Andrew Hinkes, partner at K&L Gates and adjunct professor at NYU, and Martin Köppelmann, cofounder of Gnosis, talk about the sanctions on Tornado Cash, how they will impact Ethereum on the base layer, and the likelihood of having two chains.
Show highlights:
- the legal meaning of the OFAC sanctions on Tornado Cash and why it is causing complications
- how legislation is built for the fiat world, and how in DeFi and crypto there’s no clarity
- how miners have changed their behavior prior to the Merge
- whether a proof of work chain or a proof of stake chain would be more censorship resistant
- whether laws apply to validators in a PoS chain
- the ways in which the government could provide more clarification
- the different entities involved in Proof of Stake
- the liabilities for each entity in PoS and what secondary liability is
- how the task of the bidder can be described as a mathematical optimization problem
- whether Lido is exposed to US regulations
- how long would it take for a large staker like Coinbase to stop being a validator
- how a user-activated soft fork works and the conditions in which a fork could happen
- whether social slashing is enforceable considering it is not in the protocol
- the likelihood of Ethereum becoming a permissioned system
- the role of Flashbots’ relay code and what it means for Ethereum’s censorship resistance
- how DeFi’s exposure to US-based institutions can affect the likelihood of the chain complying with sanctions
- whether DAI is just wrapped USDC
- how MakerDAO is trying to be more independent of USDC
- why US-based companies will choose a conservative path, considering the lack of guidance
- the importance of educating the regulators
- how cash and crypto are the only remaining private payment methods
Thank you to our sponsors!
Crypto.com: https://crypto.onelink.me/J9Lg/unconfirmedcardearnfeb2021
Ava Labs: https://avax.network
a16z: https://link.chtbl.com/HObAJQDL?sid=unchained
Guests
Andrew
Twitter: https://twitter.com/propelforward
Martin
Twitter: https://twitter.com/koeppelmann
Previous Coverage of the Tornado Cash Sanctions on Unchained:
Social Slashing and Censorship on ETH
Axios summary of the situation: https://www.axios.com/2022/08/23/how-transactions-might-be-censored-after-ethereum-changes-next-month
BitMex research on how the sanctions affect Ethereum: https://blog.bitmex.com/ofac-sanctions-ethereum-pos-some-technical-nuances/
Ethermine banning transactions: https://twitter.com/takenstheorem/status/1560690035955011585?s=20&t=59hWdhr8_O-hYA8uYnZJ8w
Nic Carter’s article: https://www.coindesk.com/layer2/2022/08/25/if-ethereum-starts-slashing-it-burns/
Eric Wall’s article on social slashing: https://ercwl.medium.com/the-case-for-social-slashing-59277ff4d9c7
CoinDesk article: https://www.coindesk.com/tech/2022/08/23/as-censorship-on-ethereum-begins-could-this-open-sourced-code-help-counter-it/?outputType=amp
Hetzner banning Ethereum users: https://twitter.com/koeppelmann/status/1563146729314467840?s=20&t=ubFhCQqrQyHKfflgMHEmJw
Tornado Cash
DAI and USDC
Flashbots
Proposer-Builder separation: https://www.alchemy.com/overviews/proposer-builder-separation
Episode Transcript
Laura Shin:
Hi, everyone. Welcome to Unchained, your no-hype resource for all things crypto. I’m your host, Laura Shin, author of The Cryptopians. I started covering crypto seven years ago, and as a senior editor at Forbes was the first mainstream media reporter to cover cryptocurrency full time. This is the August 30, 2022 episode of Unchained.
Need to keep up with the biggest news and market updates in crypto? Get the Unchained Daily newsletter in your inbox every weekday morning. Visit Unchainedpodcast.com to subscribe.
Whether you’re crypto curious or a C-Suite decision maker, you have to check out web3 with a16z, the chart-topping technology podcast about the future of the next Internet. Listen to web3 with a16z on Apple Podcasts, Spotify, or wherever you get your podcasts.
With the Crypto.com app you can buy, earn and spend crypto in one place. Download and get 25 dollars with the code LAURA. Link in the description.
Harness the full power of the Avalanche network with Core, your new web3 command center. Built by Alpha Labs, Core is more than just a wallet. It’s a non-custodial browser extension engineered for users to seamlessly and securely experience web3 like never before. Explore Avalanche dApps, NFTs, bridges, subnets and more today.
Today’s topic is how sanctions on Tornado Cash will affect Ethereum at the base layer. Here to discuss are Drew Hinkes, partner at K&L Gates and adjunct professor at NYU, and Martin Köppelmann, co-founder of Gnosis. Welcome, Drew and Martin.
Martin Köppelmann:
Welcome.
Drew Hinkes:
Thank you.
Laura Shin:
On August 8, the US Treasury Department Office of Foreign Assets Control sanctioned Tornado Cash, a series of autonomous smart contracts on Ethereum. This was an unprecedented move, since previously the only entities ever named to the sanctions list had been organizations or people who had the agency to also appeal to the government to be removed. Obviously, a set of smart contracts cannot defend themselves.
If you haven’t yet, be sure to listen to my interviews with Jerry Brito of Coin Center and Ari Redbord of TRM Labs for more background on why these sanctions are so consequential.
The lack of guidance from OFAC around this unprecedented move has caused quite a bit of consternation for numerous entities in the Ethereum ecosystem, and it’s raised questions for Ethereum itself at a pretty delicate time. The blockchain is about to adopt a proof of stake consensus algorithm, and the new validation process involves numerous different entities. They all now face the question of whether any action they take that helps validate blocks containing such transactions is a violation of sanctions, and the Ethereum community is also facing existential questions around how to maintain its censorship resistance despite the presence of a number of these entities which have exposure to US jurisdiction.
We are going to unpack all of this with one crypto legal mind, which is Drew, and one developer who is very familiar with the ins and outs of the new staking process on Ethereum, that’s Martin. Drew, why don’t we start with you? Can you tell us from the legal side, what is the significance of these sanctions and why has it been causing such complications?
Drew Hinkes:
So, OFAC maintains a list called the Specially Designated Nationals list, and this is a list of countries and their nationals who, for policy reasons, have been determined to be subject to sanctions. The sanctions prevent US persons from engaging in financial transactions with those listed persons, their accounts, and their property, and also renders those who have received property from those persons, it renders that property
blocked, and blocked property cannot be transacted. If you are in possession of blocked property, you are required to segregate it in an interest-bearing account and to make reports to the government ten days after you segregate and then annually thereafter.
These are laws that are designed around fiat, they’re designed around the system of banks, and they’re designed around a number of tacit assumptions. Obviously, there are limits as to the law. As our friends at Coin Center have noted, the law previously has been implemented against people, and a bunch of autonomous smart contracts operating on a blockchain wallet with no human controlling them probably doesn’t qualify as a legal person, and as your introductory notes suggested, there’s no one really to appeal on behalf of those smart contracts, and so we have some foundational questions as to whether the exercise of power under the statute is proper.
It creates some really significant issues as to those who want to comply. The smart contracts aren’t a person, so the value that flows through them maybe isn’t necessarily any one person’s property. If a person sends their assets through the smart contracts and they are sent out from one of the block’s smart contract addresses, you could view it as okay, this is block property because it came from the address, or you could view it as, I don’t know whose asset this is, but there is no person called Tornado Cash, so I don’t think that these assets are actually blocked person’s assets, and so maybe this isn’t blocked property.
