Unchained - The Chopping Block: Why Everyone Is Talking About EigenLayer - Ep. 497

Episode Date: May 25, 2023

Welcome to “The Chopping Block” – where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner, chop it up about the latest news. In this episode, EigenLayer founder Sreeram Kannan explains his vision for providing “decentralized trust” to new projects on Ethereum. Was Vitalik Buterin’s recent “re-staking” post a shot across EigenLayer’s bow? Listen to the episode on Apple Podcasts, Spotify, Overcast, Podcast Addict, Pocket Casts, Stitcher, Castbox, Google Podcasts, TuneIn, Amazon Music, or on your favorite podcast platform. Show highlights:  how Sreeram has transitioned from academia into the world of crypto why he believes that crypto is the “coordination highway” what EigenLayer is – explained to a five-year-old, a high schooler, a day trader, a developer, and a crypto professor how middleware enables technology innovation where the value accrues if there’s no token for securing a blockchain whether Vitalik is right in being concerned about the risk of corrupting validators what EigenLayer did to minimize risks and externalities to Ethereum whether protocols and applications will fork if they get hacked and whether Ethereum will “bail them out” Hosts Haseeb Qureshi, managing partner at Dragonfly  Robert Leshner, founder of Compound Tom Schmidt, general partner at Dragonfly  Tarun Chitra, managing partner at Robot Ventures Guest Sreeram Kannan, founder of EigenLayer Previous appearances on Unchained:  Do You Need to Think Twice Before Restaking Your Assets? Disclosures Links Don't overload Ethereum's consensus by Vitalik Buterin Previous coverage of Unchained on Ethereum staking:  Shapella in the Rearview: After Major Upgrade, What’s Next for Ethereum? How Will ETH React to Ethereum’s Shanghai Upgrade? Post-Merge, If Lido Becomes Dominant, What Does That Mean for Ethereum? In the Recent Crypto Market Meltdown, What Role Did Lido’s stETH Play? Is ETH on Its Way to Becoming Ultra-Sound Money? Yes, Says Justin Drake Learn more: How Liquid Staking Works A Guide to Ethereum’s Shanghai Upgrade How to Stake Ethereum With Liquid Staking Learn more about your ad choices. Visit megaphone.fm/adchoices

Not a dividend. It's a tale of two Kwan. Now, your losses are on someone else's balance. Generally speaking, air drops are kind of pointless anyways. Unnamed to trading firms who are very involved. D5.Eat is the ultimate problem. D5 protocols are the antidote to this problem. Hello, everybody. Welcome to the chopping block. Every couple weeks, the four of us get together and give the industry insider's perspective

on the crypto topics of the day. So quick intro, this first you got Tom, the Defy Maven and Master of Memes. Next to we've got Robert, the Cryptoconisour and Captain of Compound. Then we've got Tarun, the Gigabrain, and Grand Pubod, Conlitt. And today we've got special guests, Sri Ram, the luminary leader of EigenLair. And finally, there's myself. I'm a seeded-the-head man at Dragonfly. So we're early-stage investors in crypto, but I want to caveat that nothing we say here is

investment advice, legal advice, or even life advice. Please see Chopin Block. That XYZ for more disclosures. Sri Ram, it's great to have you. For those of you don't know, Sri Ram is a longtime professor at UW at the University of Washington. And he's also a crypto founder, founder of one of the hottest new projects called Eigenlayer. Full disclosure, I think robot ventures, you guys are investors in EigenLayer. Dragonfly is not.

But eigenLayer right now is kind of the talk of the town. Like everybody is talking about eigenlare to the point where you got shouted out by Vitalik yesterday. So first question, moving from academia into being now like a full-on crypto founder, center of attention, you know, in the center of the maelstrom. How has that been for you transitioning from the academic world to the crypto world? It's been a crazyness. I think I've, nothing I could have done would have prepared me for this. It is, I think, one thing to be a startup founder where you're basically off and you're quiet startup trying to go build your thing. It's a completely different thing to be a founder of a crypto project, which is much more a community-driven, have start to interact with a lot more people much early on.

in fact, so much so that that is critical or essential to the success of the project even before you build anything. So I learned these lessons through doing a project where it was only technology-focused four years back when Haseeb knows the story with a couple of other professors. And when the feeling was exactly the opposite, it was pretty much similar to my professor life where we were just inventing your protocols, algorithms, and making them. work. And I think this is a very different experience. Maybe you want to give people a bit of your background of how you got to crypto because obviously you have a unique story in terms of like all the different fields you worked in. Yeah, totally. My PhD was in information theory, but particularly we were looking at how to build peer-to-peer wireless systems. You know, wireless usually has infrastructure like

base stations or access points. Can you build a kind of infrastructure for wireless networks, just notes, talk to each other, you know, found connections on their own and multi-hawk and send information across. So that was my PhD, and at the end of which I realized that this was a pretty cool academic project. We ended up solving some theoretical problems and had long-standing problems, but I ended up not having any real impact. I didn't even know what I was expecting that, you know, suddenly we all give up our infrastructure and now move to peer wireless because we were doing clearly theoretical things at that point. But after doing that, I got into computational genomics

because, you know, I was looking for what are some of the most impactful things that we could be working on. And this is around 2011 and 12. And, you know, worked on algorithms in computational genomics over several years, had a bunch of really good collaborators. Several of them went on to win prices like the breakthrough price. But I got a call from my advisor. in 2018 around, hey, you know, this peer-to-peer thing that we were working on, that seems

to be a thing again. You know, there's this thing called Pitcoin. Yeah, I've heard of Bitcoin. And I was very much concerned that previous time we worked on peer-to-peer, it was all like, you know, Ben Nowash. And I didn't want to enter into another thing like that. So it took a long time to actually understand what was going on. Even though it was very good fit for me technically, it was not something that I was.

