← Unchained
Unchained
The Chopping Block: Ethereum's Inflection Point w/ Joe Lubin on DATs, CROPS, AI-Driven Exploits, Quantum Threats, and CFTC's Perps
01:02:24
Loading summary
Transcript249 lines
- [00:00]
A
The formal verification is the strongest argument for like, okay, there's correct by construction. You know, we have the Swiss cheese and we've like, put the holes in all the Swiss cheese, but it turns out it's multidimensional Swiss cheese because now it's in the third dimension and then eventually outpost mythos in the fourth dimension.
- [00:14]
B
Haseeb, who are you?
- [00:17]
C
Not a dividend.
- [00:18]
B
It's a tale of two Kwan.
- [00:19]
A
Now your losses are on someone else's balance sheet.
- [00:22]
C
Generally speaking, airdrops are kind of pointless anyways.
- [00:24]
A
Unnamed trading firms who are very involved.
- [00:27]
B
Talik Eth is the ultimate DeFi.
- [00:30]
A
Protocols are the antidote to this problem. Hello, everybody. Welcome to Chopping Block. Every couple weeks, the four of us get together and give the industry insider perspective on the crypto topics of the day. So, quick, intro this. First we got Tom the Defi maven and master of memes.
- [00:42]
C
Hello, everyone.
- [00:43]
A
Next we've got Tarun, the Gigabrain and Grand Poobah at Gauntlet. Yo. Joining us today we've got special guest Joe Lubin, Ethereum evangelist and protocol patriarch.
- [00:53]
D
Wow, good name. Thanks for having me.
- [00:55]
A
Thanks for being here. And I'm Aseev, the head hype man. A dragonfly. 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 chopping blocks at XYZ for more disclosures. So we had some Ethereum bearishness on the show last time, and we thought it might be good to counterpoint a little bit with some Ethereum bullishness. Unfortunately, we brought you on on a day where it feels a little hard to be bullish.
- [01:18]
B
Eth BTC is up.
- [01:20]
A
Eth BTC is up. As. As wizard pointed out before the show,
- [01:23]
D
of the majors, ether is less.
- [01:25]
A
Yes. Above the majors, Ethereum is the least bullish. Yes, bullish. Bul for those of you who are doing the pair trade, for anyone else who's just holding eth, it's a painful, painful day. So, okay, tell me. First of all, market seems to be very jittery right now. We've had just a bleed over the last week and a half, and today it legged down 6%. And I think the news that catalyzed this was that Saylor had sold 25 BTC, which is not a lot of bitcoin, and he sold it months ago. But this seems to have been enough to say, oh, my God, the up. Only, like the one buyer in the market seems to be capitulating it seems like that was maybe a shot across the bow just to signal to the market, hey, I might actually sell these, so please buy my derivatives and, you know, please, you know, like, I will, I will pay my debts. But it seems to have backfired. I don't know if maybe this was. This was understood to be the effect, but. Thoughts on the Saylor effect? And also, what does that maybe mean for the, for the eth debts?
- [02:24]
D
Well, he telegraphed it. He said he was going to do this and he did it. And I think you can be a maniacal bull for your favorite token, but you can also be a prudent CEO and do the right things for a particular strategy that you're trying to implement. So maybe a bitcoin DAT is potentially more dangerous because it's not a productive asset in the same way that an Ether dat, Sharplink and BMNR are. So we're not going to get into any sort of strategy that requires us to pay a coupon to essentially leverage the product and have to sell our token. But I also think it's. It's way overblown. They have so much Bitcoin and they had to implement the STRC strategy. There are probably pieces of news on the horizon. There's this rumor that the Trump administration's going to actually do something on a strategic Bitcoin reserve. So I'm not worried at all. I have been hearing not worried at all about Michael Saylor and strategy.
- [03:41]
A
That rumor does not seem to be calming the market at all. And so Saylor owes something like 1.4 billion a year, I think, in cash dividends from stretch from STRC and 1.4 billion a year. I think they have like 700 million cash. So the cash has come up somehow and either comes through them issuing new shares and diluting the existing microstrategy.
- [04:06]
D
And it's a testament to his discipline that he actually did do that little sell telling the market that it's not afraid to implement the strategy.
- [04:13]
A
I think that may have been like, oh, wow, never do that again. That is not the way that you should be recapitalizing. I mean, is that not the lesson that, okay, they own so much Bitcoin? Selling 25 Bitcoin and getting your share price punished by 6% is a very strong. You know, that gets internalized pretty quickly. No.
- [04:30]
C
Or I think just disclose it several months after the fact. You know, it's like, I mean, to your point. Well, I mean, also, you know, purportedly this was like, around testing market liquidity, but I'M like, are you really testing liquidity on like testing liquidity Bitcoin? Yeah, but it's like this isn't actually. Yeah. Meaningfully meaningful. Tangent. It does feel like there are obviously are like macro jitters, but now stock market's new all time high, so hard to sort of blame it totally on Iran at this point.
- [04:54]
A
Yeah, bitcoin was at 77k when he actually sold it. So market felt a lot calmer at the time that he made the sale. Now it's just the disclosure of the sale that has caused markets to just quickly leg down. And I think the fear is like this quote unquote death spiral. I mean we used to talk about this when the DAT phenomenon was really coming into its heyday. Last year. People I feel like, don't talk about this anymore. But all of a sudden I've been seeing my timeline start to get these death spiral fears. I don't know. Tarun, what's your take on death spirals?
- [05:25]
B
I mean, don't you want tokens without a preferred structure? Like if I have 10 layers of preferreds on top of a common stock that I'm dumping constantly, like why even have that as a vehicle to hold tokens? The point of tokens is that they have one share class. And then you're like buying this vehicle where it's like there's 10 different share classes, each of them have different rights to cash flow. It does feel a little like algorithmic stablecoin with too many steps to me.
- [05:53]
A
But what are you, what are you implying? Implying that it explodes?
- [05:56]
B
I'm not saying it.
- [05:58]
A
That sounds like you're.
- [06:00]
B
I'm just saying it's like it. There's a lot of weird nuance to this issuing converts thing. And I find it a little. It's an unstable equilibria. I don't know how you keep it that horse.
- [06:12]
A
And so what happens to unstable equilibria? Yeah, it seems like you're dancing around saying that it's gonna, it's gonna blow up.
- [06:19]
B
I'm just saying that that's not a zero probability event.
- [06:23]
A
What would you, what would you think?
- [06:24]
B
It's an event whose probability seems higher the lower the common stock goes.
- [06:28]
A
Right, right.
- [06:29]
B
That those are the only two things I'm saying.
- [06:31]
A
Okay, okay.
- [06:32]
D
So that said, and I agree with that largely, I still believe that DATs, if you choose the right token, are a really valuable construct for your ecosystem. Our token ether yields approximately 3% risk free yield. We can layer on sleeves on top of that. Virtually all of our Tokens have been staked when we buy them. So it's super solid, no leverage and we represent long term, essentially permanent capital that we can use to grow valuable things in our industry.
