Oneshot: Tempo Mainnet, Agentic Payments, and SEC Crypto Rulemaking | Roundup

Transcript

But I think Stripe is kind of hedging their bets. They're like, "Okay, if it's stable coins, we got Tempo. If it's credit cards, we got Stripe's core business." But we know agents are going to utilize us. We just don't know how. Nothing said on OneShot 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. Michael, Vance, and our guests may hold positions in the company's funds or projects. Disgust. All right, let's rip it.

>> Phone's off. >> I got money on the phone. >> Hands and legs inside the vehicle. >> I think we're going to make it. Um, how we doing? >> Pretty crazy [snorts] week all around. Seems >> seems like uh a lot of stuff is going on and some things are being paid attention to. Should we get straight of Hormuz up on the screen with the live tanker tracker? >> Live tanker tracker. No, I don't think this is a uh international relations podcast. I think we we can masturbate as a macro podcast, but I don't think that that is uh necessarily our forte.

>> Let me ask you a question. Do you remember when we almost hired Peter Zen to come in and teach us about global macro during the Ukraine war? >> And then you found the the clip of him saying that Bitcoin is going to go negative. [laughter] Well, I was actually thinking about that and maybe just to tee it off. Um, all the crypto we bought um kind of at the lows during the Ukraine war. Do you remember how stressful that was? >> It was uh I think the line that we finally came to was either this industry is completely over [laughter] or it's going to come back. And it did come back and it come back.

>> I think we are in a similar period of time right now. Uh there's so much prognostication going on about how this war uh is being prosecuted. But I I mean you win by killing the other team's leadership, right? Like on a literal sense. And they've basically killed everyone um them being the Israelis. And it now feels like there are some tactical strikes on these energy sites that are tit fortat. Um, and you know, the next leg of escalation is maybe like the dissalination plants, but I don't I don't know if there's a ton of desire to go there. Um, and then the last thing I would say is like the 20 million barrels a day. Very scary figure. That's like 20% of global oil consumption per day.

That's really kind of more like six. You know, they got the East West pipeline going. There's talk about tankers getting escorted. So, like I think overall it's hard to know how these wars turn out, but it feels like we're on the tail end of this in terms of the shock and on escalation >> and and I think um there's a bit of overfitting happening as well, which is and recency bias or just like the Overton window uh of everyone's memory is definitely a real factor when it comes to market psychology. And I think most people in markets right now are were were in markets in 2022 when we had the last shock. But what you know that represented was a Ukraine Russia

conflict leading to an oil shock leading to inflation. Now how much can we strip out of the inflation effect that was actually because of oil? It was definitely some. But if you remember at the same time the Fed Jerome Powell was saying well it's transitory because it's a it's a supply chain constraint. It's this is like the residual effects of COVID. We can't get stuff, you know, shipped here. This is when people were were following Ryan Peterson of Flexport talking about how many tankers were going into the port of LA. Like that that's what was going on at the time. And so I think that there's a bit of like you need to separate out the the variables that contribute to the effects and and oil was definitely one of those.

Now, was it the overwhelming majority of what led to inflation? No. and this whole market sentiment of like not only are there going to be zero rate cuts this year, there's actually going to be rate hikes now as part of part of the narrative in in Europe. I don't know if that's come to the US. US >> that's what that's what people look the the Twitter uh um PR Newswire is is saying, you know, rate cuts are now rate heights are now on the table with Jerome Powell potentially staying if there can't be a confirmed new Fed governor or Fed chair. like all of these things are just like this this is this is peak bearish for for kind of like macro sentiment right now. Um but I think my broader point is people are overfitting

this oil shock uh leading to directly leading to inflation. And and so what I will say is there there was actually a really good analogy that I heard or or comparable that I heard which was the 2010 through 2013 period where the average price of oil was $95 which in 2026 terms is like 125. So, not only is it 95 and we're at 95 today, but like in today's terms, that's like a it's like as if oil was at 130, 125. And so, what also happened during that period of time, which was sustained high energy prices, we had much higher consumption of oil for energy back then than we do now. But we also saw a rip roaring stock market and over those three years, four

years. Um and and I think that we're kind of overfitting this narrative from 2022 that oil shock and and this is not discounting the fact that wars are bad and like an escalating war with a large Middle Eastern country is where the US and Israel are are kind of contributing um simultaneously is definitely not a good thing from just like a a international relations perspective. But I think the overfitting of this oil shock leading to inflation leading to like calamity from a market's perspective is is just it's it's not accurate. Um or it's not as accurate as people are pretending it is. Um so that's that's the general take.