wanted to enter with or actually having a long-term belief in. And the thing that clicked for me is that crypto is our coordination superhighway, just like the internet is the information super highway. You need to coordinate across individuals. You need mechanisms for long-term memory, long-term enforcement. You need mechanisms for just like, you know, we say Bitcoin's main role is to prevent double spend, right? Civilization is one of the main role is to prevent double speak. I say one thing to you and another thing to somebody else. I say one thing to you today and one thing to you tomorrow. And essentially, crypto is a commitment engine that can actually end for, you know, keep people's commitments. This I see as kind of like a basic evolutionary advantage for humans,

because if we coordinate better together, we can actually build a lot more things together. So that was my core thesis. And once I kind of got into that, I said, okay, no, I'm, I mean, okay to give up working on genomics and work on proteins. So that's all I got in. You know, it's funny because as you were telling that story, I was remembering, you know, when I first got into crypto full time in 2017, I remember so much of the context in which I was taught about all the ideas in crypto were around the history of peer-to-peer technologies, BitTorrent, like, you know, mesh networking, all that stuff was kind of part of the soup that

crypto was born in. And people don't talk about that stuff anymore. Right. It's sort of become, I don't know, sort of pass A to like make comparisons between crypto and BitTorun or something. It was very much a moment in time when this very direct line was drawn between these, this lineage of technologies. But now it's, you know, crypto's kind of become the same thing. People don't really talk about that connection much anymore. What do you think the pivotal things that caused that break were? Because at least to me, it's not so obvious why that happened because, yeah, 2018,

I feel like people were still saying things like that. But I think this last bull market, 2020, kind of was like the clear cutoff point. Like maybe Defi was the thing that like... It feels like the fact that we have applications now is probably the defining

difference is that in 2017 you really had to just tell stories and the stories could be like, oh, well, this is just like, remember peer to profile sharing was so big. This is like the money version of that, that maybe had something to do with it. And I think also it was just that the tent got so much bigger that you had to tell a story that resonated with a much broader set of people, especially young people, right? If you're, you know, 20, there's a good chance you don't

even know what Bitorn is. Like, you just, you're past the generation. You're like, you don't even remember iTunes if you're 20 today. I asked some, some Gen Z crypto person who joined like last year or something if they knew it a Rye Stone was and they just gave me. They gave me this blank look like, are you a fucking alien? Like, what language are you speaking? If you don't know the reference, you should Google Ryestone. Back when I got into crypto, you could not be in crypto without knowing what a Ryestone was. That is probably the single biggest calling card of what generation of crypto you came into.

But that is another thing that's interesting about Cryptostrum is that in crypto, professors are rock stars. like it's a very different kind of thing than in most other technology fields, right? It's not just like, oh, you're very smart and there's an annual, you know, Nobel Prize ceremony thing for the smartest professor. It's more like, no, no, no, no, the biggest professors are just awesome and they command hordes of people at will

and they fight each other and these kind of... That's true and AI too. I don't know where you're talking. That's true. That's true in biology also. That's true in biology also. Okay. You know, actually I have the precisely opposite feeling here.

Is that right? Yes. I think, you know, my own experience is basically that, you know, it doesn't matter. You're a professor or anybody else. It almost is a negative credential to be a professor, so I completely don't play it. Because, you know, in fact, I got asked in another interview where they said, oh, you remember 2018, like there was this whole generation of prof coins. You know, first I thought profs coins must be like something about to do.

And then you remembered Algarand, right? Oof, oof, come on. Hey, above the belt. Yeah, so, you know, the, and then I, my reaction to that is actually that professors build infrastructure. And infrastructure, unfortunately, needed to go with creating both new, ecosystems and new tokens of value. Otherwise, you couldn't build infrastructure. And I think particularly what I'm addressing in our project is basically how to let professors or innovators in general build new things without having to build cross coins. So that is a short summary

of what I can lay this. We talked about this a little bit last time when we talked about Bitcoin and religiosity is that every new project in crypto is part technology, part movement. and you kind of have to build both. You have to build the movement and you have to build the technology, which is not always necessarily done in the same person, but it's most effective when it is in the same person. So you're getting into now the movement building part of your career.

That is the crazy part, yeah. Yes, yes. Well, you've done remarkably well at it so far because everybody is talking about eigen-Layer. So a lot of people who are listening, they probably don't know what Agen-Layer is. Maybe they've heard of it. They have a vague idea what it is.

And so I wanted to do an exercise. you know those videos where it's like those BuzzFeed videos or something where it's like, you know, this physicist explains quantum physics and five levels of complexity. So we're going to do that. And so we're going to go around and we're going to explain EigenLayer to a five-year-old, to a high schooler, to a day trader, to a developer, and then we're going to have you explain it to a crypto professor. And we're going to try to keep these relatively short just so that you sort of build up in understanding.

So I'll start how I would explain EigenLayer to a five-year-old. So if I was talking to a five-year-old, I would say on Ethereum, we have policemen who protect the town. And those policemen, they only protect the government buildings because they're only allowed to protect the government buildings. But with Eigen-Lear, they can use the same badges to protect all the stores in the town. And that's why Eigen-Layer is so useful. All right. That's my explanation to a five-year-old. When's your children's book coming out? I'm working on it. I'm working on it right now. Please don't leak anything.

I just bought my niece a book by the Stanford professor called Algorithms for Babies. So I'm looking forward to comparing it. Oh, Dr. Routen for babies from Hafib coming in. Wait, Tarun, do you have an announcement to make? No, no, no, no, no. It's for a niece, right? Yeah, yeah, my niece. Okay, okay, of course, of course, of course.

All right, Robert, explain to a high schooler. Okay. If you remember back in the days of your, when there was this concept called merged mining for all of the different SHA-256 cryptocurrencies that came out, people were mining for Bitcoin, and while they were mining for Bitcoin, they were taking the same hashes, and they were mining all of the other Bitcoin clones.

They were mining light coin and feather coin and purecoin and all these things that don't really exist anymore. And the same concept applies for all of the new generations of blockchains, which is while you're validating Ethereum, you can also be validating new L2s and optimistic roll-ups and sequencer-based things and all these other additional blockchains using the same validation resources.