- [07:07]
A
Right. Okay, so let's reorient the conversation toward Ether because last time on the show we had some Ethereum apostates and I think we got a lot of anger on Twitter that we didn't have more representation from Ethereum. Bulls. And so Joe, you're on. You're one of the co founders of Ethereum, you're one of the original titans behind how the smart contract ecosystem came together. And of course consensus is the creator of Metamask, which is one of the biggest wallets. And you're also involved with Sharplink, I believe you're chairman, correct? Yeah, chairman of Sharplink, which is one of the biggest Ethernets. So there's been increasing conversation about the state of the Ethereum foundation and that the Ethereum foundation is maybe last year there was this transformation into being a more adoption oriented, a more number go up oriented organization and that there's maybe been some reversion or regression in that mission or in that vibe Shift has undone itself. The pivot was unpivoted and now they're going back to the infinite Gardens, the Aya Miyaguchi, so on. That is represented by the EF mandate, by all the departures from the EF and there's been some back and forth, there hasn't been a lot of clarity. So Aya Miyaguchi recently put out a statement talking about, oh, this mandate was actually written by me. It was in response to hey, it really is true that we are splintering the responsibilities of the ef. That's naturally going to happen. The EF is going to get smaller. There are all these other organizations like the Uniswap foundation and Gitcoin and Moloch, Dao and Consensys to some degree as well that comprise the ecosystem that helps Ethereum succeed. And it's not just us and we mean may be what the ecosystem thinks that we're supposed to be doing that but that's never been what we said we were going to be doing. And so she seems to be affirming that view that hey, the EF mandate is how the EF sees itself and its role. And it's also pretty clear that that's not what the community wants the EF to be. So with the splintering in the ef, how do you, from your various vantage points in the, in the ecosystem, how do you see the present moment for Ethereum and for the Ethereum Foundation.
- [09:14]
D
So I've been speaking for the last few months and quite a lot in the last few weeks with various Ethereum members of the EF or people near the ef. And there was some of the internal messaging wasn't handled well and Bastian has admitted that and he's gone ahead and tried to clean some of that up. But A the Ethereum foundation needed to reduce its budget. B it needed to set it up, set itself up to be long term viable and long term strong. C it needed to do the things that other ecosystems are doing, but do it the right way, do it the natural way. And the natural way is, as Vitalik expressed in his recent post, we should be thinking of the EF as one node, a major node, a node that sets the North Star. Focus on the cypherpunk vision, focus on crops. Crops are components and these things need to be protected and continually developed. There is a lot of reverence for what Tomash brought in within the Ethereum foundation across the ecosystem, the institutional initiatives where there was a lot of reach out, an interface for companies, governments, standards bodies, et cetera, to have the right people to talk to the authoritative people, get linked to the right people, get their questions answered. That was incredibly important. That's not going away. What we see in other ecosystems that don't value credible neutrality quite as much as Ethereum values credible neutrality is this group of biz dev people. There's some pay to play stuff. There's conflicts of interest between the biz dev activities and the researchers and builders. And we can't have, we can't build a credibly neutral platform for the world without having credibly neutral organizations as some of the major nodes in the ecosystem. So yes, the EF is externalizing some of those pieces. We've seen people resign all the why
- [11:26]
A
do you say resign all the state
- [11:28]
D
because they're going somewhere. All the statements have been very positive, pretty much. And so we're going to see the EF focus on crops. We're going to see another body of core, some of the elite core developers focus more on platform scalability, usability, et cetera. Maybe even number go up with respect to the Ether token. That is, do you think second? That is second foundation number go up. Let's not call it the second foundation. There's a set of nodes, stewards in our ecosystem. There's only one foundation, okay. But there will be another group of very important software engineers, researchers, etc. That set up their Own thing with definitely not a bit of positive language from the ef. Maybe even some financial support. There may be some financial support from organizations that have long term permanent capital, for instance.
- [12:30]
A
No kidding.
- [12:31]
D
So we've had a bunch of discussions about that in Sharplink. We've actually been talking to our friends at BMNR about how we together get together to be the more decentralized commercialization engines. And consensus is a big part of this as well. So we want to be what the crypto Bros in other ecosystems are. The Don Wilson doing some great work and Tolle doing great work. I guess he's more on the technical side.
- [13:04]
A
So let me ask though, why so many people have been murmuring for a long time about the concept of a second foundation within Ethereum? I mean this is not new. There's been many years that I remember it must have been like 2023 that. I think Constantine was whispering about this way back in the day.
- [13:22]
D
Zach Cole's been doing it keeps coming
- [13:24]
A
back up as an idea. Why do you recoil from the concept of foundation?
- [13:27]
D
The Ethereum foundation just has a very special place.
- [13:30]
A
So that's why you need to. The second one. Right. What's wrong with the second one?
- [13:34]
D
I think people seem to want to anoint a monarch to lead us to hold the power. And Vitalik is a leader and a steward and just an incredible asset for the ecosystem in the world. But he doesn't want to dictate. He's never operated that way. He's always made suggestions. And we're moving into maybe an intent or goal centric, decentralized world. So lots of organizations can set their intent and systems of solvers will go ahead and satisfy those requests.
- [14:14]
A
You don't say Drun. I see you grinning at yourself.
- [14:17]
B
It's like a funny analogy.
- [14:21]
A
Okay, do you buy this? Do you buy the story?
- [14:24]
D
There are more groups that are organizing and will announce themselves at some point.
- [14:28]
A
Yeah. Okay.
- [14:30]
B
I mean, I do agree there's sort of a weird difference between the Ethereum foundation and other foundations in the sense that like historically it was just like. Is a different. It feels like the first of the. What's the word? Siftung, whatever. The Swiss. It's like the first of that era. It created the whole thing. Everyone else is all a descendant of it. So in some sense, if you're kind of the root node of that whole time, you're sort of different. And any second foundation is going to be like we're the 5 millionth foundation that has been created. Exaggerating. But many so it's sort of, I think if you're the first, you kind of want to keep that separate. Like so I kind of can see that like the idea of like copying the same thing is just like not that appealing.
- [15:21]
A
So I'd say I'm, I mean this all sounds semantic event, you know, like I'm assuming that this thing is gonna be a non profit, in which case you call it second foundation. Okay, that kind of sounds like you're maybe usurping or you're copying the original. I think on some level what the community needs to see is that Ethereum is listening. And so much of what has been, the frustration from so many people in the Ethereum outer orbit is just that, okay, Ethereum just doesn't care. Ethereum sits in an ivory tower. It looks from up high, down below, and it's only really going to respond when it sees the untouchables starting to scale the walls. This is. Oh, okay, crap, we got to do something. Okay, let's tweet that Ethereum says eth is good and then. Okay, that'll quiet them down for a while. Or we'll get the Twitter account to start posting about projects. Okay, that'll quiet them down for a while. And so I think on some level the second foundation is just like, it's become spiritually important to people. This idea that one, the original foundation will eventually go away. That's explicitly part of their.