We don’t necessarily know, if you are in receipt of blocked property that you actually think is blocked property, what to do. If you were engaged in the fiat world, you would have a bank that maintains its own ledger, and it could move the dollars or pesos or whatever into a segregated account, it would be an interest-bearing account, and they’ve got obligations to report to the government. In crypto, we don’t have banks that handle crypto for us. We don’t have interest-bearing accounts that the government has blessed. I know we all wish that there were. And we certainly don’t have any clarity as to what we’re supposed to do if we find ourselves with blocked property. What’s a segregated account? Do I have to make a new wallet that has its own separate private key and then transfer that blocked property to that new wallet?
We have seen a law that was put in place for the legacy world being implemented in a new way against a new type of actors, if it’s even an actor at all, and now we’re grappling with the fact that we’re missing a lot of information that’s necessary for everyone to understand their obligations.
Laura Shin:
Yeah. I think there’s a number of regulators that listen to my show, so if any of them work for OFAC, then here you go, sort of a wish list of what the community is looking for.
So, Martin, let’s turn to you. Before we get into kind of intricacies around the upcoming proof of stake system on Ethereum, can you just explain how the miners on Ethereum under proof of work currently have been reacting to the sanctions?
Martin Köppelmann:
We have seen that some miners have, well, apparently changed their behavior. So, usually kind of the default behavior would be that as a miner, you collect transactions and then you kind of just very simply order them by how much fee they pay and then you put them into blocks, kind of that was initially the default behavior. But now we are seeing that some miners have stopped to kind of include transactions, yeah, into blocks that are touching addresses, and what touching even means is also complicated, but let’s keep in…it’s just more that touching addresses that are on the sanctions list.
Laura Shin:
Before we even get into all the details around proof of stake, I was kind of curious if you at this point, or if the community at this point has a sense of whether a proof of work chain or a proof of stake chain would be more censorship resistant?
Martin Köppelmann:
Yeah. I think that is a very complex topic. I think there are arguments for both. I would be leaning towards proof of stake chain can be more censorship resistant, but again, there are definitely arguments for both.
So, one argument for proof of stake is that in general, the number of participants can likely be higher. So, kind of the concrete example is I myself can, without any issue, run a proof of stake validator at home, while having a competitive miner at home kind of is a much, yeah, kind of more challenging enterprise.
Laura Shin:
Okay, so you’re just saying that it’s more decentralized under proof of stake? Is that what you’re…
Martin Köppelmann:
There are definitely elements that can lead to more decentralization of proof of stake. Yeah.
Laura Shin:
All right, so now let’s dive a little bit more into the details around proof of stake. As we just mentioned, at the moment, the government took this action, but there isn’t really any guidance around how to implement it, so people are concerned about, you know, which of the steps in a proof of stake system constitutes a violation of these sanctions, and so, Drew, I don’t know if you have, you know, how much you’ve looked into the different steps here, or you know, what their level of compliance should be, but do you have any sense if there are certain actions that clearly probably would violate sanctions when it comes to validating blocks under proof of stake?
Drew Hinkes:
Well, I think we need to take a half step back and think about whether the law even applies to validators. We know that there are really two questions here. First, are validators subject to US jurisdiction, or are they otherwise engaged in conduct that would cause them to have liability under US law?
The first question has a lot to do with who is where? Is it a US-based company? Are they using US-based resources? It’s common knowledge available online that a considerable amount of the staking validators are using US-based cloud computing platforms, and so you’ve got some connectivity to the United States jurisdiction there. A lot of the larger validators are also US companies, so there’s no question in that case that you’re going to look at US jurisdiction.
So, then assuming US law applies, then we look and see whether this specific law issue applies to the conduct. The sanctions apply to what are called financial institutions. There’s a statute that defines a financial institution, and a financial institution is a US entity that’s engaged in the business of, and there’s a whole laundry list of stuff. It’s like accepting deposits, and making and granting and transferring and brokering loans and credit, and selling foreign exchange or securities or futures or options and so forth as a principle or an agent. It’s not entirely clear whether validators on a base layer of a blockchain are subject to the law.
What we’re looking at here is a mismatch between the legacy concepts that underlie our definitions of who’s regulated and what’s actually happening on a blockchain. On a blockchain, you don’t have a bank. You don’t have a registered entity that, because they have this dispensation from the government, is allowed to take certain acts where they can take custody of third-party property and they can charge interest and they can make loans and do all this. Instead, we have a bunch of people, whether they’re companies or otherwise, that are deciding to deploy technology resources in order to undertake what I would argue is sort of a ministerial function. They are taking transactions between two willing parties and they’re doing the work that’s necessary to make sure that they are validated and that they become part of the canonical history of the transfers undertaken by that blockchain system.
And so there’s a real question that we don’t really have a clear answer to as to whether these entities that are providing this service to this technology system are financial institutions at all, and if they are not financial institutions, then we have to sort of pivot the analysis from as financial institutions, what are their obligations, to if they’re not financial institutions, do they otherwise have obligations?
It’s possible, if they’re not, that a conservative approach would suggest that they should still undertake some sort of measures, because in this case the juice may not be worth the squeeze. Sanctions is, under US law, something of a strict liability standard. The why you did it or the circumstances in which you violated aren’t really that material. If you violated, you have violated, and then there’s a process by which you can introduce information about mitigating circumstances, and then OFAC will decide whether, after you violated, any sanction is necessary at all. The sanctions can range from incarceration to a fine to public description of what happened to nothing.
And so, given the ambiguity in how the law applies at all, and given the lack of clarity as to whether these actors have obligations at all, it’s very hard to come to any conclusion without more information from the government, and we could get that information in one of a couple ways. The government, OFAC, could provide us with what are called FAQs which layout additional detail that helps those who believe that they’re subject to obligations understand what they can and can’t do and how to do it. And there is also a process by which information as to what someone can or can’t do in light of the sanctioned designation, that can be issued generally, which we call a general license. A general license might be really helpful in that it might say, if you have blocked property, here’s how you designate it in a segregated wallet, and if you do these things, even if they technically violate, we’re not going to hold that to be a violation.
Now, if you have a specific issue as your specific assets, you can ask for something called a more narrow, specific license as opposed to a general license, and you get that sort of license if you have a specific piece of property that you believe is blocked, but you want some clarity as to what you can or can’t do. We need more from OFAC to really understand whether validators are subject to this law at all, and we need more from OFAC to understand really practical, how do I implement this and when do I implement this, sort of questions.
Laura Shin:
Yeah. Actually, from your comments earlier where you were talking about, you know, whether or not something is a financial institution, I happened to see that Martin tweeted about a cloud provider that blocked an Ethereum user. Martin, can you talk about that? Because that indicates that, yeah, some of these tech companies are feeling like they have an obligation to comply. So, Martin, can you talk a little bit about what you saw there?
Martin Köppelmann:
Right. So, I think it just shows how large the uncertainty is about kind of how many layers down the stack are we going? So, it’s unclear whether a validator has kind of those obligations, even those might not have those obligations, but now it turns out that even kind of the layer below, so yeah, a cloud provider in this case, it was Hetzner, so kind of they offer, somewhat similar to AWS, they make it easy to kind of, yeah, run Ethereum node in the cloud, they have started to, yeah, be much more strict against Ethereum usage. So, right now, you are, on most of their services, not allowed kind of to just even run an Ethereum node even if you are not a validator, like a kind of complete strict ban of just running Ethereum nodes, which in my view is pretty insane.
Laura Shin:
Yeah. So, and again, this is under proof of work.
