Crypto Venture State of the Union | Haseeb Qureshi, Jordi Alexander, and Mike Dudas

Transcript

Usually what happens in crypto is that the crypto markets are crazy and make no sense and they're completely insane and then public markets are this kind of the they're stayed and rational and institutional and so on. And I've never seen this place where crypto markets are kind of stayed and rational and reasonable and they're kind of sitting sidelines and public markets are just bulliant. They're insane. Like no one understands what's going on. Nobody can explain anything. Um, so it's a weird moment and I can't imagine it's going to persist for that long, but that's where we are right now. All of the energy is in public markets and not in in crypto markets.

What's up, guys? Just wanted to give a quick shout out to today's sponsor, Appos. We'll hear more about how Bitcoin, DeFi, and Apptos is heating up later today. But for now, let's get into the episode. Nothing said on Xerox Research is a recommendation to buy or sell securities or tokens. This podcast is forformational purposes only and any views expressed by anyone on the show are solely our opinions, not financial advice. Pocachio Ryan and our guests may hold positions in the company's funds or projects discussed. Hello. Thank you for that, Felix.

Hello. Hello. Hey everybody. Great. So, this is the VC State of the Union. Um, I'm Bacio. I'm going to be moderating. Um, before we get started, should we do a round of intros? Uh, I'm a at Dragonfly. Hi, Jordy from Seleni. Mike Dudis, sixman Ventures. Fantastic. Um, to get us started, how has crypto VC evolved over the past year and going forward? How do you see it change?

Uh, I think we should have the hedge fund guy answer that first. Yeah, I'm the I'm like the fake VC here. So, I can say like without feeling too bad that the state of our union is not that strong. I feel like the venture, you know, we're seeing a lot of venture funds want to get into liquid. They want to do other stuff like, oh, we can't wait for a year vesting. By the time we get it, it's it's like wrecked. So, um you know, we're we're a prop fund. We run our own capital. We love doing venture bets when they make sense. We do quite a lot of them. Uh but we don't have to do them. we don't have to deploy. So, if there's a period where, you know, the deals are not great, we'll we'll just step aside. Um, these guys are the pros, so I'm I'm just

laring. Uh yeah, I'd say, you know, it's funny because uh I've gotten into a lot of Twitter arguments over the last six months defending the honor of crypto VCs and uh I think a lot of people, it's weird because I've never seen a moment in markets where um you know, usually what happens in crypto is that the crypto markets are crazy and make no sense and they're completely insane and then public markets are this kind of the they're stayed and rational and institutional. and so on. And I've never seen this place where crypto markets are kind of stayed and rational and reasonable and they're kind of sitting sidelines and public markets are just bulliant. They're insane. Like no one

understands what's going on. Nobody can explain anything. Um so it's a weird moment and I can't imagine it's going to persist for that long, but that's where we are right now. All of the energy is in public markets and not in in crypto markets. Um one of the things that people often forget about cryptovs is that we do both. So we invest into uh about you know actually roughly on a cost basis when we invest into startups about half of our investments are into equity and about half of them are into tokens. Um and so you know it's weird that the tokens are doing not so great and the equities are doing kind of weirdly amazing. Um but it's like okay cool. I mean I'm I'm we're exposed to both. Um, I do think eventually that that energy that is um, you know, it's

kind of like, you know, when you push an air pocket in a carpet, it like moves over here. Like at the end of the day, that energy is still there, right? And eventually we'll find its way back to tokens when um, when the equity markets rationalize. But it's just kind of weird that now what people are talking about is, you know, how can I get my my shitty useless shitcoin onto the stock market so people will buy it? But that's where we are right now. and and again I think that that that will eventually resolve. Yeah. So crypto equity clearly more valuable uh over the last 6 12 months uh which has created some really interesting dynamics right like a year ago you would have thought hey I'm doing a deal I'm putting a couple million you

know if I'm dragonfly 20 million into something and uh I'm expecting the token to be you know the majority of the terminal value. Uh now it's not so clear. So you have to look at the holistic value of an entity try to figure out uh you know basically put a value on this entity and then it's unclear if it's you know the value is going to accrude to equity or token. uh things that have changed that is one public market price appreciation. Two uh some tokens that are now starting to have value acrruel mechanisms. Hyperlid obviously being you know the the clearest one and then the pressure that that puts on subsequent projects to actually push value to token holders and not just to equity. Uh what's interesting is um there's a bifurcation

right now. So I think in DeFi you're starting to see for example a lot more projects move value you know to the token. Uh in DPIN for example there's a whole bunch of confusion around okay how do we like restructure we have these real world networks. We've made promises to people who are providing supply or purchasing demand and how the economics work here. The transition's taking longer. So to summarize all of that, uh we're in a real moment of flux, uh where I think it's a good thing and that people realize that equity can have value. Uh but people realize that tokens uh shouldn't necessarily only trade on kind of like mimetic thematic large market TAM things, but you should

have line of sight to uh you know some credible token value acral. And then the follow on question to that is okay well if I'm going to invest in the token am I confident that that value will predictably long-term recruit to the token or is there a set of you know insiders who are going to be siphoning off valued equity that makes sense um does that affect the ICM thesis or the internet capitals the markets thesis at all the fact that equity might be more valuable than tokens sorry I didn't catch the first part the ICM thesis Internet Capital Markets.