>> Yeah. And and and yesterday was I think Bay 16Z had a good tweet with yesterday everyone kind of threw in the towel admitted it was over. Um and and you know it fully now is there's no cuts priced in um for the US. The Europeans are considering hikes at the next meeting. Um and so we're now at a place where European gas prices have spiked 30%. Like the reason you hike rates into that is to destroy demand. That's like the only lever they have to pull to do that. Um but that's all been priced in. So you know the you're also like oil prices increasing 20 30% also destroys demand. So you're going to destroy demand to further destroy demand makes absolutely no sense.

So, you know, in terms of the I don't know there's there's not an extreme fear or greed gauge for global macro, but when you see, you know, the Bob Elliots of the world, the Ray Dalios of the world, uh the Rory Johnson's of the world, these are like the KO oil people and like the Flexport guy. When you ever you see him on CNBC, you're approaching a geopolitical top in terms of like the stress levels. So, you know, I think it's similar to, you know, what we were talking about in the last uh shock that we saw in 2022, which is like we've been holding our noses and buying. I think that's basically all you can do. Um on Go for it.

>> Well, I would just say like from a psychological perspective, the best buys that have ever been made either, you know, basically at framework are the ones that feel the most painful. [laughter] >> I remember telling you just press the button. But it it's it's this like it's this weird feeling where you're like I don't know if this was going this is going to be an amazing decision or this is going to be a terrible decision but like I I don't feel I don't feel happy making this decision right now. That's usually the best time to be buying. I will just say and I I think this is uh with background of crypto I I always think of crypto as like the the money that is so dumb is that that it is smart. It front runs things

without knowing them both ways. And you know, crypto had an off day today and it's Thursday, March 19th, but it's been hanging in there relative to even things like gold. Like gold broke 4,700 today. So that's almost like a 20% retracement from from peak. Um, and so crypto, I think, has a constructive background led by the majors, led by Sailor and Tom Lee, led by STRC and the ETF inflow is turning on. Stable coins is a broad narrative. fixing quantum. You know, we've heard behind the scenes there's going to be some serious efforts to at least put out the narrative fire and then put out the technological fire whenever it does start. Um, and you've got a lot of people on the sidelines that have sold the 40-year cycle and that are theoretically waiting to reby.

Um, but it does feel like we're starting to really put in a convincing low for crypto. Um, and 59, I think 50 was was the absolute low, but these things are positive. you have, you know, positive tailwinds with increased circulation of stable coins, S&P partnering with Hyperliquid. Um, and then this new regulation that I guess the rule making that's come out. >> Not a regulation, >> regulation, it's rulemaking. And I mean, why don't you why don't you talk a little bit about what that is? >> Yeah. So SEC came out um kind of like completely under the radar in terms of and I can't remember if this is a day ago or two days ago but it like I don't even remember you know what happened

exactly because you know Iran was taking too much of the narrative away. Um but essentially what the what the SEC came out is provided it came out with was providing a clarification a rule making around the uh classification of different assets in the crypto space and which ones are likely to be commodities which Paul Atkins just came out and flat flat said most crypto assets are not securities. Um there is also an understanding of which ones would be considered securities which lo and behold the one that he identified is tokenized equities will likely be securities. If you put it on a blockchain and it's the Apple stock

versus on the NASDAQ it's still Apple stock. Everybody should understand that. But generally I think um we're moving in the direction where at least for the next four years this SEC is going to be in place and it is probably assumed like it is it is not impossible to get the uh rulem changed but once rules are made I think it's probably going to be pretty difficult for future SECs to go back and say actually what was said by that previous predecessor of mine was absolutely incorrect and we're going to go 180 degrees the opposite direction.