So, now let’s turn to proof of stake, and Martin, why don’t you just walk us through what that validation process looks like, who the different entities are, and then we’re going to kind of step by step ask different questions about what their need is to, you know, comply with these regulations. So, just walk us through how that process works.
Martin Köppelmann:
Yeah, so mainly we are talking about validators, so who are validators? Validators, anyone can be a validator. I’m running the validator for Ethereum, but really anyone can do it with…you need 32 ether and you need to kind of put them in into a smart contract and then you can run a validator, and on Ethereum, there are currently 300,000 of those validators, and it is the task of the validators to build blocks.
More or less randomly or yeah, randomly kind of for each next block, one of those validators is chosen, and now there are multiple roles. So first, someone, and that can be the validator, that can be someone else, needs to build a block, and that really kind of means taking a bunch of transactions, putting them in order, and kind of executing them, and then kind of, well, producing kind of this new block, so the first step is the block building. In the next step, only for each, yeah, we call it slot, so a slot is a potential block, for each slot there is exactly one validator out of those 300,000 chosen, and this is the proposer. So now this proposer has some time, a few seconds, to propose a block that was built, again, it could be built by themselves, it could be built by someone else, and now this block can be proposed.
And in the next step, so that’s the second step, building, proposing, the third step is attestations. So again, for each of those slots there is a subset of all the validators. A subset is responsible to attest kind of to a proposed block, and as a validator with the obligation to attest to block, you have kind of two choices. You can wait for the block and if you see it, great, if you see the correct block by the correct proposal, you can attest to that. If you don’t see that block, and that might be the case because the proposer might not be online, so that usually rarely happens, but it could happen, then you would attest to the latest block, you know, so that’s step three, attestations.
And in a way, then step four is this process again, and then kind of producing a new block and then building on top of that block, and I think kind of for all those four steps, potentially we need to ask questions around how does this relate to sanctions and kind of is that…yeah, I mean, the questions are obvious.
Laura Shin:
Why don’t we just start with the proposers, the ones that are, you know, propagating these blocks out? Since the, you know, process of building the block happens with the block builder, would validators even know if they’re processing a block that has sanctioned transactions in it?
Martin Köppelmann:
Yeah, so that is the…so far, default behavior is that a builder and the proposer kind of, that’s done by the same validator. That is so far the default, but there are a lot of efforts, builder-proposer separation, to yeah, to separate those processes and have specialized builders, and yeah, quite likely, that’s currently a little bit also up in the air, but it seems very much that the default software will support, again, a separation of that.
So, as a proposer, you can kind of point to builder, either to a single builder or to a network of builders, and you kind of get suggestions from different builders, and then you could pick the one that is kind of highest paying.
So, if there is no separation, well then obviously as a proposer you know what’s in the block because you have been building the block yourself. If there is a separation, then exactly, you’re right, as a proposer, you don’t know what’s in the block, so you kind of ask the builders, and what you get from them is just a hash of the block plus a promised kind of payment, and that’s all you know. So, you only know this block will pay so much. In the next step, you would sign it, just this hash, and only after you signed it then the builder would release the block and kind of then those who attest to the block, they again then see the block and see all the transactions.
Laura Shin:
Oh, so only the attesters end up seeing the transactions. Oh, this is so interesting.
So, Drew, when you listen to…well, before we get to the proposer-builder separation part of it, so if a validator is both building the block but then also validating it, what do you think their liability is here? Do they need to censor transactions or is it again what you were saying before, it’s not clear if their function is more financial or technical or…yeah, I just want to hear you kind of analyze what their liability would be.
Drew Hinkes:
Sure. We need OFAC to give us some more information to understand how all of the obligations that apply to listed addresses and to blocked property should apply in this context.
What I take away from Martin’s really clear explanation is that the process of validating is more complex with more actors with different amounts of information. One bit of information that I do want to pepper in just to make it even more complicated is the fact that there are hypothetically two different types of transactions that might be of concern. One would be transactions that include an address that is listed, hypothetically, and the other one would include property derived from a listed address which hypothetically could be considered to be blocked.
Obviously, hypothetically assuming that there was some sort of obligation, it would be a lot easier for anybody looking at a block to look at an address because it’s on this list and it’s in the block and you can see the block, then hypothetically you can potentially do something about it. It strikes me as though it would be incredibly complex and perhaps not practical to further impose the obligation on anyone participating in validation, not just to look up addresses, but also to try to determine whether a given asset that is to be transacted in a block is blocked property. Martin will probably tell me that it would be impossible to do so without some dramatic overhaul of the way that consensus is built or without the addition of other technology tools.
So, I wish, Laura, that I could tell you that I knew exactly what OFAC was thinking. I wish I could tell you that I’ve seen a draft FAQ or something. The government unfortunately does not tell me what they’re thinking. I find out with everybody else when they decide to tell us. The hope is that we will get clarity, either in the form of a general license or in FAQs that can address some of these questions, but for now, we’re stuck in a place where we’re speculating.
Martin Köppelmann:
And I want to build on that. So, even the first option you mentioned, kind of just “blocking the addresses”, even that is technically far from trivial because what is relatively easy is kind of to see, if you see a transaction you can relatively easily see it has a from and it has a to. However, the to can very well be a smart contract and you don’t know what this contract will do, and it’s absolutely not trivial to figure out in advance what this contract will do. So, it’s possible that you see kind of the from is fine, the to is fine, but then the to is a contract and it executes some complicated logic and it calls another contract and calls another contract, and eventually this other contract then finally calls or sends money to a sanctioned address.
There are some people arguing that if you would try to kind of filter those transactions out, that you, yeah, would have kind of DDoS problems, so kind of that the validator could be spammed with lots of transactions and they execute them all just to at the very end learn that they need to throw them away because they touch this forbidden address. That was, by the way, many years ago, also the argument to not do, yeah, what’s called a soft fork in those, kind of the time when the DAO hack happened. So, when the DAO hack happened, before then eventually this fork was done where kind of the hack was reverted, there was a discussion to, yeah, in a way sensor or block the attacker, and exactly because that is so complicated and you can’t really know up front whether a transaction will eventually touch it, exactly because of that reason that plan was given up on.
And so, the other thing would be I really don’t see kind of, to go then even a step further and say, kind of try to block everything that has tokens or coins in it that came from somewhere. Yeah, I can’t really see how that would work.
Drew Hinkes:
Just to build on Martin’s point, remember, these laws are created for banks. These are banks that are required to send information about the sender and recipient, the source of funds, and all the other stuff that travels along when a bank transacts with another bank based on the travel rule obligations, so finding out the source and the origination of a transfer is trivial. All that information comes with the transaction.
What we see with these assets is the complete opposite. Not only do you not get that sort of information, but as Martin suggested, you might be sending between two robots who actually have code that would result in the ultimate destination address being contingent on some third-party factor. Imagine it goes to a smart contract that says if this certain occurrence it comes to pass, then it goes to address A, and if a certain occurrence comes to pass, then it goes to address B. It may be that it’s impossible at the time of the transaction to know where the ultimate destination is.
So, again, this is dramatically different than what we see in the traditional banking system, and it’s one of these complications that, although I don’t expect that OFAC would give us FAQ that cover every hypothetical contingency, some clarity would certainly be beneficial.