- [16:22]
D
But they're not saying that anymore.
- [16:25]
A
That the original foundation will go away. They're not saying that anymore.
- [16:28]
D
No, they're saving their money for the long term.
- [16:30]
A
I mean, long term is like not obviously not infinite. Right? I mean maybe they, they'll last.
- [16:35]
D
If either price goes up, it's infinite.
- [16:37]
A
That's true for any organization. If the underlying assets go up.
- [16:40]
C
Not an explicit goal though, Right?
- [16:42]
D
So it's like it's, it's getting closer to that. It's, it's, it's kind of. This is my interpretation. This is Vitalik and Bastion going into war mode and saying there's a fight on for the planet. Essentially it's going to be a top down command and control planet with the humans who control the big machine intelligences. Or, or are we going to figure out how to decentralize the economy and the world more significantly? And crops is all about that.
- [17:11]
A
Interesting. So we're going to an Ethereum led command and control economy.
- [17:16]
D
No, against the command.
- [17:19]
A
Okay, sorry, sorry, sorry.
- [17:19]
D
Decentralized.
- [17:20]
A
Against the command.
- [17:21]
D
Decentralized protocol.
- [17:21]
A
Centralized. Okay, got it, got it.
- [17:23]
D
So we got AI to empower people.
- [17:24]
A
Yeah.
- [17:26]
D
Sum total knowledge of humanity, best practices, wisdom, access to the domain experts in all the domains in the world, available to virtually everybody. And then you've got economic, social, political, financial agency for everybody who has control of their own assets, who has a wallet, who can operate on Ethereum. So Ethereum, that is a recipe for a much flatter world, potentially.
- [17:55]
A
So it's a bulwark against existential risk. It sounds like there's also, I mean, so one of the stories this week has been one of the existential risks to Ethereum, which is quantum, also came out of the ef. So Justin Drake, I think it was just today he posted that. So if you remember, if we rewind a couple months, the big quantum news, that there was a pull forward in quantum timelines because of these two papers that were released by Google and by or atomic. So the Google paper, what was notable about it is that they created a circuit that allowed them to massively speed up the time for Shor's algorithm or make the circuits much smaller, meaning that it would be easier to basically break an ECDSA key using a quantum computer. And notably Google, in very much a break from scientific norms, they did not publish the circuit. They instead published a ZK proof that, that they knew of such a circuit that satisfied these constraints, that satisfied these properties. And so that's all they published. And this was kind of like a, hey, we're protecting you because this circuit is so dangerous that you cannot possibly see it.
- [18:58]
D
But it's over here.
- [18:59]
A
Exactly. Well, so ultimately this ended up causing a bit of a stress sense.
- [19:03]
B
We didn't destroy it for ourselves though, right?
- [19:05]
A
Exactly, exactly. It's so important that we can see it, but you can't. And so this caused a bit of a stress in the fact we're basically a bunch of different scientists within the community came together and more or less figured out what the different kind of techniques and tricks that Google had arrived at in order to build this. And this caused a bit of an amateur, almost like a folding at home kind of thing, where a bunch of people contributed to the cracking of this circuit. And apparently a bunch of people were using AI to accelerate their ability to even participate in this project. And so it's kind of an interesting phenomenon. And basically they have rederived the same,
- [19:47]
B
or actually I think they reached better. They're like, oh, they improved on it. Yeah. Like since the last two days it actually went to like 12% better, 13% better.
- [19:55]
A
Wow. So they improved on the Google results purely from the breadcrumbs at least in
- [19:59]
B
circuit complex like size of the circuit. But they use the ZK proof as almost like a RL environment. They're like, hey, can we just keep brute forcing circuits until you can reproduce the same proof?
- [20:10]
A
Wow.
- [20:10]
B
And like enough people were crowdsourcing their LLM subscriptions to do this that it actually like kind of worked. Well, no, it did work. I shouldn't say kind of.
- [20:20]
A
So we're, we're at the country of geniuses and data centers where basically your LLMs can like, well, I think better
- [20:26]
B
use of all the free subsidies of compute. You know, once they start charging per token, really, you won't be able. I don't think you're going to do this as efficiently as this was done.
- [20:35]
A
Okay, interesting. So in, in this post, Justin Drake says that he sort of makes a Q day call. Q day, of course, being the day that quantum computers can actually break conventional cryptography. And he claims that he thinks there's a 50% chance by 2032, a 10% chance by 2030, which is obviously very imminent, and that NIST is Sanders body within the US government. NIST's date of 2035 as being the date that it recommends a quantum transition to be a joke that should be discounted entirely. So it seems that among the people who are at the frontier here, very, very clear, we need to take quantum seriously and this kind of stuff using AI in order to kind of push the frontier. Even what world quality scientists at Google were able to do, a bunch of amateurs basically got together, I mean, obviously with a little bit of an advantage
- [21:21]
B
from having, I mean they had the real proof, they could basically grind against. Right. Until they found a valid.
- [21:28]
A
Right. But they were able to improve, basically.
- [21:30]
B
Actually, it's funny, this is like an ironic form of proof of work. It's actually right because they have like the goal, the target.
- [21:37]
A
That's true. And they're just basically true. That's true, that's true. But still it's like, look, the search space is obviously so incredibly.
- [21:42]
B
Yeah, yeah, yeah.
- [21:43]
A
It requires an enormous amount of intelligence.
- [21:45]
B
No, for sure. I'm not trying to. Sorry, I'm not trying to downplay that.
- [21:48]
A
Yes, this is the kind of thing that AIs are particularly good at, which is very closed form problem, very well defined goal. We know that it's possible. Right. So the same thing with a lot of these like Mythos, you know, a lot of people are trying to say, oh, Mythos is fake. Because I take a vulnerability that Mythos produced, I can point conventional LLMs at it and get it to eventually find the same vulnerability. That's only true if you know the target. If you just point it at the Linux kernel or at a browser, it's going to just zip around in circles like a mosquito and not find anything that actually is reliable. So anyway, all that is to say, Justin Drake says very clearly ETH is going to migrate to post Quantum crypto by 2029. There's going to be this lean VM construction that rips out all of the BLS signatures. KZG aggregation and ECDSA signatures are all going the way of the dodo. And so supposedly this is part of the reason why ETH is outperforming the other majors because it has this quantum roadmap.
- [22:41]
B
Nick Carter has some old tweet that's ETH BTC is the quantum index.
- [22:46]
A
Is that right?
- [22:47]
B
Of like like the market's view on on post, how fast quantum will be coming.
- [22:52]
A
Okay, interesting.