Pretty good. It's a very advanced high schooler. That high schooler knows a lot about merge mining. Merge mining, yeah. It's a high schooler who's in the script. High schooler knows about merge mining and feather forking but doesn't know about the Rye Stone. Yeah, we might be hiring if you're a high schooler

and you know about feather forking and feather corks. and feather coin. All right, Tom, explain it to a day trader. I'm going to try to work some some coinage kind of stuff in it. All right. There's this chicken and egg problem when you want to start your new coin. If you want to start Tron or Tom chain or whatever, where if you're using proof stake or some sort of stake-based consensus, you want your token, your coin to be valuable, but it's not going to be valuable until people are transacting on it, but people aren't going to transact on it until it's safe and you have some sort of weight securing the network. What if instead, you could use something like Ethereum, which is already live and mature and has a lot

of stake behind it to bootstrap your own network or maybe even replace some of this coinage. And that's sort of the idea with eigenlayer. Okay. Terun, explain it to a dev, median dev. Yeah. Suppose you're a developer. You've made a smart contract before you've deployed it. The smart contract acts sort of as like you have this sort of like magical service in the cloud

that is Ethereum that runs your contract and ensures that that the smart contract. and ensures that the state transitions that you want to have happen, happen, or the ones are forbidden don't happen. The problem is not all contracts work exactly the same in the Ethereum environment than on an environment you completely control. And that's because, you know, the validators who are putting up sort of some stake have some extra rights. They can add transactions. They can reorder transactions. They can remove transactions. They have some extra rights that don't guarantee you exactly what you want. Now, in order for you to do that, you would have to run your own infrastructure,

create a new L1 that you had to run nodes for, and then have those rules enforced in the consensus for those nodes. But suppose you don't want to have a new L1, but you have an application whose security relies on these guarantees that have to be enforced at the validator level. Eigenler is a way for you to have economic guarantees that the validators enforce these covenants that are stronger than the ones that your contract itself enforces by giving the validators extra rules,

and they get economic rewards if they kind of enforce your rules, and they get slashed or have penalties if they don't. Okay, now, Sriram, can you explain this to an academic, a very crypto-knowledgeable academic colleague? Yeah, at the root of everything we do in a blockchain, whether we think of something as a crypto-solution or not, I think the boundary, decision boundary, is whether it's based on decentralized trust or not based on decentralized trust. So you can think of decentralized trust as the raw material of, like, everything that we build in crypto. And what blockchains like Ethereum do is to take this decentralized trust and refine them into particular products. For example, block space.

So you take decentralized trust, you know, you have a group of nodes. They all work together. They arrive at consensus. they run a particular scaling and a particular execution environment and everything, and then out comes a block in every block that is some amount of space. And what happens in typical applications that build on smart contract platforms is they consume block space, which is they're consuming this refined version of decent electricity and then paying for it.

So the economy is, if I'm an application and I want to borrow trust from, say, Ethereum, I'm basically buying block space or my users applying. block space and therefore paying for the decentralized trust that is being shared. But this trust, you know, the decennialized trust underpaining Ethereum has been refined in particular ways, using a particular concern in this protocol, using a particular execution engine and so on, using a particular block limit, all of these together give rise to the particular economy of the block space. What is instead we can go all the way down and get you the raw decentralized trust and create a marketplace for it.

So what do I mean by raw decentralized trust? Decentalized trust emerges out of the combination of really two components in most platforms. One is staking, which is I'm putting down a bunch of capital. Note validators put down a bunch of capital and make a credible commitment that they are going to run this Ethereum block production service according to the rules and the covenants laid out in the protocol. And the second aspect of decentralized trust is decentralization itself, which is that not only I'm putting down a bunch of one guy put down $1 billion is still different from thousand people

put down $1 million each. So in the second attribute of decentralized trust is decentralization, which is how many different nodes out there, how many different perspectives have been integrated in coming up with this block production. So eigenlayer is a mechanism to allow. anybody to consume raw decent-less trust. So imagine, instead of having the ability to program Ethereum at the level of I can write an EVM contract, what if we give you a hook and say that you can go and tell each Ethereum note what particular node software they should run? Very low-level

programmability. So you have full control of programming the entire distributed system. So eigenlayer is a marketplace for decentralized trust. It allows you, allows validators, node validators on Ethereum to opt in and use their same stake. They already put down 32Eath and then committed to validating Ethereum blocks. They take the same 302Eth

and then say, oh, I am going to now take this 32Eath and I'm making a covenant or a commitment not only to validate Ethereum blocks correctly, but also to validate a new service, you know, Tom Chain correctly, or like Hasib's Oracle correctly, or Roberts data storage service correctly. So I can make all these governance, and since I am the one who is actually running the node software

for all of these different things, I can hold to all of them equally well because I know I'm running this, I'm also going to run that in the proper way, so I won't break any of the commonance of these many different software. So essentially, eigenlayer is a marketplace where we let the trust suppliers,

which is nodes who have put down stake, stakers, right? meet trust consumers, which is, for example, distributed service builders. Like you want to go build a new bridge, a new Oracle, a new chain, a new service, which requires N-nodes to come together to consensus and have some amount of economic security backing it. Now, anybody who wants to build any of these services, instead of going and wanting their own new decentralized trust ecosystem, can just borrow aspects of decent-less trust from this common sense. frame but that off the age.

So, Suram, we were having a conversation a few weeks ago, and the obvious objection, which led to a very interesting conversation between us, was that, okay, the idea of restaking, it's very beautiful, it's very elegant, you know, it's very efficient. This idea that instead of having 5,000 tokens, there's an Oracle token, there's this token, and that token, and all those tokens are all getting staked separately. Instead, you have sort of one group of stakers who are the most trustworthy, and they have a lot a stake and they're all using the same stake, sort of re-hypothecating across multiple protocols. The obvious objection to that is like, well, look, we're VCs, right? Four of us here, we invest

in protocols. And if you are starting a project, it's really hard, like, what do you raise money for if you don't have a token? If you're like, well, I'm going to pay Ethereum stakers to do this, and then, like, my thing will just, like, I don't know, collect a toll on 101 every time you use it. it's like, what does the token? How do they raise money? Like, where's the shit coin? Like, you kind of need this shit coin economy, don't you, in order for this stuff to work? What is your objection to that objection?