Laura Shin:
Yeah. I actually wanted to mention about Martin’s point about the DAO hack, you know, because this came up when I wrote my book, but one of the fun facts about that was that it was a high school student who figured out that this denial of service attack was possible basically by doing that, and I agree, it’s basically the same situation here. All of this is very complicated. It takes kind of a long time to explain, so what we’re going to do is we are going to talk about how all this would work once we have proposal-builder separation, but first we will take a quick word from our sponsors, and we will be right back.
Curious about the world of crypto and the future of the next Internet? Then check out web3 with a16z, the chart-topping technology podcast from the minds at Andreessen Horowitz, the go-to destination for discussions on tech as it changes our world. Whether you’re a crypto curious person looking for signal versus noise in the day’s headlines or a C-Suite decision maker seeking to understand web3 as part of your business strategy, web3 with a16z is the podcast for you. Tune in each week for leading insights from the top scientists and makers in this space, your carefully curated conversations with acclaimed podcast hosts Sonal Chokshi, former showrunner and longtime host of the a16z podcast, along with frequent guest appearances and hosting by Chris Dixon. Listen to web3 with a16z today on Apple Podcast, Spotify or wherever you get your podcasts.
Join over 10 million people using Crypto.com, the easiest place to buy, earn and spend over 150 cryptocurrencies. Spend your crypto anywhere using the Crypto.com Visa card. Get up to 8% cash back instantly, plus 100% rebates for your Netflix, Spotify and Amazon Prime subscriptions. Download the Crypto.com app now and get 25 dollars with the code LAURA. Link in the description.
Is your web3 experience hindered by inadequate crypto wallets and browser extensions? Ava Labs has created Core, a free, non-custodial browser extension engineered for Avalanche users to have a more seamless and secure web3 experience. The best-in-class Avalanche Bridge now offers native support for the Bitcoin network. Put your Bitcoin to work in the robust DeFi ecosystem by bridging BTC to Avalanche today. With Core, you can also easily swap assets, display your NFTs in style, store your assets in a ledger-enabled wallet, and put real dollars into your crypto wallet in just a few clicks. Core is everything you need for a simple, secure and convenient web3 experience. Download the free Core browser extension from Google Chrome’s app store today.
Back to my conversation with Martin and Drew.
So, as we just mentioned, things will look even different once we implement proposer-builder separation on Ethereum, so at that point, you know, as we discussed, the blocks will be created by a different entity from the proposer, so at that point can we talk a little bit about what the liability might be of the different actors? You know, we have the block builders, the proposers and then the attesters, and as Martin pointed out, it’s the proposers who actually won’t really know what the transactions are. Drew, do you have a sense of what the liability might be for each of those entities?
Drew Hinkes:
I fear them to start sounding like a broken record, but again, the laws that are in place that we’re grappling with were not designed for protocols that facilitate transactions. They’re designed to regulate actors that are legally addressable and that have obligations under enforceable law. We’re thinking more along the lines of banks rather than of those that are part of the plumbing.
Laura Shin:
Yeah, but is there a way to say something as simple as, since the builders and the attesters will actually see the transactions that their liability might be higher or…
Drew Hinkes:
Again, the keystone issue is does this law apply at all? Are these the sort of actors where the law would potentially apply? I guess we should talk about secondary liability which is notwithstanding whether you have any obligation you’ve registered or not, if you are aiding and abetting third parties in transacting blocked property, for instance, you may have liability. Again, we have no idea how OFAC thinks about block validators or miners or those that operate the base layer.
You can make an argument that, for instance, banks are regulated, but the SWIFT protocol by which banks communicate with each other is not regulated. OFAC doesn’t tell the SWIFT network, don’t send a message here. It tells participating banks, don’t honor messages from this bank. We saw with Russia, for instance, Sberbank, which is the largest bank, I think, in Russia, used to be a member of SWIFT and there was political pressure placed on SWIFT so that the member banks, which include non-US banks, would discontinue communications over SWIFT with Sberbank. So, it may not be the cleanest analogy, but there is an argument to be made that we should think about the base layer, irrespective of whether there are entities that are providing services on the base layer, as if it’s a communications protocol and not a set of legal actors with obligations. We don’t know how OFAC thinks about this, so it’s too speculative for me to really take that anywhere.
Martin Köppelmann:
What I can do at least is try to make the kind of technical argument why they should not. So, the task of the builder can really be described as a pure mathematical optimization problem, and actually, yeah, it is also possible to represent the problem very much abstracted away from the concrete addresses. So like, mathematically speaking, you have just a bunch of items and you want to pack them together in the most efficient way so that you in the end generate the most fees. Again, you could convert this problem to really pure, yeah, mathematical optimization problem. So, if solving that or working on this optimization problem is allowed, then probably you could argue the builder doing the builder’s job is fine, because that’s really all they do.
So next, the proposer, what they do is really just query a bunch of those solvers, or kind of those who try to solve this problem as good as possible, and all they do is pick the one that pays most. Again, that’s all they do, and at this point, when they sign it, they don’t have any knowledge about what transactions are in this block.
So now, the third step would be those who attest to the block, and again, here they are in a situation where they have only the choice to say kind of this block exists and I’m attesting that it exists, or well, they could kind of decide to lie about it and say it doesn’t exist, but yeah, those are, again. the only two choices. So, as long as they kind of don’t maliciously act against the protocol and pretend something doesn’t exist that does exist, they also don’t really have any choice here.
Laura Shin:
Why don’t we now talk about the fact that some good percentage of the proposers/validators are going to be these large entities that have US exposure, Coinbase, Kraken…I don’t know, actually, I tried to figure out, does Lido have US exposure or do we know if it’s going to follow sanctions?
Martin Köppelmann:
Yeah, so with Lido, it is slightly more complex again, because Lido is somewhat a layer in between validators. So, how Lido works is that plenty of people can kind of deposit their ether to contract, and now Lido, yeah, kind of a DAO or entity, whatever you want to call it, is splitting up that ether to many individual validators. I don’t know the exact number, but it’s at least 20. Twenty, then, validators said this ether is, yeah, kind of delegated or given to, but in such a way that they don’t control the ether, but they can perform this attestation and propose a task. So, because yeah, you can set up a validator and you have a kind of a specific withdraw key which kind of means you are in control of the ether, and that’s not given to them, but the key that is used for those attestations and proposing that is given to them. So, asking deeper, I assume one of those 20 validators was in Lido, I very much assume there are also US entities in it, but I don’t know.
Laura Shin:
In general, if we’re going to have…so if we add up Coinbase, Kraken, Binance, unclear, they at least have Binance US, Lido also unclear, but all of those together, that’s like 60% of all validators. I don’t know what your perception is around what the risks are posed now to validation if we have some of these major entities that at least some of them do have US exposure.
Martin Köppelmann:
Yeah, so there are a few things that could happen. So again, I guess the happy case is it turns out it’s fine for them and nothing happens at all. There is then of course the case that they feel that they are obliged to kind of not include those transactions. That would probably mean they can’t use proposer-builder separation because again, if they use it, they don’t know what they are proposing or they would need to specifically work with builders that, yeah, kind of build according to some rules.
Then there is still the case where even if it’s 60%, even if it’s 70% of the validators would do that, what that then effectively would mean, that specific transactions will essentially kind of take longer, because if you do a specific transaction that is filtered out by some rules, you will probably need to wait until a validator comes that uses different rules and will include your transaction. Of course, like if we are talking about those hypotheticals, then I guess the question is even are nodes then even allowed to broadcast those transactions, but okay, putting that aside, so kind of that could be one case that those transactions will just be kind of slower because you need to wait for such a validator. The thing where it really becomes, could threaten Ethereum at its core, would be if kind of those validators that follow different rules would not even build on top of blocks that violate those rules.