- [22:54]
D
So still going to take a fair amount of time for Ethereum to do three major migrations. There's some hash based cryptography work that needs to get solidified. And so let's hope that the world doesn't crowdsource any further solutions to accelerating quantum.
- [23:17]
A
But it does seem like the problem of creating hardened clients is more accelerated by AI than the problem of actually building quantum computers, which is a lot
- [23:25]
D
of engineering heavy hardened clients. That's a different problem from cracking protocols, and that's a different situation from taking machine intelligence, the best machine intelligence, and enabling the builders of conventional software or decentralized protocol software to make use of that. And so in the Ethereum ecosystem, we're already building pipelines where we can formally verify specifications and formally verify implementations of those specifications. So actually there's a lot to be worried about in the next little while. But I think there's a decent chance that we go in the direction of a golden age for software where the builders who do have advantages, they get access to formal verification and just software that will help them think through everything. Most of the vulnerabilities in our ecosystem sit either between the chair and the keyboard. Dumb people doing dumb things, making dumb choices, writing dumb things into parameter files, or in between protocols. And so our consensus diligence team, our security audit team, they've built their own AI, essentially, and they've got all these agents that swarm on protocols that they're looking at, and now they're starting to look at because our ecosystem is composable. You really have to look in the cracks between Protocols and try to determine where vulnerabilities might sit. And that's where a bunch of these more sophisticated attacks, machine intelligence driven attacks. It's three subtle things in these three different situations here.
- [25:12]
A
Yeah, I mean there's been, been a lot about this lately. So the Open Zeppelin founder, what is it? Manuel Aro. Aro. Okay. There's an accent in there, so I don't know. Okay. Manuel, he tweeted that he believes all of Defi is now unsafe and he's recommending that all of his friends and family withdraw everything from Defi. Now this tweet got a lot of attention, got a lot of people talking, a lot of people very angry about this claim, especially given that you know, his, presumably.
- [25:41]
D
And it doesn't matter because Eliezer Yudkowski told us we'd all be dead pretty soon anyway. It's, it's the same kind of thinking.
- [25:47]
A
You think so? I mean, how does this land for you, this idea that all of.
- [25:51]
B
I mean, I think it's obviously I, I kind of agree with Joe, where there's like this horrible two year period and maybe two years too long. Maybe it's less than two years.
- [26:00]
D
Yeah, yeah.
- [26:01]
B
Somewhere between one and two years.
- [26:04]
A
Not the same claim.
- [26:04]
B
No, no, but there's this horrible period of like cat and mouse games. And they're all these kind of like what you would call like a Stackelberg game, security game, which is like, you know, the, the defender, like the person writing the code sets their, chooses a strategy. They choose their strategy by like how they wrote their code, what, what parameters they chose, whatever. And then the attacker plays a game, plays a move, and then the defender adapts, updates, attacker plays. And the question of those games of that form is like, what equilibria do they reach? Do they reach the equilibria? They're like the defender wins or do they reach the ones where the attacker generally starts winning? Most of these types of setups, the defender does eventually win, but the attacker has this like long period where they like are. It's like win, win, win, win. You know, it's like a little bit like when you're training a bot to play chess. Usually the first 5 million epochs, you're losing almost everything. And then all of a sudden it's like you found, you know, you will. You look at the thing and it like finds one thing. It kept messing up for like all the Fast 5000. And then all of a sudden the, the thing starts working again. I kind of think it's going to be like that except much faster. So it's just like we're going to. We're going to build
- [27:17]
D
in many business domains like the finance industry. You have banks, you have guarded, armed stagecoaches, you know, vaults and things like that. And the builders think through a few little pieces. They offer their services, and the crackers, they look at every little angle for how they can get to that money. And they essentially perform a consulting service where they go in and they show the builders where the vulnerabilities are. They take their consulting fee, assuming they don't get caught, in which case they have to maybe give back their consulting fee. Right, right. And the industry proceeds from there. And the industry.
- [28:00]
A
So this works well, if the consulting fee is 10%. Right. But if we're talking about like a ronin hack situation, the consulting fee is everything.
- [28:08]
D
That's the consulting fee for the entire industry.
- [28:11]
A
For the entire industry. Right. That only works if we are all socializing the lessons that each individual defender is learning.
- [28:18]
D
But don't we do that?
- [28:21]
A
That's a fair question. Do we do that? So right now, Thorchain, which is this kind of infamous, I don't know, let's say cross chain spot trading protocol, it's been shut off for like two weeks now because they got hacked two weeks ago and they can't turn it back on. Literally, it's like so broken. And then yesterday, I don't know if you guys saw this, but V12, which is a AI detection tool, automated security scanner built by Zelic, they basically found a bunch of vulnerabilities in Thorchain. And they were like, hey, we want to report these. And Thorchain was like, no, we're not going to pay you anything. They said, oh, okay, we'll just disclose them then because you're down anyway, what does it matter? We'll just tell everyone what's broken about this chain. Now, Thorchain might be a uniquely kind of bad piece of code. I don't know in particular, but it seems emblematic of there's a degree to which we have learned a lot of lessons in industry, but we've not learned them uniformly. It does seem like there's a lot of places where there just isn't the security spend isn't the sophistication. There's a lot of talk right now about how bug bounty programs are overwhelmed by all of the AI driven reporting, where somebody who runs a bug bounty is like, okay, we pay this much for crits, but we're getting 50 reports a day that are claimed to be crits. And we just can't triage them all.
- [29:38]
B
My thing is, you think about how this happened in traditional finance. People lost a lot of okay, let's just take the railroad boom, the original Great Depression, which was the late 19th century railroad boom, there was tons of fake at that time also it was like very much like the stablecoin era where there's like lots of private money. Banks were issuing their own dollars also all these like companies were issuing their own bonds. You were getting scammed left and right by, by fake bonds and people like issuing too much of their dollar. And at that time an interesting thing that came out of it was like everything blew up right? Like and the, what was initially called the Great Depression until the 1930s one took the moniker afterwards. You saw the creation of some of the largest insurers who after that time were able to sell corporate bond issuers on hey, your purchasers are only going to buy your bond if they have some notion of coverage. And then the private money stuff also had a similar thing where the FDIC insurance offered by the government effectively moved. No one wanted to hold the private money anymore. And so obviously FTSE is a little later. But that was the thing that slowly drained that system. And I sort of think the main problem with DEFI right now is composability makes insurance very hard right now if you think about insurance, every time there's an attack, the insurance is at the end point. It's the last place that got attacked having to pay out for the attack. But if I think about a composable financial product, there's many inputs to it. Like there's the bridge, there's the oracle, there's like the issuer maybe has to sign something. If it's like an rwa, there's like this whole supply chain. And I think the thing that we're missing that probably will be the innovation from attacks is how do you distribute that type of cost across the system as opposed to just trying to fix it at one place. Right. The traditional financing is like I have a single entity, single point. And we're trying to fit the square peg of try to do coverage at one point versus across the network. And I think that's what you're going to find is going to happen over time. Now I'm not going to say it's going to happen tomorrow, but I think that will get.