Yes. My objection is that it doesn't stand empirical evaluation. Why do I say that? Like, you go back to 2014 and 2015 and say that, hey, you know, you could go raise money and instead of all apps starting to congregate around. like a platform like Ethereum, they could all go and do the same thing and start their own chains. And we, you know, we talk about DAP chains today, but DAP chains were the only thing that were possible before Ethereum. Like, you just had a bunch of DAP chains.

And Ethereum, by creating a shared security substrate, by reducing the actual cost of security, actually made it more viable for all of these different services to be built. They were all built on a par security standard, like everybody's sharing the security, which means they inherit several aspects of composability. And all of these effects basically made more and more people build on Ethereum till to some extent people realize that, oh, Ethereum doesn't have block space anymore for me to build new things. That's the only reason you could go start a new L1 is this story has run its course.

Oh, you know, there's no more block space to build on top of Ethereum. And then you say, oh, no, I'm going to create a new blockchain. It's doing the same thing. but now like produces block space in abundance, and then go and create like new layer ones, like, you know, all the other layer ones that we know of. But to me, the logic of aggregation is much more powerful than the logic of segregation.

So, but to actually answer your question in more practical terms, we have thousands of taps that you guys have funded. We have maybe like 10 layer ones that you guys have funded. Like, you know, why is that? That's because it's much easier for a data, to establish its use case and value, and there are many more daps that are likely to be actually viable than there are like layer ones that are likely to be viable. Most layer ones, like our own, my own, previous startup called trifecta, which which was

dead on arrival, right? It's dead on arrival, not because that technology, it's dead on arrival because the bar to have a layer one is pretty high. It's pretty, pretty steep chart to both that. So another analogy that I'd like to draw on here is what happened in the internet. In 94, if you want to go build your own like, you know, internet web application, you need to build your own server, you need to build your own identity stack, you need to build your own payment stack, you need to build your own DB, and then you build whatever application

you were building. That's how, like, all the old applications were like hotmail or whatever. But in 2023, if you want to go build your application, you go, you go. use AWS, you go use OAT for authorization, you use like a stripe for payments, you use some other DB for, just tie all these together. And then you're really focusing on the thing that you really value. And if you look at it, if you think of things like middleware, nobody really made money on middleware, but like all the VCs over the last 20 years made money on SaaS. Why? Because SaaS is SaaS services that built on top of AWS were highly profitable because you can focus on one

narrow thing and kill it. And it's much, much easier to predict whether somebody can focus on a narrow slice and kill it than it is to figure out if they've solved the whole monolith from like top to bottom. So I think the modularity that is that eigenlayer empowers basically makes it feasible for many people to come in and focus on narrow slices and kill it. Like I come in and come up with a secret chatting. That's just a secret chatting layer. Somebody else comes comes up with a data layer, somebody comes up with an Oracle layer, somebody comes up with an authentication layer. Now if you're building an application, just tie all these together. And so as a VC, it's much easier to calibrate on sharp focus bets. This is the best

team to build the secret challenge layer. This is the best team to build the Oracle layer. And DAP tokens exist in plenty, and they are also valuable because they provide valuable services. And so that's the same thing I would say. Just because their tokens use for security doesn't mean it's more valuable. And also, if you build two systems, one system has a much higher cost basis, and another system actually has a much lower cost basis. Systems, the lower cost paces just wins in a free market. You can't wish it away. The existence of a shared security system means that the cost of security has gone down. And that means that systems building on top of this leverage an economy of scale that's not available to systems that have to emit a 20% emission

annually to keep their security going. So that all those factors together lean towards actually building an aggregated systems. And we are very friendly to tokens. Like you can design very complex token incentives, even inside, you know, something like eigenware, you can have the tokens absorb a portion of the fees that go into your system. You can have your token be used as the payment token. You can be, you can use your token as an artisional staking token in addition to Ethereum, giving rise to dual staking models.

So you can build all kinds of complex things. And it really depends on the applications in your case rather than a one-size-fits-all. Yeah, I guess another tiny thing I'd add is, you know, if we look at the last five years, one of the things we've seen is sort of like the unbundling of how transactions are propagated and how transactions move between chains.

And that unbundling is like a one-way function, right? It's like second law of thermodynamics of blockchains, of modularity. There's always some extra complexity that comes from hyper specialization. And in some ways, bootstrapping that specialization, as these things get more complicated from nothing, is much harder than trying to build it on the existing network. I mean, the simplest example is comparing value extracted on Ethereum versus value extracted everywhere else. and how that entire supply chain works. It's just actually easier to do it at the place

where there's the most liquidity and the most volume versus trying to restart. And you could argue that, you know, name coin failed, but ENS succeeded, right? This is like kind of in the same vein, like a bunch of L1s may fail, but the roll-ups that proceed may succeed. It's not that like all L-1s have failed,

But I think there's a lot of L1s whose value prop over Ethereum roll-ups is just a lot lower. One of the things that was an aha moment for me in our conversations for around was this idea that you alluded to again that, look, the sort of 100 IQ take is, ah, if I'm a founder and I use restaked ETH instead of using my own token to secure my chain, then I can't really have a token. My token has less utility. And therefore, you know, I'm going to get punished by the market because my token is going to be. be not very useful. And I think in a way, the market is kind of conditioned right now to think that way. Look, if this token is not used for security, then what the hell is the point of this token? Throw it away. You know, this thing's not worth anything. But that, it's kind of a nonsensical view because ultimately, in a world where you use restate teeth instead of using your own token,

like your own token can still be the governance token that ultimately governs the protocol and in some way is, you know, attributable, whatever, you know, boons the protocol ultimately accrues, go to the the token holders in one way or another, if it allows you to reward those stakers with less marginal inflation than you would otherwise need, like at the end of the day,

the only thing that really matters for the value of your protocol is how valuable it is in terms of how much money you can make, how much rake you can actually take in as a protocol, and then, too, how much you're paying for it, like how much emissions you're getting out. And like those two numbers