Is that like the salon? Does anyone Does anyone know this ICM? I don't know that that's more of a slogan than like a token. Sorry. Am I crazy? ICM. I just came from the World Series of Poker. So I I've been hearing about ICM all day, but it sounds like a different ICM. I don't think that's an ICM. That's a different ICM. Don't worry, that's a slogan. It's not a value cruel theory. Um, so it is for for Salana. It's like, hey, if Salana is the future of international capital market, sorry, is internet capital. Internet capital. See, sorry. This is this is not a real thing. It's not a real thing. I want a completely different Can you guys go away saying the word ICM? We have failed you. Do not use Do not use ICM. That's not real.

It's a slogan. Look, well, I I'll say this. I remember there was a tweet recently by I think Andy from the rollup where he was like, "When do you think this focus on revenue narrative is going to end?" And all the responses to him were people laughing at him being like, "What are you [ __ ] idiot? Like revenue will never like this is it's over. there's no there's no we're not going back from this. Um, and I think it's uh, in a way it's it's, you know, it's it's kind of funny to dunk on that that tweet, but I think he's actually probably right that um, it's very normal for markets to not focus on revenue when they believe that the opportunity is really big and in the future, right? So like the AI companies today or at least, you know, as of a

year ago weren't making very much revenue. Now the revenue is starting to ramp, but for a lot of these AI companies, they're still the revenue numbers are very tiny if you're not open AI. Um, but the market understands like look that I know the revenue will come. I'm not worried about that, right? Uh that's why all these companies that do random AI things are getting huge premium because it's like look, I I can see how big the opportunity is going to be. Um right now people are flabbergasted that Circle is trading at this crazy multiple on public markets. Uh and everybody in crypto is like you idiots, it's supposed to be worth 6 billion. We did the math. Here's our, you know, here's our multiple matrix. Um and what the public markets are telling you or you know, whatever. Obviously the the the the number there is very very

volatile but what public markets are telling you is that we believe that there's going to be a much bigger future for circle than what you guys see right now and you guys are too anchored to this stable coin crypto market and we see a stable coin global market which is going to expand way beyond you know whatever it's a trading pair on Binance. Um, now one of those two parties is going to be wrong and we'll see in the coming years uh how that's going to actually play out in practice. But there will come a time when people are actually excited about onchain use cases and onchain uh value creation and that's going to cause this again this disconnect between revenue and growth potential. Right now I think what what the market is telling you is that we

don't see growth potential on chain and that's okay. There are times when that optimism and people people no longer can see one or two or three steps to out on the chessboard. It's like look, I don't think there are any more moves left. I happen not to believe that. I happen to believe there's a lot more that's going to happen on chain than what we presently see. But the market doesn't agree with me right now. And that's why like the I I think actually Andy is right. Right now market is hyperfixated on revenue. But I think that will change if and when we see that there are more use cases and more growth potential on chain than what we're currently seeing right now. I think it's very sector by sector. So there's some sectors where you know if you're not focusing on revenue you're focusing on the hopes and dreams and the

butterfly butterflies unicorns. I think AI is so new like AI crypto we're still figuring out exactly what it is. So it has that potential to just trade at some crazy multiple something like DeFi that we've had now for a couple cycles. I think the revenue is expected like buybacks are expected like what Mike said like you're we were seeing the pump fund you know token they're going to have to do buybacks otherwise if it's just a meme no one's going to buy it that's the default and so some sectors like deepen you know I think we'll follow suit like you'll need some actual tangible um like economic model to come out yeah so just to quickly wrap up the uh I do I think even some of the best

founders feel like in the very very near term we're tapped out on the number of people who are willing to spend money and invest money and put risk capital on chain. Okay. Even the biggest institutional uh you know announcements that we're seeing are not like hey we're going to deploy a bunch of our capital into somebody else's asset on chain. It's hey we're going to issue this asset and we hope some people are interested. So we're in this like sort of pause where we're I think in uh inventing uh some really interesting things to put on chain. I saw Republic today is putting a SpaceX stock. Uh it's a really fruitful and exciting time to experiment with what can be issued. Uh it's not clear outside uh it's not clear that there's