- [32:01]
D
There is something that's happening that is going to rearchitect the architecture of the Ethereum set of networks. I'll call it the extended Metropolitan Ethereum Ecosystem. I think of metropolitan ethereum as layer 1 and a bunch layer 2s, and then there are these other Ethereum technology networks that we'll call them extended for now with some near synchronous and asynchronous composability techniques. So Zisk and Gnosis are doing it in the Ethereum Economic Zone project. Our Linea team is doing it. So we are now just able to run zero knowledge proofs across a few linear networks and even into private permission BESU networks. So some of the. There are like 200 Besu networks in the world currently and Citibank runs Citi Token services. So these are some big businesses running on Ethereum technology in a private permission context. So we can take zero knowledge proofs in those situations, we can aggregate those into a zero knowledge proof, into a block on a layer two or into a block on layer one. And we can create these atomic execution zones where we can have a bunch of transactions happen across a bunch of different networks. We can have tokens move across these networks without the need for bridges. Because you're just doing zero knowledge proofs. You've got somebody verifying that all these zero knowledge proofs line up and that is a essentially bridge free architecture that removes a lot of the vulnerabilities. It's really hard to get inside a zero knowledge prover and hack that thing. We'll see if that's true. My point is that I think we're moving into a much safer architecture based on zero knowledge proofs. And the advantage is we're bringing composability into a situation where we sort of intentionally introduced fragmentation.
- [34:07]
A
Sure.
- [34:07]
D
So for instance, you can do real time unification of different liquidity fragments across these different networks.
- [34:17]
A
Tom, let's bring it in here. What do you think about where we're going?
- [34:19]
C
I mean, I think the Thorchain point, I mean, I guess I also kind of cop it to general open source development where it does feel like hey, new improvements generally get rapidly integrated into whatever the live code base is. Whereas in crypto there is this very long tail of cropped. And we do occasionally see these really old pieces of code that get vulnerability discovered like six years after the fact, there's not really an opportunity to deploy. But hey, if these best practices are being integrated into common libraries that are then used, even OpenZeppelin being one example, the average quality has improved, which I think why we see also more attacks that are kind of on this user application level versus even six years ago when novel smart contract attacks seem to be so so common. So I guess it's all to say I agree. I think the problem is the deployment and rollout process in crypto is just super long. And so it might be a while until everyone is, I don't know, running really hardened, formally verified software. In the interim, you're just going to have this like, you know, these laggards that just aren't. And I think, you know, again, even this isn't even like a crypto problem. It's like a general open source problem. Like using this now and we've had what, like three months of crazy supply chain attacks and people are like, oh, maybe now, maybe we should have like very strict version pinning for all our dependencies and Cool, that's now becoming more commonplace and more talked about versus, yeah, we'll go to whatever the latest version is and assume that it's going to be hardened and good and reviewed and all that.
- [35:43]
A
Yeah, there's so many cases in the history of technology of really, really old software just breaking right. So you remember Heartbleed, which was like a bug in very basic networking stack. There was Spectrum Meltdown, which was like every single intel processor had these very, very deep vulnerabilities in the microcode. These are almost certainly not the last of their kind. And blockchains by their nature are not really designed for quick patching. They are these kind of slow, discursive. Okay, we all argue in a discord. And the validators, we can't get them all on the phone or on a single zoom call, at least in some ecosystems.
- [36:21]
B
I was like, it's funny you bring
- [36:23]
C
that up in the same AWS region.
- [36:25]
A
Yeah, yeah, yeah. Well, it's funny actually because, I mean, this week there was this story about Sui, that SUI was down for 48 hours and it was down three times in a row because they had some vulnerability and then they patched that which caused another vulnerability and they patched that which caused another vulnerability or something like this. It was like a sort of cascade of failures. And this sounds kind of goofy, but I'm like, we'll probably see a lot more of that when there are these. Like, okay, AI discovered this and we were talking before on a previous show about how we're seeing not only AIs be able to Mythos quality models that are Mythos now. If you look on polymarket, it's like 70% to be released by the end of summer. So if Mythos is going to be in the hands of ordinary people and the amount of people like this quantum DIY stuff, that's what security review is going to be. That's what Black Hatting is going to be. That's what Bugman is going to be.
- [37:18]
D
Mythos is not. I'm sure it's much improved over the previous state of the art, but so is 55 chatgpt and whether or not.
- [37:31]
B
Yeah. Honestly. And if you might seem the greatest
- [37:33]
D
marketing campaign and you just give them more cycles.
- [37:36]
A
Yeah. But a lot of what I want to say was that we're also seeing not just the quality of models increase, but also the speed of models increase. Right. Where there's fast mode for all these models now where they're just lowering the batch size that you're just Getting twice or 2.5 times the tokens per second. But then we've also got these models on ASICS now that are spitting out Thousands or even 10,000 tokens per second. At which point it's like, okay, this is multiple days of work happening within minutes. That is just grinding every possible attack vector against this thing. And our intuition's about, okay, we have humans in discords coordinating with release schedules of how we're going to update this code. And if the attackers are happening at AI speed and the defenders are happening
- [38:18]
B
at human speed, but the defenders will also be the AI speed. Eventually. Right. That's the point.
- [38:22]
A
Eventually. Eventually. Right. But we have to converge.
- [38:25]
B
I, that's very different. I'm giving you this time gap. I'm like, there's like the defenders being able to use it effectively will take longer than the attackers.
- [38:31]
C
Right.
- [38:32]
B
That, that's, that was why I brought up this kind of.
- [38:34]
D
I mean, it's like the real answer is to build a world of abundance so that we don't have nation states that feel like they have to steal a bunch of money to build nuclear bombs.
- [38:44]
A
So you think the answer is that we have to just give a bunch of food to North Korea? I think it'll stop attacking.
- [38:48]
D
I think it's a cyber war issue.
- [38:50]
A
I mean, how do we solve needs
- [38:51]
D
to be played at the nation state level. Potentially.
- [38:53]
A
Yeah.
- [38:54]
D
Or we can.
- [38:55]
A
Well, but, but I, I, but I disag this because North Korea has been such a prominent attacker of crypto because they are abnormally well resourced from a cybersecurity perspective. They have actually really good talent. And we don't have a great explanation why they punch above their weight despite being a fairly poor and largely low state necessity. There's a lot of countries that would love to have a cybersecurity program that could just make billions of dollars, but Only North Korea has pulled it off. Sometimes you have these weird quirks of just, okay, there's like one guy who trained a bunch of people. I don't know, who knows? But in a world where everybody has access to frontier intelligence, then the limiting factor really is just money.
- [39:31]
D
That sounds like a world of abundance to me.
- [39:33]
A
Well, it's a world of abundance, but it's also a world where everyone has a gun. Right. And it's like, okay, you come to your views about gun laws and guns.
