commissioner there and the new chairman Paul Atkins. They are they they are legitimate. They they are they are very um genuine when they say they want to tokenize everything which is actually like Paul Atkins gave a speech about this and they are also very genuine. They talk about financial privacy. Commissioner Hester Pur gave an amazing speech in defense of financial privacy where she where she said we

should uh celebrate rather than attack the availability of technologies that will recreate transactions that are like the transactions our forefathers understood when they wrote the fourth amendment that if I pay you in some coins or some some dollars there's no record of that transaction and that there's no way to surveil that transaction in a warrantless way. So between tokenizing everything and technologies for financial privacy are important and should actually be celebrated rather than attacked. The SEC sounds ready to do a lot of proactive things in the space to enable the kind of global and open and fair and liberal financial system that we want to build.

And >> this open and liberal financial system and SEC in the same sentence is not something that I had on my bingo card. But it's >> and and lots of people are are now aware of this. But I think I still talk to people who are like, should I go in and talk to Commissioner Pur about the the very potentially important tool that I'm building or is that scary? Um, should I find a way to say that like we can still preserve the role of say transfer agents and other intermediaries in the system because they're wary of like going full token tokenized and I'm like not necessarily like everything's on the table I think there and like there isn't necessarily right now amongst the commissioners a desire to reintermediate when that reintermediation

of securities transactions is inefficient and unnecessary. Um, so I think like just fully tokenized equities are on the table and I think private transactions are on the table. Now, there is an interesting thing here where obviously we're not gonna have a world where, you know, North Korea could steal a bunch of US equities in tokenized form and they're not going to find a way to cancel those shares. So we will still need tools for identity and compliance on these perhaps permissionless new securities um ecosystems. But that's again part of why I'm interested in finding like privacy preserving alternatives to the ineffective KYC regimes we have today so that we can

find ways to like uh figure out what the issuer security issuers liabilities should be or obligations should be visav the wrong person getting a hold of the US equities that they issued. But there's ways to address that in a tokenized environment that aren't like just rebuild a transfer agent who's a person who's just like got a list of names as shareholders and is like well we can't send that to him you know like something better than that. So >> so tell people where we can learn more about Coinenter and uh support your work. So I think it's very important work. >> Thanks Sebastian. Um so all of our work is publicly available at coincenter.org.

this report that I mentioned uh on digital identity is one of the more recent things we published and also I think more people should take a look at that amikas brief we filed in the parabeno case about me boost whether you're interested in minor extractable value or not it's it is it sort of helps tee up why there is this existential threat to neutrality of the base layer um in part because of the transparency of the transactions on it and the inevitability of regulators or prosecutors inferring legal duties onto validators and then arresting them for failure to obey those duties. So that amus is also up up on our site right now.

>> Yeah. >> Super cool. >> Thanks a lot. >> Thank you for taking the time to uh speak with us. >> It was a pleasure. Yeah.