demand to take the risk uh to put the dollars on chain. you're seeing is uh actually more folks putting money into, you know, the BTC ETF and the Treasury stocks and the circles and you just sort of dabbling in the theme uh versus the the underlying thing that will create long-term value. I think we're going to probably have to wait a little bit of time for that to start to wash out. And you're already seeing it, right? As a lot of these folks who are issuing the Treasury thing say, "Hey, we're registering our shares. It's pretty obvious we're going to sell over the next 6 to 12 months a significant portion of, you know, the great deal that we got in this pipe. Uh you're going to see people, you know, start to

not make money investing in these public equities that they expected to. And I think then you'll see folks really rethink what do I want exposure to on an underlying basis. Uh but right now onchain finance internet capital markets just isn't the thing that's like you driving the public's imagination and incremental capital. Bitcoin defi is heating up on ApdOS with nearly 400 million BTC assets on chain. Apttos users can now earn yield on their Bitcoin via EcoABTC and OKXXBTC without the typical bridge risks and high fees.

To explore BTC yield opportunities on APOS, follow the link in today's show notes. As always, investments in blockchain technology involve risk. terms and conditions apply. Do your own research. Jordy, you mentioned that you guys do both liquid and venture depending on the where the cycle is. Where do you see the returns coming um over the next year or two? Like which one? Um I mean we have been overweight Bitcoin just uh understanding that it's a time where macro matters a lot and it's just you know war happens it doesn't matter like it actually is better for something that you know can be outside of that whole system of war.

Um, the question is, you know, as Bitcoin becomes bigger and bigger, where does the sort of beta come from? Historically, it's just always been the usual like Ethereum will go first and then some other stuff. I think that's kind of changed now. The role of Ethereum is different. A lot of things are doing well on the app side now. Um, so there's going to be a lot of opportunities. We're very bullish on AI as a sector. Sure. I know Hib and I have like sometimes, you know, been less sure around like how much will come out of that. I think quite a lot will come out. Um, so I spend a lot of my time now, you know, listening to like all the AI kind of labs and what what's coming next and

trying to figure out what what could be uh linked into crypto products and how will that So, you're buying Bitcoin and listening to AI podcasts? Yes. Got it. Good good good recommendation. You're actually going to do really well this year. very normie stuff. I mean look there's unfortunately like some of the age-old stuff that even though the industry is now like quite mature in some ways we haven't figured out the some of the basics you know why is onchain internet capital markets not huge like one thing is you know we had this by bit hack it just goes to show you that keeping money on chain there's just like these tail cases where you think you know this thing is super safe and and but somehow they've changed the front end like there's things that still

happen that um give like large institutions pause on like keeping billions of dollars and trading their stocks on chain and then on the other side you have these L1s where you know we talked about DeFi like you know that's starting to figure out its its value acrruel L1's are still very nebulous in terms of like what is the what is the value of like an L1 and and like there's people trying to talk about like economic capture and me and different things but none of it is really concrete and um you know a lot of the venture bets for the last 10 years that have you know poly chain from the beginning has done this for many cycles is just you know you bet on L1's and L1's always have this like hope that someone's going to come and build

something on the L1 and it's going to acrue value everywhere and it still kind of works I I haven't said that it is fully broken but maybe it's not as easy as it used to be to just bet on L1's should we do this again should we run it back I I I feel like we've already done this to death I don't know if we can I don't know if there's anything more to say. Yeah, the returns are decaying, but Monad and Mega ETH, what do we think? Yeah, there's I mean, look, like I I was I I've said this too many times. You're probably all bored of me saying it, but like blockchains aren't done. There's still more to do. Like, there's no way that in the year 2025 or in the year 2021 when Salana was first launched that it's like, "Yeah, that's it. We figured

it out. No more innovation. We we there's nothing more to do." Um, some of these things get upgraded in place. I mean I was just talking to somebody about um you know Salana trying to go to multi-proposer and it's going to take like three years you know it's going to be it's like so P like Salana is becoming Ethereum and obviously Salana people really don't like to hear that because they think Ethereum is the enemy but like you you eventually grow up and become your dad like that's just the circle of life and Salana is becoming its dad and Salana people will eventually get like there will be the next generation and they'll look at Salana and be like oh Salana's so slow and governance is [ __ ] broken and nobody can get anything done. It's so political and everyone's already gotten rich. All that [ __ ] it will it will

just happen again and again until eventually it stops and we will all retire. Yeah. So, I would say we've so if you had asked me even when we were on stage a few months ago, uh we had never in the history of 6 ventures invested in you know base layer chain private in private markets and and we have since since that time. Uh so our our I don't know if it's a thesis but like the things that interest us is well one you just look and the returns in investing in uh infrastructure have been really good and what we've noticed is uh a move towards sort of apps launching first or use cases launching and then chains later right Athena ando you know bunch of