- [39:42]
D
If everybody has a gun, then everybody should be nice to each other.
- [39:46]
A
That's not in practice how I think American society eventually has turned out that way. Yeah. Or people have guns pointed. I mean, if you go to many places in Africa, they also have a lot of guns and they're not very nice to each other. So I don't think it's like a settled equilibrium that having extremely powerful offensive tools in the hands of everybody means everyone will be nice.
- [40:07]
D
You have to have more than just guns in every hand. You have to have social cohesion.
- [40:14]
A
Yes. Which blockchains are not exactly known for. So here's what I'd say. Look, I'm somewhat partial to Manuel's call and I think we're all kind of echoing a little bit like, hey, we're going through a rough period. There's going to be this pace where attackers learn faster than defenders learn. Defenders will eventually learn, but they will not learn as fast as the attackers learn. The best resource defenders will get there because they will adopt these tools just as fast as the attackers will.
- [40:40]
D
But the attackers are actually finding vulnerabilities in the cracks and via social engineering. So it's not going to be that hard for us to get smarter now that we're alerted to different attack patterns.
- [40:53]
A
Right.
- [40:53]
B
I mean, another thing I think maybe more optimistic version of this is all of this attacking has an immediate cost, but it has a long term benefit of like, hey, what percentage of coverage do I have over the space of attacks? And it's like some of these attacks actually dramatically increase such that the next, you know, if I look at the future set of potential attacks, I've shrunk it so much that it actually gives me a lot more confidence in the hardening.
- [41:22]
A
Right.
- [41:22]
B
And so there's a sense in which this is a very expensive bug bounty in one lens, but there's also a sense in which it might actually lead to something that really is impenetrable eventually, which I don't think we had the as strong conviction of getting to like, I Don't know if you've read Mozilla's blog post about Firefox with Mythos and ChatGPT. It was like they were a little overconfident I thought where they're like oh, we think we'll be 100% vulnerability free by end of year or something. But like they showed the curve and then they showed the surface area covered and you were just like watching how little they actually really audited of their code base. They like had certain parts right where there's like HTTPs and, and encryption where they were like okay, we over audited there but other places we didn't really care.
- [42:08]
A
Yeah.
- [42:08]
B
And like that's where they were just like churning out finding all the zero days. And so there's a version of the world where like we're paying a huge upfront cost but maybe the end state is actually a much more impenetrable system. And the optimism for blockchains and open sources, they're going to be the first ones to reach that Valhalla hopefully. Yeah, I mean the closed source stuff is going to actually be where you're going to find the bleeding many years later because everyone's like ah, it's too expensive, I don't care, I'm not going to, I'm too smart.
- [42:34]
D
So I think you're right. In traditionally in code there's probably 15 bugs every thousand lines of code somewhere around there there are probably other numbers and, and to the extent that machine intelligence will enable private repos and traditional software to clean itself up, that's great. Ethereum I think of as an anti fragile technology and an antifragile ecosystem which is really what you're talking about. So as long as we start paying attention to it and apply the proper resources, I think we're going to squeeze out the bugs from software that's already been written and we're going to start building formally verifiable software and it's going to be hardened software world.
- [43:17]
C
I agree. I actually think I'm more worried about these old closed source code bases now that hey, it's getting much, much easier to decompile these old binaries. I'm think of so much of the world, yeah, we love Linux here, but so much of the world doesn't run on Linux. And that's the stuff I feel like definitely does not have the security standards that should because there's so much reliance on, on security by obscurity. And so I think there's going to be like just those are going to
- [43:43]
B
be soft targets and who's getting paid to go maintain this old code that might be have a ton of bugs? It's like everyone is like, especially in a big orders, like everyone's looking for
- [43:53]
D
whatever has the economic interest is going to.
- [43:56]
B
I think, I think there's not as much.
- [43:58]
D
When was the last time that humanity faced an existential threat and said, eff it, we're all going to die. Let's not bother fixing it.
- [44:07]
A
Okay, here's where. I mean, look, I appreciate the optimism that you guys are projecting, but I'd say the model that you're describing of like, okay, well there'll be a rough period, but then we'll hurt immunity, we'll get through all the bugs and then everything will be super robust.
- [44:20]
B
Yes. This is a pandemic.
- [44:22]
A
Yeah, well, I think it actually is a good analogy.
- [44:24]
B
Kind of feels like I think it
- [44:25]
A
is a good analogy. But the problem with this is that the attacker keeps changing because the models are getting smarter and there are bugs that require much more creativity that the current models are not leveraging. So if you take GPT 3.5 and you try to get it to come up with an attack, it can only do something if it's right there in the five lines of code. But you get Mythos and it's like, oh, I can grab something from over here and something over here and it can construct these longer chains. The successor to Mythos is going to be able to find categories of bugs that only super geniuses.
- [44:58]
B
But the successor to Mythos might also be able to write the code without the bug by speaking against itself.
- [45:05]
A
We require it in principle to be able to understand these very long scale
- [45:08]
B
effects of some point in this trajectory. Humans stop being able to understand anyway. So we're going to outsource.
- [45:15]
A
Well, we'll understand when we're being exploited like that.
- [45:18]
B
Yes, that's like pnnp. I can verify quickly.
- [45:23]
A
Right, right, right. Exactly, exactly, exactly.
- [45:25]
B
But I think that's the direction we're going to head in anyway.
- [45:27]
A
Look, I think this is almost like. It's almost like analogizing to moral progress. Right. If you think about if you were in the 1800s and you were like, okay, people in the past were very immoral, but today we understand what morality is. And so we're going to just get rid of all the evil things that people did and then 50 years pass and you're like, wait, what? We didn't think that disabled people should be in the moral circle. We didn't think that black people should be in the moral circle.
- [45:52]
C
Think of it more like scientific progress where it's the type of work required to have a real breakthrough discovery in 2026 is so much harder even though you have so much knowledge to stand on top of and tooling.
- [46:03]
A
That's only true if the models aren't getting rapidly way smarter. And it seems like they still are.
- [46:08]
C
I'm saying you're standing on so much improved tooling and knowledge bases and speed now as a scientist versus 150 years ago. But hey, overall progress has slowed down. I guess the question is, do you think it is possibly to formally verify a non trivially small application or do you think that's just like, that's just impossible?
- [46:29]
D
It is.
- [46:30]
A
I think the formal verification is the strongest argument for like, okay, there's correct by construction as opposed to a. We have the Swiss cheese and we've put the holes in all the Swiss cheese. But it turns out it's multidimensional Swiss cheese because now it's in the third dimension and then eventually that post mythos in the fourth dimension and it's just almost impossible to cover all the holes.
- [46:49]
C
The slop it up more method I think has a lot of like.
- [46:51]
B
Yeah, you know what's really funny is like listening to you say the last two sentences you said. You sound like Vlad. Like ethereum. Vlad.
- [47:01]
C
Vlad.