are really the only numbers that matter, which is sort of the, you know, the cost of administering the service and the price that you receive for giving the service. You know, if you want to really dumb it down at the lowest level. And ultimately, paying for security

with re-hypothicated, less volatile collateral, like ether, as opposed to paying for security with one single-use collateral that you invented that's super volatile, you're always going to have to pay more in super-volatile, low-liquidity collateral, which is, I think, to me,

that feels like the most robust, non-obvious argument for why it is that this is a better equilibrium for how people buy security. Right now, with tokens, people are bundling, I want to buy security and I want to govern my protocol using the same token. But you don't need to bundle them. There's no reason in principle why they should be bundled just because they

are bundled for other layer ones and for many other protocols. But at the application layer, I feel like if eigen layer succeeds at this mission, it's going to do so by breaking that link in the mind of honestly many traders. This idea that oh, well, that's what gives a token value

which is that it's used for stinking, which doesn't really make sense. Yeah, I think we're already starting to see this with the layer two's commanding some value, right? Like the layer two tokens have, you know, are not necessarily the security tokens and don't even have a path to be staking tokens for some of these, but still people have started, I think there's a flippening in the mind or the narrative, which is that, okay, you know, these tokens have value because these systems create value and there are some fees that gets attributed eventually back to these token holders.

So I completely agree with that. In fact, that was the dominant way in which we ended up with the protocols. In fact, there is that other like perverse thing with that bundling that you talked about, Hasid, which is that the total amount of security you can buy is bounded by your token market cap. Like, what is this thing? This is a very weird thing. This is the total amount of, if you stake your own token for security, the total economic security you can buy is bounded by your token's market cap.

Those two things should have no relationship. There's no web two company which cannot serve a bigger web two company, because there is no relationship between your company's market cap or value or whatever and who you can serve. But when you rely on things like economic security, the smaller you are, the less people you can serve because you don't have much security. And kind of like it's like shooting yourself in the foot on day one when you're starting a protocol.

you might as well like if you're you know if you're successful in building a protocol even with restaking eventually you may decide not to use restaking because of whatever reason and you can go ahead and be your own like layer one or like just your own security token or whatever it's like these options are all available if you know the marginal advantage of using a lower cost of security is not dominant at that point but i think for most protocols there is a kind of like a bootstabbing problem that Tom alluded to earlier, which is that you can't even get started. Because if your token market cap is like 100 million, like your economic security is almost 100 million, if everybody stakes, this seems to be like a very bad association. And DAPs have clearly avoided this.

And I was just thinking about like, how did DAPs avoid this fate? Like DAPs have the entire Ethereum's economic security, even in their market cap is zero, right? Uni systems not as secure as the unique token. And that is roughly what we want to emulate for all these other systems. So speaking of security, that brings us to the post that I was alluding to earlier that Vatallic wrote earlier this week.

So the title of the post, and I'd love to just get everybody's reaction to it, the title of the post is don't overload Ethereum's consensus. So there's a few kind of big themes in the post, but the main thrust of it is basically to say there's a lot of experiments that are taking place of trying to leverage Ethereum security model and Ethereum security through all of its stakers,

but also some implicit assumptions about Ethereum's social consensus, meaning that when things go wrong, is Ethereum potentially going to fork? Is Ethereum going to favor one thing or another? Is it going to do a soft fork? Is it reasonable to expect Ethereum to do,

to bend in certain ways in order to protect its overall ecosystem or protect individual players? Or even if you're doing something that's not directly affecting the normal operations of Ethereum, applying certain kinds of pressure to stakers and making it so that stakers

should be a little bit more likely to do X or Y or just even be thinking in a certain direction more may be ultimately deleterious to the protocol being neutral and being unopinionated about the applications

that are living underneath it. So where are you guys' reactions reading this post? It got a lot of attention, a lot of conversation going before I jump over to Sriram. Well, my super, super high-level take was that I think the post

only focuses on the risk and not the potential benefits. The risk being, hey, this is a total distraction for Ethereum validators. And if you, I'll use this word, corrupt them with outside economic incentives, it could maybe mess up the economics of Ethereum at like the core foundation without saying, hey, if this supercharges Ethereum validators, isn't that a really good thing for Ethereum? if there's even more use where like this is the like super collateral. Like, you know, so I think it just focused on risk and downside and not. I feel, okay, let me give you a sort of counter analogy is like,

somebody's like, yeah, okay, you might be worried about money in politics. And that sounds really negative. But what if you gave so much money to politicians that all the best people became politicians, maybe it's look on the upside. I think what Vitalik's pointing to is like the risk of corrupting validators. it's like valid something like you know a theory is public good

if we throw too much money at it it's kind of like money in politics right it's sort of corrupting by its nature now you can't stop it right like how do you stop it I don't really know other than just like writing you know kind of upbraiding posts like this yeah it's um

I think it like in Singapore they have some of the highest paid sort of public servants kind of for this reason I mean they're not politicians but it's like you track really great talent and it sort of reminds me of like the Trump quote like I'm too rich, they can't bribe me. I think there was like some kind of merit in specific applications,

but like, I don't know, I think a lot of it, the post setter focuses on, on consensus. And overall, I think I agree with the design philosophy of like keeping core consensus very simple. I think a lot of other chains have experimented with sort of adding a lot more at the base layer. But at the same time, I guess I sort of zoom out and think about more like application sort of social consensus capture. And I mean, you know, CB had this post from a while ago around Ethereum being unforkable because of defy and specifically talking about USDC. And I think that in my mind feels kind of similar here where it's like people don't want to use Ethereum.

They want to use the applications that are on Ethereum. And if there are fundamental problems with the applications that are on Ethereum, of course social consensus is going to direct it to a new fork. Like if there's a vulnerability in uniswap and all the value is drained, I think people are going to move to the fork of Ethereum where the bug is fixed and, you know, the bug is fixed. And, you know, MewSwap's not drained. And so I, like, I empathize a little bit of sort of the, you know, sort of design philosophy, point of view here.