others um hyperlid so we invest in plasma you know I like to call it tether's blockchain I mean that's a thesis I'm not overthinking it uh but a really good UIUX and economics around issuing moving uh and interacting with stable coins. And then uh we like to do the barbell thing at six investors. We invested in Doge OS. I love memes. You know, we're investors in pump. I uh worked on launching Bonk. Uh Doge OS is uh L2 on top of Doge. And it's like, hey, you know, let's use that community, that energy, the mimetic um fervor of of folks around the Doge meme and launch apps around that, you know, things speculative apps, news apps. Uh you can I can't even imagine that things will happen. So I agree with the se, but we pro, you know, I think you're investing

more in like hard tech around, you know, what's happening. Is yours soft tech? What is your tech? Uh mine it's not bar chain, please. Um, but it's more around L1's that have sort of app specific and community specific use cases or L2s. Sorry, I just throw I L1 L2. So if you if you're building soft tech, you know where to go. I actually like that. I might start using it. Um, Mike, with regard to the stable coin chains and the RWA chains, what do you think they need to be able to succeed?

Yeah. So, uh, well, I mean, it's a really broad question, but, uh, so I think on a very basic level, it's crystal clear that stable, stable coins, uh, are a 10x, you know, better product than, for example, you know, holding a balance. Like I worked at Venmo 12, 13 years ago and you kind of had this like Venmo balance and it was held within one siloed app and then like hey maybe they do integration with some merchant and oh I could spend you know my Venmo bucks there. Uh stable coins just completely reinvent that, right? Uh so they're a better product and that they're open, composable,

permissionless. Well, in some cases there are some permissionless ones. Obviously there are some like USDC and USDT that aren't. Uh but you have your choice of what type of table you use. they run in open networks. um they're very cheap and so that primitive is is better for example than you know you just look at banks where like I have to the banks have to face each other for my money to move between them you know if they're moving internationally there might be multi-steps there's a lot of latent capital that has to sit for you know my money to move from the US to Pakistan people have to keep reserves it's very inefficient you can solve some of those things with stable coin over time we haven't solved them today I won't say that we have but we have a promise to do that. The networks are 10x

better. Uh the product itself, as I said, is 10x better. It's not, you know, Venmo bucks. It's open bucks that anybody recognizes anywhere. Uh and it can, you know, generate yield that like and it's extremely and hyper competitive and easy for the next, you know, incremental person to issue a stable coin. Uh and therefore, uh it sort of pushes people to give the benefits to the end customer, whether it's a business or a consumer. Um, and you know, there's sort of settlement that's much faster. But the point is this stuff only works if we can create an economy where like people want to keep their money in the very long term in stable coins and there's things to do with them. And that's starting to

happen. Uh, because at the exit points, uh, they're not as relevant today, but like you can use a card to exit from stable coin in the local market that you're in. Uh, and we're going to get better products than just cards. you know there's in Pakistan there's people in every individual market you could think of that are creating better off-ramp products and it's going to happen much much much faster than you would think the next thing is machines are starting to transact at an ever you know faster rate machines can transact in stable coin it's really difficult for them to transact in fiat you know there's fees on every transaction there's a million reasons why it's more difficult so stable coins are going to go from 100 what is it 250 billion issuance to many trillions like

literally in the next 24 months It's going to blow people's mind. Um it's not going to acrue as much value to Circle as people think. that's going to acrue value to a whole host of other applications in aggregate I believe um around them and it's going to acrue value to the end user and it's going to lift up a number of businesses from operating you know not for example in dollar denominated not having access to you know uh if I had to wait 120 days to you know settle my trade or you know settle my invoice or settle my wholesale bill I'll be able to do that you know maybe in one quarter of the time. Uh, sorry I'm all over the place with the answer, but the point is I could keep going for probably another 30 minutes

and give you a bunch of reasons why they are a fundamentally better product. Netflix versus Blockbuster. It's this is a Netflix moment for money. Buy circle. That's basically summary. Yeah. Yeah. Joe Joe Mckam would agree with you. With nearly 400 million BTC on chain, supported by a secure, fast, and affordable move VM environment, Apptos has become the BTC growth chain. Apttos users can now earn yield on their Bitcoin via ECOS's ABTC and OKX's XBTC without the typical bridge risks and high fees. With support from Echo, OKX, and Stax's SBTC, Apptos is the place to go for best-in-class yield on Bitcoin.

To explore BTC yield opportunities on Appdas, follow the link in today's show notes. Algoit 3.0 0 includes all the tools and open- source resources you need to unleash the power and performance of the Algrand blockchain. Access a resilient network with instant finality, consistently high throughput and low fees, and layer 1 features like atomic swaps and native assets. No smart contracts required. Whether you're building D5 protocols, enterprise solutions, or nextG DAPs, Algo Kit has all the tools you'll need to build faster, safer, and smarter. Start building with Algoit today by following the link in today's show notes. As always, investments in blockchain technology involve risk. Terms and conditions apply. Do your own research.