- [47:02]
A
What does this have to do with Vlad?
- [47:04]
B
He would say a sentence with the absurdist trolls 3D. He would 100% say that.
- [47:12]
A
Okay, that's really weird.
- [47:13]
B
Especially because you started with correct by construction.
- [47:16]
A
That is true.
- [47:18]
B
Correct by construction. 3D Swiss cheese. I was like, haseeb, who are you?
- [47:23]
A
That's funny because actually Vlad, I think Vlad absolutely hates me. But I haven't heard his name in quite a while.
- [47:27]
B
But trust me, you just said a sentence.
- [47:29]
A
I think I just. I did channel my interval.
- [47:31]
B
Joe can also tell me if I'm crazy.
- [47:33]
D
No, you're right.
- [47:33]
A
Yeah, yeah, yeah. Okay, okay, fair enough. I will take that. So there's one other story I think kind of related to cybersecurity. Kind of loosely related. That. That was very interesting. So Zama, of course, privacy protocol. It's. It uses fhe. Fhe fhe in order to achieve its privacy. They have this privacy protocol called overnight Finance. And there was somebody who deposited 12.5 million USDC into their confidential USDC wrapper. This was like 99% of all the USDC deposited in there. And a federal court issued a restraining order to the founder of this protocol, which is called overnight, this guy Maxim irmalov. Over the $15 million treasury, which is apparently a rug from someone, something else. And so Circle froze the CUSDC contract in response to this court order. This caused a bunch of kerfuffle online of like, oh, my God, Circle is freezing privacy coins. Does this mean that you can't use Circle USDC privately? People eventually untangle that. Oh, no. This was in response to a court order. Now, what does that mean that this court order happened? And so this all seems to have revolved back to this guy Diogenes and his company fund, I guess, called Patagon. So Patagon, he apparently was the group that was pushing for this injunction or this restraining order on the protocol, because he had bought the underlying protocol's tokens, I guess, and was almost like, what's the name of these financials? Yeah, like a corporate raider buying distressed assets. Yeah, like an activist investor, effectively. And apparently this is not the first time that he's done this, of buying a distressed asset and then pushing for a court order to try to get some assets returned to the underlying investment. He got called up by Zach XPT on Twitter for doing this and got a bunch of people kind of upset about it. Anyway, the story, it all resolved. The court order was dropped, and I guess whatever's happening there is happening. And so Circle has stepped out of the story. But I just thought it was one of these interesting little glimpses of maybe where we're going, maybe this is going to be the new normal of, like, every time we see something like this happen, we're like, oh, wow, that kind of looks like traditional finance. And it might just be like, oh, this is just how everything looks.
- [49:52]
D
What was the law firm? Was it the absurdist troll law firm that went after Arbitrum?
- [49:59]
A
I don't believe so. I think it's this. I don't know what the law firm is, but this is not a classic.
- [50:04]
D
That can be a vector.
- [50:05]
A
Yes, yes, yes. That was not.
- [50:08]
D
Lots of troll law firms are going to go after protocols for different reasons.
- [50:12]
A
Right. But the long awe of the long arm of the law seems to have now extended into privacy protocols and into.
- [50:20]
B
Well, I think this is the difference between a privacy protocol that's an overlay on the network versus a privacy protocol that's, like, built in.
- [50:27]
A
Right.
- [50:28]
B
I think that's.
- [50:28]
A
Well, when the underlying collateral is feasible for a privacy protocol, that's the underlying weak point. Okay, so one of the other things that we should talk about. We have a few minutes left. Obviously, while everything else feels very gloomy in crypto right now, Hyperliquid has been hitting just all time highs. I mean we keep saying this on the show but once again it crested $75 which brings it to a fully diluted valuation of what is it, 50 something billion. So it recently in FDV terms flipped Solana. It is now number nine of the most valuable crypto assets and it's one of the only things that's been going up while everything else has been just getting hammered in the markets. And so there was this guidance issue.
- [51:07]
D
Venice, near Venice. There are some things that are going up.
- [51:10]
A
There are some things that are going up, but not a lot lighter. Lighter as well. Lighter as well in the Ethereum ecosystem. Shout out to lighter. So anyway, the CFTC recently issued some perps guidance and approved the first US perps product from Kalshi of all people. Kind of a curious first choice to make. People are a little bit surprised, like why not Coinbase or one of the US exchanges? But Kalshi had a UST or sorry, a BTC perp which is approved on Kalshi is the first product to come on the US markets. And supposedly this sets a precedent for further product reviews, although they kind of clarify that. Look, the kinds of things that we are going to be approving are probably not equities and also not ag products because of course AG controls everything in this country. But it seems like the door is now open for US domestic purpose products. This caused hyperliquid to pump, as it seems almost every news caused hyperliquid to pump somewhat curiously because it's not obviously good for hyperliquid. It's like a little bit of a complex story about whether or not. Okay, is hyperliquid going to enter the US or they're going to.
- [52:16]
D
It means it's coming to America at some point.
- [52:17]
A
Sorry.
- [52:17]
D
It means it's coming to America at some point.
- [52:20]
A
You think hyperliquid is coming to America?
- [52:21]
D
I think that's it.
- [52:22]
A
You think they're going to create a regulated US entity that gets a CFTC license and KYC's customers domestically?
- [52:29]
D
I think that will happen.
- [52:30]
A
Okay.
- [52:31]
D
And I don't know the timeline.
- [52:32]
A
Okay, so you think they kind of pull polymarket style where they have the over the overseas offshore product and a domestic US product and you think the domestic US product is going to be
- [52:40]
D
competitive style like offer your American product and keep your business going in the rest of the world.
- [52:47]
A
Do you think that product stays competitive? Because I mean that's ultimately going to be a centralized exchange.
- [52:53]
D
So hyper liquid, great team known them for a while. Great product. We actually have one of the bigger interfaces in Metamask to it and they're going to keep doing great things, but there's going to be a huge amount of competition. So lighter's like great Ostium. What's the other rfq?
- [53:13]
A
Variational variation. We're investors in variational as well.
- [53:16]
D
There's a little company called ice, New York Stock Exchange. These organizations are tooling up. They're actually building or extending Ethereum technology. And I think it's going to be a massive sea change in how the world participates in finance. Essentially most of the world is financially illiterate and are regularly led to the slaughter by the financial industry. So cheap essentially to the extent that finance can be more entertaining, more engaging in the form of prediction markets, perpetuals, simpler product, a bit more like a casino than an actual commodity future or an equity. But to the extent that it can draw people in and get them learning things. You can ask your favorite machine intelligence, what is this thing? What does it represent? How do I use it? Owen, tell me a little bit about the math behind it. So I think this is part of a theme that we've been exploring here today where the world may get more decentralized, people may get more empowered. And this sort of entertainment layer, the perpetuals layer and prediction markets layer is a great way to draw people in and get them thinking numerically.