And so I I think what basically people are hoping for with clarity from just like an implementation perspective is is now been has now been declared the the kind of going standard rules set out by the SEC or or the vast majority of there there's there's large elements that have very specific applications. to the stable coin uh yield question which I think is the biggest sticking point that has big implications for Circle and Coinbase. Um but I I do think that broadly from just like a a system or a universe perspective of assets. The SEC basically came out and gave us clarity this week and nobody seems to have paid attention to that. Um so I I think that this is

obviously extremely bullish. Um, I think paired with the partnership of S&P 500 and Hyperlquid showcasing that there is institutional interest to bring these assets or bring these indices and tokenize them on chain like that is a fundamental I I think unlock as well. Um, you've got Mastercard buying BVNK for I think $1.8 billion. um which it had been rumored to be in the works for a long time and then it was kind of like well we didn't hear anything so we didn't expect it to close and something's died on the vine at the 11th hour but this one did close and also paired with you know the whole launch of tempo which we we should probably talk about as a whole entire subject but just

like agentic payments feels like you know the the narrative of stable coins is starting to move in that direction um at least from just a adoption or or institutional kind producing of platforms and frameworks. Um, so yeah, it just feels like an extremely bullish week for crypto, but obviously no one is paying attention and if anybody was like there are so many instances in the last 6 12 months where it's like if this happened in 2021, we would have just blasted off into the stratosphere. Um, and so I I it's a tale of two cities once again. Um, I think, you know, every asset is going to have its 15 minutes of fame, uh, when it's on

Hyperlid. And, you know, just like watching oil go from 95 to 120 weekends ago was like a watershed moment for a lot of people because they had to do it on Hyperlquid and those screenshots get shared. Um, and you're going to have that play out with everything. And, and I think, you know, the guidance is very bullish. I I I think one of the more bullish things that we had this week was talking to a team of builders in Asia uh who really fundamentally believe in hyperlquid and per as a concept as something more fundamentally fairer than uh zero DTE options or even holding spot. uh there's so much belief in these underlying concepts as well from the entrepreneur side that like at the end of the day that's that's kind of what we're missing relative to the AI side of

the house which I would say maybe has too much of that where it's just like people just like believe in this stuff and like believing in financial concepts often comes with a uh negative connotation um but I think a lot of where we see people going is just more focused more narrow and you know we uh were in a group chat together Michael this morning and And I and I'm just trying to pull up what this person said. Um it's like, you know, before crypto was legal, people could raise hundreds of millions of dollars by selling tokens. You know, after crypto is legal, you know, you can still sell the tokens, but there's no interest. And I think this is kind of like a false dichotomy in many ways where look at Hyperliquid.

You know, it's $42 billion FTV was launched within the last two years. It's team of five. It's team of 10. Like those outcomes are still possible. What is not possible anymore are the hey we're worth a billion because we're worth a billion because we're worth a billion like it's just like not happening anymore and I would say that's playing out in the AI side. Um but that is so fraught in so many ways like the neolab funding and and I think there's good neolabs and bad ones but the vast vast majority of them I would say these are the old L1s of last cycle and it's going to be down only if and when they ever get liquidity so you better hope that someone buys these things. I wanted to double click into this too because I think there's a lot of interesting parallels but you know I we've talked

about this I can't remember if we talked about it on this podcast but um in 2016 2017 uh well specifically 2017 with the ICO craze I think you know that I I think of that as being basically a recognition of um companies staying private longer like all of the interesting tech companies in 2017. I mean there were there was like Uber and Airbnb and and some you know consumer application companies but the vast majority of you know if you worked at a startup there's an 85 90% chance that you worked at some enterprise SAS company which is probably just not like the most sexy or interesting place to be. And so this whole like ICO craze was brought on by people in there's a a latent amount of attention and there's a

latent amount of capital that is looking for opportunities but also looking for interesting new places to play. you know, speculation, gambling. And so I think that that, you know, was probably how you could look at a lot of the attention that it it garnered. And you had this ability to have like seed or series A level, you know, return pro profiles of these tokens that would happen in like months or weeks, um, sometimes days. And so it's just like this this thing kind of build on top of each other. And I think what then occurred is a period of time where in 181 1920 you had basically the start of the crypto venture industry which where you know paradigm got started I think polychain was maybe a couple years

before we were in 2019 um but like basically everybody got started in that like 2 three year period and that was in recognition that we're at the tail end of the technology cycle of the venture class basically applying applied to what is probably better utilized at at the time of a private equity business model where you're able to underwrite these sustained cash flows. You're able to lever them up. Um you you're able to like very accurately predict all these things because you have 85% of your revenue coming in before the year starts with these annual contracts or these multi-year contracts. And and so it gave way to this whole new kind of like venture landscape for crypto.