I feel like we kind of mentioned um in between the questions, but what sector specifically are you guys most interested in going forward? Mike, obviously stable coins, let's see. What about you guys? Did you say besides stable coins? Yes. Wait, I can't give a 10-minute speech about stable coins. You can. Okay. Um I think so. One of the things you mentioned was uh was cards. That's actually something we're looking at. Um, so there there if you look at crypto cards like credit cards, there's there's two kinds. Uh, the kinds that probably people in this room maybe think of like, oh, Coinbase has like a credit card. Um, it's it's like not really a crypto credit card. It's like a credit card like gives you rewards that are relevant to Coinbase of like, oh, we'll give you Bitcoin rewards, but it's kind of just like a different skin on like, okay,

you're instead of getting, you know, airline miles or whatever, you're getting Bitcoin and, you know, hopefully Bitcoin goes up. Um, these are fine, but they're they're not that interesting. Uh what we're seeing actually a lot in emerging markets is that these crypton native cards where they're actually drawing against an onchain balance that's getting paid in real time and then settled through the stable coin processor. So Bridge has one of these. There's another company called Rain. Um there's there's a few other companies are doing actually Marquetta which is a you know kind of old school traditional um uh payments company is is also getting into this game. And this is growing like gang busters in emerging economies because you know if you if you go to a place like you know you go to Argentina or Venezuela or you go to you know Turkey uh there people will

actually take a crypto payment. Not everywhere but in many places they will take a crypto payment. Um and usually they're using Binance Pay. So Binance Pay right now is the biggest in the world. You go to any emerging economy it's mostly Binance Pay that people are using. And uh you know you scan a QR code, you pay somebody on Tron or whatever it is uh and you and you pay them some tether. um these credit cards like if you go to a place that's basically like not a failed state and you want to pay with your crypto, right? Like let's say you're in um India or you're in China or you're whatever and you have some tether and you want to pay a normal person at a normal business.

These crypto credit cards are the only real way you can do that endto-end loop where you have stable coins onchain in a wallet or a fintech that you're interacting with has stable coins, but you're still connected to Visa Rails or Mastercard Rails such that you swipe a card and your onchain balance gets deducted. Right? To me, that is the only way that I could see stable coin payments actually really going mainstream where you have your balance on chain and because of the way the crypto works, you can like instantaneously swap whatever asset you have on chain. So you don't have to have Tether sitting around or circle you know USC sitting around. You could have you know some weird exotic memecoin or whatever and as long as you can swap it instantaneously um and you know settle and and you know the company that's uh

issuing this card knows that this collateral is good for it then you can do all this in real time right like my experience with a bank is that like my bank doesn't do [ __ ] for me unless I move it there or move it into a money market or go and do something smart with it. I have to manually go in and say, "Oh my." So, who's putting the capital there? You are putting the capital there. And why am I doing that? Am I earning a yield? Like, why do they have all this latent capital that allows them to immediately uh you know, sort of settle? Oh, so sorry. So, the depending on how these companies work. Um, so for example, Rain can actually settle directly with Visa. Okay.

Um, using stable coins. Most of these companies, my understanding is that uh they have to have some kind of capital buffer such that they can see what collateral you have and they can predict your spending patterns and say okay I need to have at least this much cash like USD that I can then settle with Visa in order to make these payments. Uh so these are less capital efficient version. I think within a few years, everybody's going to be settling with Visa directly in stable coins. Y and that is going to allow you to have an onchain balance that's earning yield in real time. And the moment that you spend money, it gets deducted or swapped or whatever in order to actually spend money at a at a retailer or whatever it is. And this is how you connect the old world and the new world, right? What

we've learned in the history of venture is that making people do a new thing just [ __ ] sucks. Like it just doesn't work. No one will do it, right? So like, okay, if you're in an absolute disaster situation like in Venezuela, then people will figure out how to use Binance Pay, but most places in the world people will they just won't. Merchants won't do it. Um, so by piggybacking onto old infrastructure and kind of overloading it, to me, that seems like the most plausible path. And the reason they're able to avoid sort of like the chargeback and credit risk is because the stable coin is a bearer asset and they can see my balance and they have access to it.

Correct. So the chargebacks are normal because all the credit card companies Yeah. the chargebacks still happen but uh the chargeback happens from you know claw back from the merchant and the merchant gets punished and you know that's the way credit cards work. Sorry that's my that's my stable coin speech. Perfect. Jordy. Yeah. I mean um I think Ether is doing quite well now with their card product letting you spend your ETH. Obviously, Crypto.com had this like period where everybody wanted a Crypto.com card cuz you're getting a bunch of uh rewards on it. I think you could spend your crypto and now Mantle is working on something and they bought Fiat 24 which is u like a real bank in Switzerland and you can settle there and they're developing into something called your bank. So that's that's just been announced should be