Laura Shin:
Well, do you have any indication of whether or not they’re thinking about trying to not build on blocks that contain sanctioned addresses?
Martin Köppelmann:
Yeah. Yeah, so far, everyone absolutely said that they would not do it. Yeah, the Coinbase CEO was even saying on Twitter, before doing that they would halt or kind of stop all their validators. So far, absolutely no one saying is that they would do that, but yeah, that would be theoretically possible and that could really threaten Ethereum.
Laura Shin:
Yeah, I did see that Nic Carter of Castle Island Ventures did an op-ed for CoinDesk and in it he wrote, the board won’t allow Brian Armstrong, the CEO of Coinbase, to shut down Coinbase’s massively lucrative staking business because some .eths were mean to him online. So, I mean, who knows? Obviously, we don’t have somebody from Coinbase here, but you know, it is a good point that Brian may have said this on Twitter, but when it comes down to it for business reasons, who knows what they would actually do.
So, one thing is, you know, because Brian did raise this issue of them shutting down their staking business, I was curious, how long would it take a large staker like Coinbase to exit? Like, I read Eric Wall surmised maybe a month or longer, and then I didn’t know, like during that time, would they still be proposing blocks or like, how does that part work?
Martin Köppelmann:
I think we need to also separate here two things. So, it is possible right now to kind of stop being a validator. What is not yet possible is to withdraw your stake. So, we are still in this period where we are still pre-merge and even with the merge, the transition to proof of stake will not be completed, because again, eventually it should be possible to kind of, yeah, stop being a validator and essentially, eventually having your ether again on the Ethereum blockchain, having it transferable. So, this moment when you can, yeah, withdraw stake, that is expected to happen, yeah, six to twelve months after the merge.
But what is possible right now and is immediately possible is kind of to stop your validator duties, yeah, and that means a few things. That means, first, you will no longer be choosing to, yeah, propose and attest to blocks, and it’s important to note that if you if you stop attesting to blocks while you are still an active validator, you are penalized. How much that is depends. So, if very few nodes in the network or validators in the network disappear, those penalties are fairly small, so kind of being one day offline is just the equivalent roughly of what you would earn for being one day online, so kind of it wouldn’t be a big penalty if out of a year you are two days off offline, and that’s important because we want kind of also to have it possible to be a validator if you don’t have a professional setup and you kind of just run your validator from home.
However, if a large percentage of the network kind of is not online, then those penalties can be very significant, and that’s also kind of needed because if a large percentage of the network would be offline, and that is a little bit how it would look like from a network perspective if some validators would decide to not build on top of specific blocks, because for those who do build on it, they kind of just see their version of the history, and in their version they just see that kind of some percentage is missing, and they don’t necessarily see that they are building their alternative version of history.
So, kind of on a chain that would be, like let’s say then uncensored, those validators that are not attesting would kind of potentially dramatically lose their stake, because on that chain they are offline and a lot of them are offline, so they will kind of…essentially, their stake will disappear in a way, until that chain is then able to what we call finalize again.
So, finalize means that enough of the active validators attest to a block.
Laura Shin:
And wait, so are you saying that in that instance then, that would cause a fork, or are you saying it would cause a reorg, or…I don’t know if I fully followed.
Martin Köppelmann:
Yeah. I mean, both of those things are possible. Again, we are talking about this hypothetical scenario that some validators would start just ignoring blocks although they are correct proposed and they get correct attestations, but if you would have the policy that in your kind of view you would see them as invalid because again, they violate some additional rules that are not part of the Ethereum protocol, then yeah, you would probably build a block and that would potentially cause a fork, yes, and then kind of on the other fork, you would lose your stake.
Laura Shin:
Okay. Yeah. One thing I wanted to just mention about that was obviously…so, as people saw, like Coinbase, you know, is a large staker at the moment, my guess would personally be that if for whatever reason the money that they had staked was slashed, I feel like the company would somehow make it up to their clients. Even though some number of them are probably institutional, it would probably be a lot of money, but for the retail, I imagine they would.
Why don’t we just move now to forks? I do have some other questions about those entities, but since we’re talking about forks right now, why don’t we just talk about this, because there were people talking about doing what’s called a user-activated hard fork to prevent censorship on the chain, and this would be kind of a user-led movement to punish validators who do censor the chain.
But Martin, just from the way you were describing things, do you think that that would then essentially lead to two different chains, or what’s your take on what would happen if people were to try to do that?
Martin Köppelmann:
Again, it depends on many factors. So, if a minority of validators would start, yeah, what I would call adding additional rules to Ethereum, because that’s what it is, if you say kind of specific transactions are not valid, then by Ethereum standards it is a valid block, by your standards it’s not, so if a minority would kind of ignore valid blocks, Ethereum valid blocks, then they would essentially create their own fork and that would be kind of in a way fine for Ethereum.
The issue comes when that becomes the majority. So again, this is very hypothetical, but if a majority of validators would agree on the same rules, and that’s very far from trivial because like, even if they all agree they, well, want to honor those or kind of want to kind of act according to those sanctions, it’s absolutely…I mean, there are so many edge cases where it’s not at all trivial kind of to then make a clear distinction between valid and not valid, and that would be required to have a coordinated fork.
So, if that would happen that the, I don’t know, 70-80-90% of validators would kind of add new rules and kind of try to go for such a fork, then yes, there is a theoretical, again, all very theoretical, option of a user-activated fork where users could decide to essentially manually remove those validators from Ethereum or kind of remove their ether balances, remove them from the validator set and continue with a validator set that acts according to Ethereum rules. Again, all very hypothetical.
Laura Shin:
And I just have a question because it wasn’t totally clear to me, people were also talking about something called social slashing. Is that the exact same thing that we’re talking about or is that…
Martin Köppelmann:
Yeah. That would be social slashing. Exactly.
Laura Shin:
Okay. Well, as I read, social slashing isn’t something that’s in the protocol, so it would sort of be like people taking it on them upon themselves, like vigilante style to do this. Like, is that even enforceable?
Martin Köppelmann:
Well, I mean, in a way, that is what happened in the DAO fork. So, I mean, to be clear, of course the fork that reversed the DAO hack was also not in the protocol, but yes, it is ultimately a community power or kind of a network power to make forks. Every fork is a something that’s not in the protocol or was previously not in the protocol.
Laura Shin:
Okay, so I’m curious to hear from the both of you, just what do you think are the risks that we’ll end up with two chains, sort of like the OFAC compliant chain and then the censorship-resistant chain?
Drew Hinkes:
I think it is unlikely that we end up with a chain where under no circumstances, for instance, blocked property could ever be transacted for a very simple and obvious reason. If you’re in possession of blocked property, you’re required to send it to a segregated account. If I’m in a centralized bank, the bank can just move some asset on a subledger and say, okay, I created a new account, it’s interest bearing because I have the legal power to do that, and I’m going to file my report with OFAC, and I’ll keep filing my annual reports, and the bank customer just sort of goes along with their life, and if OFAC wants to investigate, they might, and they might inconvenience the customer, and great or not great, however it shakes out for them.