But like, I don't think it's, it's like, it feels kind of like a, again, like a moot point. Like people are going to go to the chain where there is stuff that they want to use. And so if that stuff that they want to use and is going to use a theory and consensus, like, that is kind of the end of the show, I think. I think in some ways. The theorem already made sort of the Faustian bargain by moving to proof of stake in the sense that you can't get a lot of the same risk guarantees anyway. You know, there's this availability tradeoff. There's kind of the idea that people can't join and leave as easily as you can in proof of work.

This is the idea that geographic distribution is much worse. There's like a ton of known tradeoffs. Those tradeoffs probably have higher economic value than a lot of the applications. level risk in my mind. And you already made those decisions. So from my perspective, it's like, well, you chose a system that allows this extra functionality, takes on this other risk.

You can't really stop it from happening. In the same way that, you know, the last time there was a don't overload consensus style post, it was like, hey, we're never going to allow delegation in Ethereum. Lo and behold, what happened was Lido. So I just think the inevitable hand of capitalism and market forces will inevitably drive these things into existence, whether decided or not, you know, desirable or not, because validators are not really honest. They are rational. And those rational validers want more money. That feels, okay. That feels very fatalistic, to ruin. And I don't, I don't totally agree with it because I think, let's use the analogy of like money in politics. I think that actually is a relatively, good analogy because it is in a total free market, you would just be able to bribe politicians

to do whatever you wanted, right? And like, people talk all the time about bribery attacks being one of the fundamental issues in any kind of democratic system that has to be overcome in one way or another through, you know, anti-bribery mechanisms, such as strong laws against bribery. In crypto, obviously we are in a permissionless system or, you know, whatever, modulo, more or less permissionless. And that sort of means that you can take the view of like, okay, well, then, because permissionless, you know, whatever, there's going to be bribery attacks. And there's going to be, you know, people just, you know, making validators do, there's going to be feather forking. There's going to be all this other stuff. For the most part, we actually, we don't see a lot of

these behaviors. And I think the reason why we don't see a lot of these behaviors is twofold. One is that a lot of the people who are actually running the big validators on Ethereum, they're not purely market oriented. Like, they are actually relatively ideologically aligned with Ethereum. Just because, look, if you are a total ruthless capitalist, you're just not going to, I don't know, it's like a weird line of business to get into. You know, it's not the most profitable thing in the world to run Ethereum validation services.

You know, it's like kind of a weird, nerdy business to get into. Maybe it's a little bit more true in the proof of work mining days. But in the world of proof of stake validation, you know, there's no like random guy who just happens to have a, you know, retired power plant next to your house. Like, you're in the same footing as everybody else since you just love running validators. and those kinds of people don't want to watch the world world world. On the other hand, on the other hand, the reason you have staking derivatives,

and MEV is redistributed to many people who could never capture it themselves, is because the ruthless nature of this led to the securitization, that securitization led less sophisticated actors gain access to those returns for a small fee. Totally. And I'm saying that this is just the same March we're going in the same direction.

The question is not is it marching in that direction. Totally agree with you that all markets tend toward efficiency over time. But the difference is in what way are the institutions

behind Ethereum and the culture that is enshrined in Ethereum in what way are these things modulated by the institutions that are held up by Ethereum? And Vitalik, of course,

has a large role in that because he's like kind of the religious leader or whatever forefather of Ethereum. And I think the, like what he's doing in this post

is not just saying like, guys, please don't be totally self-interested and, you know, but I know you're going to anyway. It's not just like a, it's not like saying, hey, guys, please don't extract any V. Like, if you do that, it's really going to hurt Ethereum. Like, why are you hurting me? That's not, I feel like that's, it's very different than what he's doing. What he's doing is he is advocating for a certain set of institutions. And that set of institutions is enforced collectively by almost everybody. Like, I think it's very clear that Ethereum does have institutions.

Right. And these institutions have been conveyed through both leaders like Vitalik, but also just like the kind of ever-present culture that is everywhere in Ethereum that like even random people at DevCon, you will find them and they'll just have these beliefs and they'll be walking around with them, even though they also want to make money and they also are self-interested in different ways. and the maintenance of those institutions is really subtle and complex, and I feel like that's what this piece is ultimately about. If it was really just, hey, you know, don't extract any V because like it, you know, that makes you an asshole, then it wouldn't work. And I would say, okay, well, this is doomed. This is obviously just a futile attempt to delay the inevitable. But I wouldn't say that that's true of what Vitalik is doing here.

I'm not saying that the post said that as much as like there was a little bit of like a tis-tis-tis-ness to it, right? They had this tone of like a like little like, you shouldn't have done that, you know? I find, I find that to be the part that I'm like, well, this happened before. Like, I don't know what you're, you know, fool me once, I'm the idiot, pull me twice, maybe you're the one. Let's bring you, Surreal, what is your take on this whole thing? Okay, I have somewhat different take. Maybe it integrates all of your things and adds a different dimension to it. I think the first thing is, the title of the post.

The title of the post was do not overload Ethereum consensus, but actually, if you read the content, it is do not overload Ethereum social consensus. It's do not overload Ethereum social consensus. And he's basically trying to do what, you know, Haseeb was finding out, I think, in my view, is basically saying to take the government type analogy. What he's saying is,

don't build a bank, take too much risk, and then come to the government's for a bailout. That's what he's saying. You go do your thing. You go do build restaking. You go build crazy things. Don't rely on Ethereum is going to fork and save your ass if something is not. Basically, that's the selling point. He's trying to kind of get everybody to agree on. You know, that may be a big project. There may be uniswap. There may be comfort. Maybe eigenlayer is big. Doesn't matter. if you get screwed on a different layer, go deal with that layer. We're not forking Ethereum 5.