- [54:39]
A
Tom, how do you think about this for hyper liquid and or the other?
- [54:43]
C
Well, I will say I think LIDAR is the only perp that's Delaware Corp and issued the token out of Delaware Corp. So maybe something to consider. I think obviously the Kalshi approval is like, like the first of many. I was actually surprised there were not more like Robinhood for example, that has been talking about perps and perps offshore. The question you're all sort of saying is just like how do you do this in a smooth unified way? Is it always going to be segregated onshore, regulated offshore, or is there going to be some way of these basically acting almost as brokerages? Or is there a way to sort of have a nice siloed model or something like that that can still send interfaces both ways and we're seeing maybe some progress with that. There's a whole no action letter for Phantom and so their definition as a broker and cool. It doesn't strike me as so much of a gap to say, well hey, if you're offering this other type of asset, why not something else too? And so I think I even look at frankly at polymarket, which obviously is not us accessible right now, but it's mainstream awareness and penetration of the US being like, hey, there's maybe more acceptance or understanding of on chain derivatives being, you know, something that people actually want and a way to sort of offer something that's global and the advantage of doing that.
- [55:51]
A
Tarun, what's your take?
- [55:53]
B
I mean, I think the key thing, key innovation for hyperliquid that no one has copied is distribution. Obviously like between getting other people like Metamask or Phantom, distributing your product between people distributing assets in the HIP3 and I guess soon HIP4 world. I think no one in crypto has ever pulled off that type of distribution play. Like, I think like an L, one is supposed to do that.
- [56:21]
C
Right.
- [56:21]
B
But now one is like almost too many counterparties that you have to agree, get to agree. Whereas like they somehow found a way to do it without complexity. And I, I think the other perps, exchanges obviously are trying to do the same thing. But it, if you're the first one and you're the one mentioned in Congress all the time, it's like, it is. I feel like it's very hard to imagine that you don't end up trying to go in the U.S. yeah.
- [56:46]
D
So circle did a pretty good job arranging distribution for themselves, but sure, in a decentralized context.
- [56:52]
B
Yeah, yeah, yeah, sorry. I. I guess, I mean more of the, like you can build new markets on there, but it's not like you have to rebuild everything from scratch, like sort of. It's like, it's like somewhere in between a blockchain where you have to write the full stack and like someone giving you a bunch of SDKs and APIs that are like you only get a certain amount of input. I think that's arguably why Hype EVM has failed in a lot of ways is like no one wanted the full programmable thing at that. I'm going to get roasted for saying
- [57:23]
A
that, but I think it's the one thing you might not get roasted for. Given everything else, it's very clear that Hyper Liquid, where they are their strongest is where they lean into just pure markets and using the hyperliquid infrastructure at
- [57:37]
B
its best, as opposed to also not building everything themselves.
- [57:40]
A
Right, right, right, yeah, exactly. But also not trying to be just a generalized platform. The generalized platform part of Hyper Liquid is not what's worked. What's worked is the market specific platform. We're a platform for new listings and new markets and order books of all kinds. And that's where clearly it has this Tremendous staying power and a lot of entrepreneurial activity. Like we see it all the time. We still get tons of pitches for people building stuff on Hyperlake. We liquid increasingly. They're not building on Hyper VM increasingly they're just like, hey, we're going to launch a Hip three, Hip four, whatever on A, B, C, D, E, F, G, whatever. All sorts of the universe of financial instruments and being able to attract that kind of capital formation on top of your very constrained platform of like, look, you have a little spot in our order book. Here you go. Is incredible. It's an incredible testament. I've never seen anybody else.
- [58:28]
B
It's like the bucket shop position of an exchange. Right. Because an exchange is usually integrated as one and it's like, like I've partitioned it into small components where I have like distributors who are getting order flow, who are sort of like prime brokers, but not really. And then I have market creators who are like, you know, some, some category of users. I have like unit as like the spot and it's like unbundled exchange. Right. And like sold off some portion of it. But that let you grow faster. Yeah, that aspect I think is the thing that clearly they were the first ones to do.
- [59:05]
C
Right.
- [59:06]
B
Everyone else I think is racing.
- [59:07]
D
But the competition is coming.
- [59:09]
B
No, no, I agree.
- [59:10]
D
PI is going to get way bigger and the RFQ space is going to be really interesting.
- [59:15]
B
I just think the distribution model is the thing that is the unique thing and they got the brand recognition the way Polymark has a brand recognition despite not being in the US and it's a very similar. I mean the number of congressional mentions is similar to Polymarket order of magnitude wise. So it's like, I think it's like kind of hit some marketing part that I think the competition actually has to catch up on.
- [59:40]
A
Yeah. It also has this thing that like polymarket, the number of people who are betting on it itself is much smaller than the number of people who know about it. Look at it, use it. Right. So it's got this kind of Polymarket effect where it has enormous amount of free marketing where you just go look at it and see what are the oil prices today, what's the pre IPO price for cerebras or for SpaceX. And that is such an enormous marketing engine of earned media that you just can't buy.
- [60:07]
B
I think that comes from the distribution
- [60:09]
A
model of being very wide, but distribution begets distribution. And that flywheel is something that we've really only seen from Polymarket and now from Hyperlook we.
- [60:20]
B
Yeah, I just don't know if like L1s have ever been able to really. They want, I think they want to achieve that, but I think, I think Solana had that.
- [60:27]
A
Solana definitely had that over the last like 2024, 2025, where it was just everywhere. I mean Ethereum has had that in previous eras as well.
- [60:35]
B
But I'm not sure if it was Solana that had it or like applications that showed you Solana stuff. Right? Like there's a slight fandom or pumpkin or whatever.
- [60:43]
A
Yeah, yeah, you're not wrong. Okay, we're coming up on time. Joe, give us something to feel positive about in the Ethereum ecosystem because a lot of people, regardless of what we've said on the show, they're feeling the blues. They're feeling like it's tough out there in Ethereum land. Why should we feel positive about the ecosystem?
- [61:00]
D
The agents are coming. The hybrid human machine economy is coming. I can't say anything about what MetaMask is going to announce at some point soon, within a small number of days, but it'll facilitate that consensus. The institutional side of the company is working with or in deep discussions to work with a bunch of FMIs and major financial institutions. They're using Ethereum technology, they're using Linea technology, setting up different linear networks. We contributed besu, our execution client, to Hyperledger Foundation. We just contributed Linea, our layer two to Linux Foundation Decentralized Trust, which is the same thing. There are a hundred well announced projects that have built businesses on Besu. There's another hundred not so well announced. We can bring all of those into the Ethereum economy proper. We're participating in building that and we're going to link them all via different forms of composability into an extended Ethereum ecosystem. It's happening pretty fast.
- [62:05]
A
Okay, beautiful. I love that. That. Well, with that, we'll be back next week. Thanks everybody. It's all for now.
- [62:12]
D
Thanks guys.