In this context, if a recipient of blocked property, and let’s ignore the fact that we don’t really understand if transactions that come from Tornado Cash listed eth addresses are blocked or not, the safe, conservative approach would be to say, yes, but there are real questions if it is, leaving that issue aside, let’s say that there was property that came from say the Chinese fentanyl ring whose eth addresses were designated on the SDN list in 2020 by OFAC. Let’s say that you unwittingly, unknowingly, by no fault of your own, received some eth from them. If you wanted to comply, you would try to figure out well, how do I segregate that? Any effort to segregate requires a validator to validate the transaction. I can’t use a magic wand and just move my eth without having it confirmed by the network. So, unless OFAC tells us how, we logically would need validators to validate a transaction of blocked property. Now, how could they do this? They could give us a general license that says, if you want to segregate your blocked property, do this, and that would hypothetically give the validators some comfort that they could validate. Again, broken record, we need OFAC to tell us what to do. But if we are to adhere to what it appears the law requires, it seems unlikely that we would ever be in a world where absolutely no transactions of blocked property are allowable for the simple reason that the law seems to require them.
Martin Köppelmann:
Yeah, so to make that also very concrete, that is one thing that happened a few days after the sanction list, yeah, got announced, that someone started sending tiny amounts from one of those addresses to many, many, yeah, kind of prominent Ethereum addresses, so kind of Ethereum addresses of celebrities, of, yeah, I think probably…
Laura Shin:
Shaquille O’Neal. Brian Armstrong. Yeah.
Martin Köppelmann:
Exactly. Exactly, so all those addresses are, well, potentially blocked or whatever, kind of theoretically, and it would be fairly trivial to do that at a much larger scale than it was done. I was previously, well, obviously joking, but saying kind of in theory someone could write a bot that watches out for any larger deposit to an exchange because it’s fairly easy to see, and while that transaction is in the in the mempool, so kind of not yet confirmed but visible, well, someone could send one cent to the sender. So now, kind of at that moment when the exchange receives the money from that sender, kind of the exchange would need to acknowledge that just a second before, that address received one cent from, yeah, kind of the sanctioned address.
So, I guess where I’m trying to go, it is incredibly hard to kind of come to any practical kind of solution for how to deal with those sanctions, so that is again also why I would answer your initial question, how likely is it that we will see a fork? It’s very, very unlikely, because it’s, again, very, very unlikely to find to find a technical, practical rules for Ethereum, how Ethereum could meaningfully exist. So, I think to implement those, yeah, kind of those sanctions in their strictest form, I think you would need to turn Ethereum completely on its head and say it’s by default a permissioned system, so kind of…right now it’s kind of…right now anything that is…by default, kind of any kind of transaction is allowed, and if you would want to do this strict censorship, you would need to have, yeah, just a whitelisted list of transactions, of transaction types that you are allowed to do. So, you could say, yeah, sending ether between two addresses that are not on the list, that’s fine, but then immediately as it comes to contracts, you would say, yeah, only contracts where we know they only call those kind of…so again, I think that would be very different from what Ethereum is.
And by the way, it is possible to build such a permission system on top of Ethereum, but I don’t really see that it is possible to build that into Ethereum. So, you can have a permissioned system on top of a permissionless system, but not the other way around.
Drew Hinkes:
One quick clarification just for the listeners, this is complicated stuff, and we don’t usually spend a lot of time thinking about sanctions because it’s not something that usually kind of invades the conversation like this, but just a little bit of a clarification. An address receiving blocked property does not make the address blocked.
When a bank account receives a transfer that should be blocked, some of the cash that’s received needs to be segregated. One of the questions that we have, and this is an eth question but more of a crypto question writ large, is if an address that has five eth in it receives 1/10 of an eth, what exactly are we segregating? Eth are fungible and balance based, so can we just choose any 1/10 and move it to some other address and move on with our lives like we would with cash, or is there some obligation to try to identify the fraction of eth that was received, if that’s even possible?
If we’re looking at this on a UTXO-based type of system like Bitcoin, it’s actually easier if you catch it because you can actually identify the fragment of Bitcoin that you’ve received, the actual UTXO, and segregate that, but again, this is a question that we don’t have an answer to from OFAC, and this is one that we can probably get an answer to through FAQ. But the big picture is you don’t become a listed account, right, a listed address simply because you’ve been greased or spammed.
Laura Shin:
Yeah, why on social media you probably saw a bunch of users being like, I didn’t even engage with Tornado Cash. Like, why am I being blocked from this or that DeFi protocol? Or even someone was like, I was hacked, that’s how my money, my account got associated with Tornado Cash, because somebody stole the money from it, ran it through Tornado Cash, but like, that was before it hit Tornado Cash and they were being blocked, so…
Martin Köppelmann:
Andrew, I’m just thinking, that’s important that you say that, but the unfortunate reality is that a bunch of projects kind of went so far and did that, so essentially said, and again, the reason is the uncertainty, they then call it, yeah, risk-based approach and say kind of this address received money out of Tornado, and then we just ban it in our service.
Drew Hinkes:
Really quickly, what I’ll say in a podcast under a set of hypothetical questions and conditions is different than how I’ll advise a client when I have specific facts, but we’re talking about…most of our conversation has been about those who are serving third parties and want to transact on a base layer. Different companies that provide different services at different layers may have different levels of exposure and different legal obligations, so I would caution perhaps some sympathy or pathos for different folks in the industry that are looking at this challenge with a different set of obligations. Everyone is hamstrung to a degree by the fact that we’re trying to implement 20th-century rules on 21st-century technology.
Laura Shin:
Yeah. One other kind of ludicrous scenario, you could call it, that Eric Wall, a bitcoiner who’s also very interested in Ethereum pointed out, he wrote this amazing blog post, The Case for Social Slashing, which people should read because it walks through all different kinds of game theory around sanctions on Ethereum, but essentially, you know, he concluded, okay, so you know, here, if we end up with this chain that does follow sanctions, then who’s to say, like, what if, you know, people want it to follow the sanctions of this other country that, you know, has interests different from the US or whatever? And so, like, yeah, it just gets to this place where it is kind of ridiculous and it sort of shows how unenforceable this is.
Drew Hinkes:
I want to push back on that for just a second. The US is big in a lot of ways. I think we still have the biggest economy in the world and I know for sure that we do the most sanctions in the world, and most other sanctions lists are based, in fact if not in concept, on the United States. We are the most sanctions-heavy of any jurisdiction. So, while the idea that we could splinter Ethereum based upon the whims of certain specific subgroups into a bunch of groups of validators that lose the ability to remain in consensus with others because they change the protocol rules so they can’t form new blocks together could lead to a number of new versions and distinct chains, we have seen that the majority of validators will continue to be supporting the chain that most will use.
And in the context of Ethereum, where there are so many material commercial considerations that are built and operated on the second layer, above and on Ethereum, we’ve seen commercial interests indicate, even if there is a POW fork that remains after the merge, the authoritative, canonical supported, commercially viable version of their product will be on the POS system as it goes forward. If there were to be nine different sanctioned versions, I suspect that those actors would do the same thing.
Laura Shin:
Yeah. Before we actually get to that, I just want to ask one last…oh, Martin, did you want to say anything?
Martin Köppelmann:
Yeah. Just one comment to kind of the United States as the country with the most sanctions, maybe that’s true, but it’s certainly not the country with the most restrictions. So, I mean, like even if it’s not called sanctions, other countries have all kind of restrictions what you are allowed or not allowed to do, and I would argue on that level, well, the United States are more on the, you’re allowed to do more. So, kind of if Ethereum would need to kind of make sure that all the, I don’t know, kind of restrictions that are applied in China for currency control, or kind of all those things, would also need to deal with them, again, that would make the system, yeah, impossible to operate.