That's the boundary. The boundary is basically what is the core concern of the protocol. So to answer like Haseeb's analogy of politicians and bribing politicians. Like you can pay a politician to write computer programs in addition to whatever they're doing in their office. So the concern of the government is to make sure that when they're in office, they do the thing that they're supposed to do. And that's why we are slashing at the Ethereum Corlear to protect it. And the slashing at the Ethereum Correlator protects against safety falls and protects against

liveness falls. And safety falls are like programmatic, right? You're assigned two blocks. You're going to get slashed. It's programmatic. But liveness falls are non-programmatic. And that needs the use of social consensus. He's basically saying we're reserving social consensus for.

for the one fault that is at the code of the protocol. Not for all the other things that you may want to use it for. And I think it's a really good, really, really good boundary setting, which is what are the concerns of the protocol and what is the concerns of the free market? Consensus of the free market and don't externalize like risk from the free market into the core protocol, right? That is basically what I think the article is trying to say. And I think the shelling point or the equilibrium that it's trying to set is when somebody comes and claims that, oh, we are so big that if we go wrong, like, Ethereum's going to come for us, that's not going to happen.

Like, I think it's a really good, good demarcation of boundaries. And if you look at the article in read carefully, like he's explaining all the things that are low risk, things like staking for a new chain, he's saying it's low risk, right? specifically called our like take Dogecoin, run it and validate it. So the way we view it, in fact, we have internalized all of these inputs for the last like 15 months. The way we build our protocol is already assuming that this is how things are going to be. We cannot rely on Ethereum's like social consensus. So all mediations happen in protocol and hangar there. And this made us like, you know, think much more carefully about the various risks and how we handle it all in our layer.

we can't say, like, throw our hands up and say, like, Ethereum's going to handle it for us. And I think it's really good that he wrote this up because now, like, when other people look at restaking, they're like, oh, it's such a simple idea. Like, we know, it's as simple as just take a liquid staking token, put it in, forked salana and build it on Ethereum, right? And we didn't do this. Like, we could have done this like one year back. We took a lot of efforts in minimizing all kinds of risks and externalities to Ethereum.

We have to worry about what are the slashing risks and how it affects validated. We have to worry about what are the centralization risks to Ethereum. How do we minimize that? We have to worry about what kinds of slashing can we allow on protocols, how to make these slashings all objective and attributable rather than subjective and, you know, let Ethereum deal with it. So all of these things have been internalized into the construction of the eigenlayer protocol and how we approach the services that have been built on top of eigenlayer.

but somebody else from the outside may not know or understand that these are the boundaries that we've tried to maintain. But so that's why I was actually quite happy to see that kind of a boundary setting as to what is the concern of the protocol, what risks have to be internalized into the systems that are being built, whether they're externalized into broader things. So I love that analogy, especially given how timely it is, given all the stuff going on of the banking crisis of this idea of like, hey, it's almost in a way a self-fulfilling prophecy that if you build something and you assume that Ethereum is going to bail you out, right?

If you sort of, you get this culture of people thinking in their minds that, ah, we're going to get bailed out if something goes wrong, then you get collective risk taking, right? You get basically this sort of moral hazard where everybody takes on lots of lots of risk. And then when things blow up, the pressure to try to bail something out becomes really strong, right? If anything, it becomes almost self-fulfilling that people kind of expect that this thing will get bailed because everyone's been saying that it's going to get bailed out. And so in some way, you need to sort of stop that spiral before it begins and say,

hey, guys, just so you know, we're not going to bail any of this shit out. If this thing fails, you guys are on your own and you're fucked. Now, that being said, you can say that all day long. If optimism gets hacked, or if, let's say, God forbid, Uniswap gets hacked. All the money in Uniswap gets drained. Like, I think there's a good chance that Ethereum would be. Dude, 100%. 100%.

But I think that's exactly. But it does serve the purpose that you exactly said. which does not encourage excessive risk-taking. This is basically, you know, by setting the selling point, you know, is basically driving people towards be more conservative. And then, like, of course, something happens. Then we figure out what to do about it, right? But you can't go with the a priori assumption that everybody's going to get bailed out.

So you go do your fun stuff. Although, I think there was even some more stuff written that says even that might not have, might not happen or at least intimated that. when Vatalek talked about the difference between Tycho, which is like a small L2 and optimism arbitram, and like with the argument that like, hey, the developers and optimism arbitram overlap more with the Ethereum core developers than Tyco.

So they will, in theory, have this ability to kind of cause this, their fork, their preferred fork. But, you know, we have to resist that. I thought there was a very distinct thing that, even in this. He did say that, right? That's the thing. It's like, you know,

if you're Janet Yelling, you've got to go up and say, we're not going to bail out any banks before the bank failure start. But then once the bank failure started, he was like, all right, let's go, fuck it, let's bail out the banks.

You know, you know, I wouldn't compare the two of them. Well, okay, so, Turner, do you think that if optimism was hacked, all the money was drained,

that Ethereum would fork? Where would you place the odds at? 55-45-45. It's more likely than I would fork. No, more likely, would fork, but like not by much. I think it would actually be quite a contentious issue.

I don't think it's like a hundred percent chance thing at all. I think the best historical context for this is the Dow. Right. The Dow happened to be roughly the only major application on Ethereum when it was hacked. So it's not like it was like one of many. it was roughly the only actual application with like substantial assets. And it was consequential enough that it was the first and only crisis response hard fork for Ethereum. And now we have Ethereum Classic and it's years in historical rearview mirror.

I actually think if something like optimism were hacked or a different L2, right, I don't know if it's as consequential as the Dow, to be honest. You know, the Dow was one of one. It's not one of five. And I just think like the total impact is less. It would be a brutal, brutal, brutal, brutal learning experience. But like I actually don't really think Ethereum would fork. I think it's like a 3% chance.

Really. And obviously it's all, you know, spitball. But I don't think the magnitude of the event would be as serious as when the Dow was hacked. I think that's definitely true. A interesting variation of the question is whether it will fork for Lido. Just in terms of percentage TVL of like, you know, what is going on, I think, you know, the DAO was 15% of all EAP, right? Like that is the level. I don't think we have any applications kind of reaching that.