very cool. Um like I said, I've been listening to the cryp um AI podcast. Um, you know, we've invested very heavily across like a lot of crypto AI companies and just have this belief that this is a sector that over over the next 3 four years will continue to like flesh itself out. Um, one of the things we're seeing certainly is around uh memory and personal data. right now chat GBT just gets all your history and then you know at some point that'll be a massive mode for them as they'll be able to actually do useful things for you using all that data that they know about you. Uh but I think the element of like owning your own private data and your history of your chats is going to come into play

and history uh is something that you know you can own through crypto ultimately like there there'll be some form of ownership that you can uh you can have and you'll still be able to have as good a product that will cuz I think the quality of product without having any history about yourself and just you know opening a new chat GBT chat is going to be very different than you know I know everything from chatting with you for two years and like something much more personalized. So I think crypto is going to play a role in that. Um some companies that we've invested it are working on that uh category and you know we look at everything all the way to like the biggest things like

what is money and what is bitcoin and sort of what will be currency in a world of you know digital intelligence. are going to be some kind of intelligence currency. A lot of people are thinking about this and some are even, you know, trying to develop intelligence currencies. Um, so we remain open-minded obviously like I've always been quite skeptical that you can just devise money out of thin air and just create base money. That's something that many people have tried and failed. Um but in this specific situation where the world might change so much from um from a new form of intelligence, I think a new type of currency might come into play. So there's just like the broad spectrum of

uh of crypto AI that's very interesting. Why do you think crypto AI hasn't done as well, I guess, as people imagine it would have done over the past few years? I I think it's already done too well for how much is real. Like if you look at you know there's a lot of people love bit tensor and it's it's an interesting coordination mechanism but um you know the work that's being done on these subnets is not necessarily um you know worth billions of dollars yet let's say uh and then you have things like virtuals which is still doing quite well I mean it's it's kind of back in the billions of uh valuation and that I would say is more of um you know an experimental ground where you're people are just launching different

things and Um, so it's done quite well. I I wouldn't say it's not done well. I think what maybe you're expressing is like what we seeing across Chris all of crypto, which is in each category there's a lot of people building and there's like, you know, a lot of them that fail and a lot of them raise money. Um, it's just a little bit saturated. So like even stable coins, I mean stable coins maybe like there's a lot more to do, but each category you look at you know what people have figured out is like there's a bit of an arbitrage building in web 3 versus web 2. So you just get like a lot of builders trying to take their shot at, you know, this category and one of them of course crypto AI and there's just been so much funding thrown out at like you know

people trying stuff. Um but I think that's across crypto. Makes sense. Mike Hib, do you guys have any what are you guys focused on in terms of crypto AI? Is that something that's still interesting to you? Yeah, so uh so to your original question, I mean there are successful I mean it just depends on what what you define crypto AI as. Like worldcoin is wildly successful. I think has to date we're not investors but has exceeded you know my expectations in terms of number of people signed up uh market cap and you know it's a very specific use case which is you know identity human identity in a you know machine transaction and interaction world uh you have things again I don't have exposure

to it but grass uh you know so sort of like decentralized I mean they they've moved beyond the original thesis but you know gathering data and information uh and a and a distributed manner that you can't do in a centralized manner for a whole variety of reasons. You'd be blocked this that the other thing. Um and so there are a number of areas uh and so like this whole category is only a couple years old, right? Like like the Silicon Valley version of like foundational models is is you know call it 15 years old. um maybe expanding a little bit, but I do think what is happening and the amount of value and and novel uh insight honestly should talk about this because I'm in like a

Telegram group with him where he's a third of the comments and probably twothirds of the things that I learned. Um so I'll kick it over to him very soon. But I would say crypto AI to me has exceeded expectations in terms of value. It's just not the stuff that people are talking about. It's not the the virtuals platform AI agent tokens that are useless, right? It's um things like we're finally starting to see like with uh Plurales and a bunch of other companies decentralized training start to actually occur in an interesting way.

We talked about decentralized data gathering. We talked about um you know human identification verification. Uh there are a number of use cases that uh we have phenomenal entrepreneurs and founders who could go out and work at the open AIs of the world but have chosen to say hey just like people who could have gone 15 years ago or maybe 10 12 years ago and worked at big banks or big financial institutions or hedge funds but chose to work on Bitcoin or Ethereum we're having that right now in AI. Um but it's not the stuff people are talking about because they're not launching tokens. They're not cynical.