Laura Shin:
Yeah. Before we move on to some of the things that Drew was mentioning, which refers basically to DeFi, I did want to ask about one last piece about the validation process, which is the fact that Flashbots will be the default relay on Ethereum at the time of the merge. The company has already announced that its relay will be censoring transactions that involve these sanctioned addresses, and so they said that they were going to be open sourcing the relay code to encourage other ones to be built, but obviously it’s a short timeline. They only announced this roughly like a month before the merge was going to happen. I’m guessing that Flashbots still will be the default relay at the time of the merge. I don’t know, Martin, maybe you’ll tell me I’m wrong, but I was wondering, if that’s the case, does that mean that effectively Ethereum will be complying with sanctions for some period and will not be censorship resistant for the first, I don’t know, month or two of launch?
Martin Köppelmann:
Yeah, so a few important pieces here. So first, it’s still a choice of every individual validator whether or not they use builder-proposer separation, so it’s absolutely still your choice to build your own blocks and therefore, yeah, in a way, not use Flashbots or MEV-Boost. The second thing is, yes, I think you mentioned it, so Flashbots used to be, for a long time, the most prominent builder, although there have always been others. There has been a project called Eden Network, whatever, there have been others, but they kind of open sourced their software. Yeah. It’s called MEV-Boost and that kind of, yeah, facilitates this builder-proposer separation, and yes, they will probably run a builder, or if you want to go even more into details, they will run a relay, so kind of that connects various builders to the validators to the proposers, but yes, there was already an announcement that there will be at least a handful of other relays. Some have said, of course we will only see, time will tell, that they will not apply sanctions or kind of filter for those transactions, but bottom line is there will be some validators that simply don’t use it, that simply build their own blocks according to their own rules, and yes, there will be a variety of builders with different strategies.
Laura Shin:
Okay, so now let’s turn to the DeFi issues, and really, I think, well, you can name…there’s two different factors here. First is that a number of these DeFi entities do have US jurisdictional exposure, and then on top of that, as I’m sure you’re well aware, USDC is managed by a consortium that, you know, the two most prominent members are Circle and Coinbase, also both US entities, and USDC is something that’s just interwoven through the fabric of DeFi. You know, I would almost say that DeFi is, like, somewhat dependent on it. I just saw it was the top asset for borrowing and lending on Compound, it backs more than half of Dai, et cetera. So, how do you feel that, you know, these issues around both different DeFi protocols’ exposure to the US, and you know, the dependence on USDC, how do you think that’s going to affect how things play out? Because a part of me feels like it almost pushes things in the direction of two Chains, but…
Martin Köppelmann:
Yeah. No. I mean, that is a very, very big question of, again, how practically USDC decides to deal with the situation, so you mentioned correctly that at the moment, that might change very well, but at the moment a very big percentage of Dai is backed by USDC, and I think that’s a little bit provocative statement, but some have called Dai wrapped USDC, and Dai, for example, doesn’t have the ability to restrict individual accounts. So, with UDSC, an individual account can be frozen, with Dai, it cannot be, and of course, in general those sanctions or kind of restrictions, if they are very easy to circumvent, they will just, I mean, not be effective or kind of not really do anything.
So then, of course, then always the question is, if they are being circumvented, what will the next step be? And of course, ultimately, the step could be, for USDC, that they would somehow need to be, well, would probably be pressured or kind of see an obligation to restrict the use or restrict the form of how Dai is using that. That is a big open question. There are multiple outcomes here, possible outcomes. One could be that, yeah, Dai is kind of trying to get independent from UDSC. Another one can probably also be that more measures are implemented that makes it somehow okay for USDC to continuously use this role, but yeah, certainly lots of open questions.
Laura Shin:
Yeah. I just want to point out though, that when Rune Christensen, the founder of MakerDAO, suggested that they sort of de-peg Dai from the US Dollar to try to reduce that dependence on USDC, then even people like Vitalik said that they thought that was a terrible idea, so.
Martin Köppelmann:
I think that was somewhat joking. He was saying kind of, yeah, they should just market dump or kind of just use all the USDC to buy ether, and I think that was the part that…I mean, that was definitely a joke and that was the part where Vitalik said shouldn’t do that.
Laura Shin:
Oh. Oh. Oh. Maybe I didn’t read it carefully. So, I thought he was saying…so wait, are you saying that Rune wasn’t serious about trying to de-peg?
Martin Köppelmann:
Oh, well…okay. No. I mean, there are multiple things. So, I think there’s clearly an understanding or there’s a clearly an attempt in the Maker community to become more independent of USDC. The question is, is that possible and still keeping it pegged to the dollar? So, there are some stablecoin projects they are not dependent on USDC, but they have often given up on the one-to-one parity with the dollar. They are fairly stable against the dollar. They, I don’t know, kind of go up a few percent, but still, they go up, I don’t know 3-4%, go down 3-4%. So, if you allow for that volatility, it’s much easier to design stablecoins that are not kind of backed by dollars, and I think that was or that is a potential question for Maker. I think currently there is no majority kind of supporting that, but that’s at least an idea, or kind of that might be a consequence of saying, if we really want to get rid of fiat-backed collateral, then Dai might need to get rid of the one-to-one pegging to the dollar.
Laura Shin:
Oh. Okay. Interesting, and Drew, did you want to add anything about kind of the power of USDC here?
Drew Hinkes:
I’ll keep my comments brief on this point. If you’re a US-based entity, if you’re regulated or if you are clearly within the US jurisdiction and you’re looking at what you’ve seen from OFAC, for the 18th time in the call or on this podcast, I’ll say we don’t have enough information to understand what our obligations are. However, the magnitude of the penalties here are so severe, this is potential jail time. It wouldn’t surprise me if most significant US-based companies were taking a conservative approach. It doesn’t mean that they’re not going to also try to find ways to challenge the designations here. It doesn’t mean that they’re not going to introduce lawsuits. It doesn’t mean they’re not going to ask for licenses. It doesn’t mean that they’re not going to deploy lobbyists over to Treasury to try to get FAQ on things that they care about. When you have this sort of lack of guidance and lack of understanding as to how to implement what you are obliged to do, a lot of times lawyers will counsel doing nothing rather than taking some sort of aggressive action, because if you take the wrong aggressive action, it could actually exacerbate the problem rather than solve the problem.
I would love to see the FAQ issued, you know, the day that this podcast is run. It would be amazing and it would probably moot everything that I’ve said because it will answer all of the questions. However, government entities are not always moving at the speed of light. A lot of times it might take them a while. They’re going to engage with industry. They’re going ask questions. They’re going to listen to the best podcasts in the industry. They’re going to come up with their own ideas and they’re going to battle test them with important stakeholders, and so the process of coming up with the guidance that we need, the FAQs or issuing general licenses to empower us to take certain actions that on its face may be prohibited is not going to be fast. So, for regulated US actors, they are probably going to be conservative and they’re probably not going to move terribly quickly.
Laura Shin:
And out of curiosity, is that typical, in situations like this involving sanctions, for the government to just issue, you know, some kind of ruling like this and then later explain how it should be implemented, or does this sort of indicate that there was sort of hasty action taken before they had time to think through everything?