Even something like Lido is not yet at like 15% of ETH. But it is, you know, I think that's the number to look at is what. percentage of ETH is basically, you know, at risk. And whether there was any kind of soft or other like assurances that, you know, we are behind this, right? And I think... And so here's the thing. Here's the thing. Here's the thing. Going back to the idea of institutions and politics, I think in a way, Ethereum would actually be kind of glad to see Lido get hacked. There's like, aha, we told you guys. You shouldn't have, like, this is why you shouldn't play with fire. We didn't build this in the protocol for a reason.

Don't overload consensus. Don't overload consensus. Exactly. Don't start getting too confident in your little play things that you build on the beach. You know, like what it represents for the Ethereum core developers, or just not the, not the core developers, the intelligentsia, let's say of Ethereum. What it represents is that Lido is a financialized protocol. Like Lido is basically like these guys, like Turin said, look, it's the market forces, you try to get yield, MIVB, blah, blah, blah. We want, we want state ETH to be tradable.

I'm saying this in a positive way, right? They've literally democratized access to MEV returns. Sure, sure, sure. Like you as a random Ethereum holder had no way of even realizing them. Right, other than staking, which is what they want you to do, right? They wanted you to staking. Well, no, no, no, but you needed a kind of use a staking derivative because you yourself, even if you ran a home staker, very unlikely would have been a good flashbots auction

participant, right? Like my point is they did actually serve this democratizing yield thing, which is important. But my point is, if that exists, the market will. find a way to make it happen. I just don't believe that anything other than that will be the outcome. But let me just finish my point is that for Lido, Lido is I think somewhat ancillary to Ethereum's goals. In a way, it's almost like a little bit of a thorn in their side. Whereas the L2s are the story. Like that is the point. We are going to scale Ethereum through L2s and these are the anointed ones, but basically optimism and arbitrar, mostly optimism is anointed because, you know,

it's all ex-EF people. And if that goes under, it's not just that like, oh, there's this amount of money that was lost. I think it actually would be less about the overall number, like the amount of ETH, because from the perspective of LIDO is like, oh, you took a 30% haircut, oh, so what? You know, they'll boo-hoo. You shouldn't have been fucking around with DeRotos, you know? Whereas with the layer two, it's like, oh, I thought this was trustless. There's like all sorts of things on the L2, not just ether. The kinds of stories would just be so much worse, I feel like if an L2 exploded. And that's why I think it'd be different, even though in monetary terms would be Yeah, I think I kind of disagree. I think an L2 is effectively like.

a bridge and we've had really big bridge hacks and we have not done a fork. And so like, you know, whatever. Whereas Lido, 30% of all eth, you could cause some serious mayhem with consensus and overall chain stability. Again, I do feel like this is kind of like young manials at cloud. Like, you know, it doesn't actually matter because let's say Lido is hacked. Someone is going to try to do a fork that reverts the hack. The question then is, well, which chain are us DC and USDT going to accept withdrawals on or that's application value capture

and so you know no one really cares what the Vitalik thinks it's like who can basically anoint the chain and there's other actors there's application level actors that can do that. Or it's your point. No, no, I know, I know, I know, I know.

No, no, it's bringing around full circle. It's very true. It's very true. So again, I guess the reality is that we're all under the cult of Jeremy Aller whether we know it or not. he's ultimately, he's behind the scenes pulling the levers to decide. End of the day, I think, I think Vatolic, like any religious leader, as you pointed out, who's successful, has to write sermons.

This was a good sermon, right, in some sense. I just am more pointing out that I think, I think there have been a lot of platitudes and diatribes about and like on both sides of it, whereas I just feel like it's just saying kind of the obvious thing of like, well, this is probably going to happen. Here's some road rules. don't do the bad ones. Exactly. I mean, that's kind of the role of a sermon, right?

You say something that's already been said. Athop's fables for proof of stake. Totally. The point of it is you've got to say the sermon at the right time, right? It's like the perfect moment to give a sermon, which is the characteristic of a good preacher is not that they innovate on the source text, right? They don't.

They shouldn't. But rather that they pull out the right lesson at the right time that the congregation, that the congregation needs to hear. at that given moment. And I guess this is the time that we're all messing around with consensus. So it's a good time for us to get those rules

of the road right back to us. Wait, wait, Trieram, I have one question I've wanted to ask you, but I figured since now it's recorded, I can make other people have to listen to this answer,

which is why does your logo looks like a, you know, like 2D error correcting code, but it's too correlated. Like, what is this tree thing? Is that forking? Is that something the application?

Can you bring the t-shirt up closer so we can see it? Yeah. How would you describe that? I don't even know how to describe it. It's like an L with like a pixelated Y sticking out of it. No, this is not a, this is a, it's an eigen layer. It's an eigen layer. Oh, obviously.

It's obvious. But what we did is actually this logo goes with how the symbol is written. The word mark is written. And there are many ways to write this wordmark because, you know, we want to just eigen is, you know, in German means your own, so eigen layer is your own layer. You can make many things out of it. So you can write the eigen layer of wordmark. Actually, we have many, many ways of writing the word mark.

And there's an algorithmic relationship between how to write the word mark and how this symbol automatically changes. So we have a family of symbols and they're all like, put there's an underlying principle behind a couple of them. Yeah, yeah. Shat security, modularity, open innovation. Good answer. You got the beginning of religion going.

I like that. That's very important. No, it's like, you know, Christianity there was like 17 different symbols you can use to identify yourself as a Christian. This is Calvin's brainchild. I mean, of course, Adam. You guys are Calvinists. You guys are Calvinists. I see.

Okay. Okay. Speaking of predestination, I think it's time for us to wrap up. But Suram is great pleasure having you on the show. And looking forward to see. It's not live yet, right? When do we get to see Maynet? We're going through a staged mainnet where we are basically going to turn on just taking first,

then we turn on delegation and node operators, and then we turn on new services. So we expect the first turn on to be in the coming months. Okay. Sounds very controlled, very pious. I'm glad to hear that. I'm sure Daddy Vitalik is appreciative of your guys' conservatism. Awesome. All right. Suriram, it was a pleasure. Thanks, everybody. Thanks, everybody. Thank you so much. Thank you so much.