They're actually being really well funded so they can build things over a period of 2 to four years. Research, do really deep research, use the money they're raising on things like compute, hiring, uh, and growing, you know, their fundamental thesis and model such they're ready to expand in a world where, uh, you know, the folks who have been at it for a decade are raising 500 million capital, you know, every six months. Okay. So I will take the counter point from these two gentlemen that I think crypto AI is obviously not working and I think anybody who's like really clearly looking at what's happening in the space is like no one's using any of these products. They're not solving any problems. Like there's a lot of founders groping in the dark trying

to find something that anybody will give a [ __ ] about. And if you talk to actual AI people, they will you will tell them, "Hey, what crypto AI products do you use?" And they're like, "What the what are you even talking about? Get out of my face." Like let me get back to my podcast. So like so let's take let's take some of the examples that were that were just mentioned right so we're talking about things like um uh worldcoin which is an identity mechanism uh who here has seen anybody using worldcoin to solve an identity problem to do what what's that using it to do what solve an identity problem who has seen world coin nobody right I I can't name any I can name a lot of people who went and got their eyeball scanned in emerging countries who got paid to do it

but uh right Now, nobody at World Coin would tell you, "Yeah, Worldcoin is working, right?" Because they they're not stupid. They understand what it looks like to be at a startup that's doing well because they're all ex startup people who are who had products that people actually wanted like right now. So, we have a stable coin startup that's in the sort of world coin app store and is actually acquiring users because they're actually verified humans and not bots. Like, so which app? What's that? Which app? What's that? Off-R. Offramp. Yeah. Yeah. What is Stablecoin? It's a It's a stable coin banking product. Has tens of thousands of users. Tens of thousands. Yeah. But like but Worldcoin has how how many wallets? Millions. Tens of millions.

WorldCoin has about like 10 million verified users. They're um they're scaling like the rate at which so it's like millions a month, right? I think what you're missing is that these these things are not going to be useful until there's a network effect, right? Like no one's using Facebook if it's just like Right. Fair enough. So I'm not I'm not arguing that that WorldCoin will never work. That to be clear that's not what I'm claiming. What I'm claiming is that it's not working right now, right? something like pluralis is decentralized training. Um same Jensen there's a bunch of these that are doing decentralized training. So uh training and where where anybody can add their GPU and train a model together. Uh right now decentralized training is extremely expensive.

It's very very slow and you cannot do the largest state-of-the-art models in a decentralized setting. Um people in the regular world think it's cool. They're happy to talk about it but like is anyone using any of the models? Has anyone even tried any of the models that were trained in a decentralized setting? They're not that they're not at the point where they're trying to get people to use the models. They're at the point where they're trying to actually decrease the cost 10x every totally totally totally but these things will fundamentally never be as cheap as doing it in a centralized way. And the decentralized training people know that. They don't claim that it will. They've never claimed.

Right. But using Bitcoin is not as cheap as using fiat currency. Correct. Correct. Correct. So what they're hoping is that people will care. They're hoping that people will care about the way in which your model is trained. So look around. Do you know anybody who cares the way in which their model is trained? Not today. Have you met such a person? Right. Okay. So, here's the So, here's here's the fundamental issue that I have with a lot of things in crypto AI. So, I want to be very clear what I'm saying. What I'm saying is not nobody should be building in crypto AI. I'm not saying that. I'm not saying that all these startup founders should go home and pack up their bags and whatever. I'm saying they should be very clear that their startups are not working right now. They should be they should look that clearly in the eye and

understand your startup is not working if you cannot clearly describe what problem you are solving and for whom. Okay. Um, usually what happens when somebody is right about the thesis but wrong about the timing is that they just lose. Period. Right? Almost always when you are too early to a market, you are wrong and you die and you fail. And even if you're eventually right, if you're two years too early and eventually the market that you were going after materializes, you are not going to win the market. The person who started right now, the moment that it's clear who the customer is and what you need to build for them, that person wins. you don't get to win, right? So if you're if you're five years early on VR, if you're way too early on AI, if you're way too early on whatever, almost always you are

not the winner almost always. So this is just a universal law of startups is that startups are learning machines. That's why they exist. They can learn faster than big companies can. Even though they have fewer resources, they have fewer people, they have fewer customers, they have less distribution, but they can learn faster. So two things I would say that the learning faster cycles are getting tighter and tighter and tighter and tighter. So it's not like 10 years ago when you know I I think like what you're talking about I don't think these companies are three years early like they might be 3 months early like the the dude this is this has been going on already for like two years how are they three months early because things are accelerating at the like literally the like right now they're accelerating absolutely like Mike Dudis is actually

using like chat GBT you know it's impressive this boomer open AI is accelerating yeah openai is accelerating but like look here here's what I'm saying is that there will come a time when agents are good enough to come on chain. There will come a time when decentralized training becomes really cheap or personalized models or any of this stuff will really start working and like you and I will use a personalized model that's entirely so small and so efficient that it can be quantized and put on our phones. But those don't exist. The agents don't exist like these things are not on our phones. None of this stuff exists. And until it exists, you cannot build these products yet. You will build the wrong thing. I guarantee you, you will build the wrong thing. And once the things happen, that is when the startup learning loop can actually take place.