Drew Hinkes:
So, I can’t speculate as to the deliberations that went on within OFAC. It is typical or not uncommon for there to be a designation and then, after a period of time, FAQs that provide additional color as to how to comply or that provide additional information that’s helpful to interpret the meaning of the designation. If you go on the OFAC website, for instance, you’ll see that there are searchable FAQ that are actually grouped by topic, and there’s, I don’t know, 12 or 13 that actually address virtual currency, and some of them are actually quite helpful. If you’re a centralized entity, there’s actually some guidance in the FAQ as to how to deal with blocked property. What it does not tell you is if you find yourself the recipient of blocked property in a self-custodial wallet, for instance. There is some guidance if you’re in exchange, but not if you’re a self-user or a self-custody user. So, it is typical that there is some deliberation internals of the organization before they put out the FAQ, sometimes they revise FAQs to update them, and as I said before, there are certain issues here that are going to be widespread where a general license, which would be published and public and anybody could read it, would be an effective way to address some of the issues.
Laura Shin:
Okay. Well, clearly there’s a lot of questions that this has opened and we’ll sort of have to see when it is that the government finally issues some FAQs and also what those say.
In the meantime, are there any parting thoughts as we head into the merge in a few weeks?
Martin Köppelmann:
Yeah. I want to just make a high-level comment on the whole situation. I have been passionate in a way about, well, kind of Tornado Cash, and I a little bit want to explain why and kind of maybe a little bit zooming out. So, to me, what we see happening is that, yeah, if you use the blockchain more and more, and I am a heavy blockchain user since many years and that is absolutely the exception that is only kind of a very small number of people who are kind of today using blockchain on a daily basis, but if you do that, you leave an incredibly large trace of, yeah, kind of transactions. You essentially, yeah, you could imagine, let’s assume every payment you do, every bank account you have, kind of everything would be public and visible by anyone. That’s essentially what is happening on the blockchain, and again, it doesn’t affect you if you kind of only bought some coins two years ago and they are just lying there, then that’s fine, but if you if you do daily transactions, and again, we are moving there, then that is the trace that you are leaving.
Now, Tornado Cash was a project or is a project that kind of, well, didn’t necessarily invent a new technology. The technology was around in different form. Very similar technology is used in Zcash and was available for some time, but they kind of just went ahead and took the pieces that were largely available and made some usable product for Ethereum for the first, where it was practically for the first time on Ethereum doable for at least advanced users to, well, have some privacy, and that can be as simple as, yeah, if I want to send someone money and I do that all the time and I kind of also do transactions all the time that are associated with, I don’t know, kind of my address that you can immediately find if you look me up on Twitter, and if I just want to make sure that I also have some addresses that are not linkable to that, if I don’t want to kind of expose to the whole world all my crypto, I needed to use a tool like Tornado Cash.
Now, of course, the problem was and that is a real problem, that only a very small number of…or it’s maybe also a false statement, but a fairly small number of users are kind of in that situation where their life is in large parts on the chain, so in a way, the number of users kind of who really needed, in a way, Tornado Cash, was small, absolutely not…I mean, absolutely real users, so in still thousands, but kind of in the grand scheme of things, small. Now, unfortunately, also a very large and real user was, of course we don’t know that for sure, but it seems like this North Korea state-sponsored hacker group at least…I mean, at the end we…that’s almost always very hard to know what’s true in those cases, but let’s assume it’s true, then that is a very real problem and that is very real harm that in a way was done, and from that perspective, on some level I kind of understand those sanctions.
The issue is now that it creates now a situation where it is almost, yeah, kind of impossible to work on, yeah, kind of good solutions to this problem. So, the problem is, how can we on the one hand provide people some level of privacy and on the other hand still hopefully make it as hard as possible for hackers to kind of obfuscate their stolen funds, and there were actually, yeah, some interesting proposals of how you could tweak Tornado or something like Tornado, various proposals. One was kind of to say you need to make a deposit and then there is some waiting period where those who have already deposited can kind of decide whether this new deposit is accepted, or you could kind of reveal, when you withdraw you can reveal that your deposit was not coming from let’s say those stolen funds or something like that. The issue now is you won’t even find someone who is confident working on that because it’s so unclear what you’re allowed to do or not to do, so it’s kind of unclear. Are you allowed to work on that technology? I mean, obviously, there was, to my knowledge, absolutely no law kind of that prevented the Tornado guys from writing this code and publishing this, and now one of them is in jail, so that’s of course a horrible situation and makes it very, very hard to, yeah, come up with good solutions to those hard problems.
Laura Shin:
Yeah. I think so many of the points that you made are super interesting and valuable and should be listened to by regulators who, you know, have any say on what’s been happening.
Drew Hinkes:
I had two quick parting comments. The first is it’s unclear whether OFAC understood the full implications of its designation of the eth addresses associated with Tornado Cash. There’s some language that suggested that it was considering Tornado Cash the same way it considered other centralized entities operating mixers, and if that’s the case, then it’s likely that they did not understand the broader implications of designated smart contract addresses and all the complexities that they were bringing into the market. If that is the case, then we should all be looking at this saga as a really important opportunity to engage with regulators, to educate regulators and to partner with the government so that there is a good dialogue between the industry and those that are making decisions that impact the industry. I can’t underscore enough how important it is for us to educate those who are making the rules for our products and systems and services and businesses. If they don’t understand what they’re doing or what these systems and products are doing, it’s unreasonable for us to expect that they always get the rules right. It needs to be a two-way street, and it’s something that the industry should embrace because it’s to their benefit.
The second point goes to what Martin said before, which is at this point we are running out of ways to anonymously or privately transact. In the United States, we have this amazing thing called the Fourth Amendment which gives us tremendous protections over our persons, our papers and our homes, and we have progressively seen those rights eroded over time because there are regulated intermediaries that have been deputized by law enforcement as basically agents for their purposes. National security is incredibly important, but we still are allowed to use cash, and cash at this point is one of the few ways that we can transact without being subject to surveillance. The other is crypto. So, while Martin’s point with respect to we should be working on privacy and anonymization technology, but we don’t even know whether we can anymore, I think that we should continue to work on it. I think that projects like Aztec are important and continue to push the envelope for private transactions, and I think that we need to continue to fight for the rights that we have in our respective jurisdictions to continue to be able to transact anonymously. We can do it with cash. We should not allow our desire for convenience in payments to erode all of our rights to privacy.
Laura Shin:
All right. Well, great words to end on. Now, do you guys each want to say where people can learn more about you and your work?
Martin Köppelmann:
Yeah, I’m usually…kind of my main work is around Gnosis. We are also working on, yeah, the Gnosis chain, the Gnosis blockchain, a chain that also puts decentralization and having many validators that run the validator from home first, but about me, you find me pretty active on Twitter @koeppelmann.
Drew Hinkes:
This is drew, you can follow me on Twitter @propelforward. You can look me up on my law firm’s website, K&L Gates, and of course, if you’re in New York and feel like learning, feel free to enroll at NYU where I teach a couple different classes on regulation of crypto assets, blockchain systems, governance, and so forth.
Laura Shin:
Perfect. You guys, I have loved this conversation. Thank you both so much for coming on Unchained.
Martin Köppelmann:
Thanks for having us.
Drew Hinkes:
Thanks.
Laura Shin:
Thanks so much for joining us today. To learn more about how the Tornado Cash sanctions affect Ethereum at the protocol layer, check out the show notes for this episode.
Get exclusive access to even more of my content through Bulletin, including interviews you won’t find anywhere else, weekly news roundups, and more. Go to laurashin.bulletin.com/ subscribe.
Unchained is produced by me, Laura Shin, with help from Anthony Yoon, Matt Pilchard, Juan Aranovich, Pam Majumdar, Shashank, and CLK transcription.
Thanks for listening.
Comments are closed.