But until you have a customer, until you have a feedback mechanism, until you actually know, am I building the right thing or the wrong thing, you will build for retweets. That's what most people in crypto are doing right now. Now I I put something like virtuals in a different category. Virtuals knows what they're doing, right? They're building a meme marketplace. It's memecoins. They know exactly what they're doing and they're doing a great job of it. Good job, Virtuals. But I don't put them in the same category as the other people who are really trying to build when they don't understand what their customer is going to look like when it arrives. That's my problem with crypto.

I I mean I think um I'm much more patient and I think the companies we've invested in, we invested in like, you know, Prime Intellect and Eternis. These guys have moved to SF. They're building their teams. They're getting talent from the AI labs. They're interacting with AI labs all the time. They're figuring out, you know, where the puck is going to be in a year. I think what Mike said like I'm feeling it as well like there's a hyper acceleration happening because of like how quickly this technology is is progressing and so yeah it is like 3 months it's 6 months like we're talking about like there will be things happening very quickly and I would rather be in there um even if you know there's there's not like use cases right now although I can think of one use case so you know unfortunately some people

have been getting hacked and it's been through these zoom calls and and and Google meets calls and apparently you One of the uh people that I know got hacked with uh I think it was like a dragonfly fake VC call. It's like an AI Dragonfly. I think it was like an AI hase. Literally me. Yeah, I think I think it was it was it was meant to be like a pitch call. They they thought they were they were pitching to Dragonfly. You know, it was some AI. Now, if they had worldcoin and they had world verified hive ID, I think, you know, my friend would not have uh gone drained. So there's still hope for use cases.

The world coin needs to verify all the cryptovcs and then we'll be safe from fishing attacks. If you know I think this is the real here so we're okay. But if it was digital we wouldn't know and worldcoin would have solved that. So we're also investors in worldcoin in case you missed it. To your point though I think part of it also is just how you explain what your goals are. And I think a lot of these folks are trying to project that their models uh and their products are being used today by world-class teams. And to your point, I don't think it's the case. Uh a lot of them are still research experiments that are trying to drop, for example, the cost of decentralized training uh by 10x every number of months. If they're clear with the market about what they're doing, they could be

succeeding. But by saying, "Hey, we're going to compete with centralized models, uh, you know, that narrative just doesn't ring true." And that's where I mean, and this is the thing. This is crypto. The easiest person to fool is yourself, especially when you have a token and the token's going up and retail's buying into it and you're getting all this stuff about you. So, like, as long as you are not deluded yourself about what you're doing, if you understand, look, this is a research project, great, awesome. Go do research. I'm glad you're getting funded to do your research. But if you believe your own hype, you're getting high on your own supply. That is the formula for just abject failure, right? And the the the example I can give uh I think of of uh that that will resonate is like look I there were so

many people in the early days of DeFi that that that we funded that were luminaries, brilliant people who broke open the field and solve fundamental problems and Hyperlid arrived 10 minutes ago and just ate the whole market. Why? because they knew exactly who they were building for. They knew the customers were already there and they could just build the right thing. And that is how almost every market in history works is that once you know exactly what to build, the the the person who can take the straightest line toward building the right thing wins the market. So the first search engine didn't win, the first social network didn't win, the first uh DeFi exchange didn't win, the first PES exchange didn't win. Um I think it's very likely that the first crypto AI thing isn't going to win.

And the second big smart contract L1 is going to win. Salana. Fair enough. I should have brought up AI a lot earlier. Um guys, thank you so much for coming on. This was fantastic. Do you have any parting words, but do you guys have any parting words for the audience? Maybe more positive AI takes. Damn. Yeah. All right. So mine would be be honest about who you are, what you are, what you're building. Uh you know we we talked about crypto AI and overpromising and uh what we've seen a common success theme is uh folks were just deeply honest. So you know I I'll just use the

canonical example pumpf fun. Hey it's a token launchpad. It has trended towards memecoins being launched. It is tended toward human emotion around liking, enjoying trading, enjoying real time, enjoying frenetic, you know, permissionless casino gaming, gambling, and they're generating a lot of revenue. And they're continuing to do that despite people saying day in and day out, uh, this is wrong, this is extractive, we hate it, but human behavior is saying they love that. you know, hyperlquid slightly different, but you know, per is definitely not a market that most people are playing in and profiting in on a regular basis, but they've found product market fit with something that people love. um seek out those behaviors if you want to be

successful uh that people love because it's increasingly clear that we're moving away from commodity, you know, blob and block space, uh being valued at astronomical multiples and we're moving towards uh you know, products that people love and and use as part of human human DNA and and they're willing to pay for those um positive stuff. I mean, I think the capital formation game is kind of nearing its equilibrium. We're seeing even like ICOs come back and Echo is kind of doing some platform now. So, it's very interesting. It'll start, you know, making the VCs not feel so bad having to defend themselves cuz, you know, people will be able to if they want to get default capital, they'll be able to access it. If someone's adding

value, it'll be clear. So, um, you know, I'm happy that we're we're kind of nearing that equilibrium soon. And to end on a positive note, I love you. [Music]