And that invariably like gets way over its knees as it does with every single technology wave within venture. And so like bad things were funded. 2021 was crazy. And we kind of had to put the genie back in the bottle a bit. But there's now like the durable value of what sustains after all these tech cycles. And that is what we're seeing with stable coins. It's what we're seeing with the adoption of things like Hyperlquid and S&P 500. It's basically everything that we're talking about. And we have the regulations on our side. The other thing that's going on in San Francisco this week, it's in San Jose, Santa Clara, is GTC. And what this means is that everyone and anyone who is investing in AI comes out of their holes or comes out of the woodwork and and basically is like, "Hey, I'm here." Um,

there there were two photos that were shared. I saw one. >> Wait, wait, wait, wait. Picture Bitcoin Miami in 2021. Now, >> now maybe a little nerdier. And I think that's kind of where we're at. At least scale. held at the Shark Tank in San Jose, which is where the San Jose Sharks play. [laughter] So, but there there are two photos that were shared which uh I I just found to be hilarious. One of them is actually a billboard here in San Francisco which is Mayfield fund is so there's this whole history of like you go up and down 101 and it's every like hot tech company whether it's the AI companies or the enterprise SAS companies they have a billboard but there's a billboard you know um I think in San Francisco that is

for the first time I've ever seen this Mayfield which is one of the oldest venture capital firms is is promoting the fact that they have $3 billion of capital raised ready to deploy for for you and your AI startup. Like I have never it is usually the billboards companies trying to attract the venture dollars. I've never seen the billboards for the venture companies trying to attract or venture firms trying to attract the companies. The second photo was one of the um sessions at GTC was a reverse pitch where it was the venture firms pitching the startups as like a a pitch competition. [laughter] So

[laughter] >> if there was ever going to be a top in venture dollars attracted to AI, we are we are at least breaking breaking new levels. I don't know if this is the top and it goes back down, but I it is just crazy. Um, and so I I think this is kind of like what I wanted to kind of utilize this to represent is AI is at this level where it can do no wrong from a venture perspective. Um, at the same time, it is also one of the most hated industries in the history of tech, uh, from a consumer perspective. And and so there's this like more dollars than anybody knows what to do going into it. Um and and at the same time, and I think this is also

another interesting fact, uh which I heard on a podcast this week, which is a great point. Um we're talking about the private credit cycle. Basically, all of these, you know, like redemption requests, like cliff, whatever it is, you know, it had like a five or 7% gate, but it got like 14% requests last month. And you've got like all these outflows, and they don't know how they're going to handle them. And and like is this the start? Is this the end? Who knows? Everybody's talking about private credit and like the the private credit blowup is is sort of one of the biggest narratives. >> You know what? Nobody's talking about >> the low the the equities that those back. [laughter] >> Yeah.

>> And if if there was I I think I think it was Mark Rowan um from Apollo had a comment this week where he said somebody asked him about the marks for the private equity book. >> John Zto. This is John Zto. >> Right. Right. Right. Um, and he said all of the marks are wrong. >> There is there is real stuff like like the the spidey senses are tingling in other industries. Uh, whether it's enterprise SAS that's been levered up and the loans are like in in kind of disarray or it's the we always go this direction with new tech waves and like we put way too much venture capital into things and then it gets way out over its

geese and bad mistakes are made. got crypto like in the middle just kind of like biting its time and like building and >> and building >> and like >> let's talk about the so the private credit thing is a really good point. Um, for people who don't know, there's been all these redemptions. For Michael's point, the actual companies that have issued these loans. Um, a lot of them are these software companies. John Zto from Apollo said that he thinks that the average software company returns between 20 and 40% of the value of the loan, which is a lot of write downs. Um, that it's like, you know, if you're doing a loan, you expect to get back 1x. It's not like an equity expectation return.

So, that's a problem. >> Um, and we see that on the Sky side, too. like Sky now has 11.5 billion dollars of stable coins starting to walk out thoughtfully the credit mix and the duration mix but for now it's kind of just like you know just stable coins and and I think that's where everyone is comfortable with and when you have $6 billion of stable coins immediately accessible for redemptions it's like the that's like the golden state of a credit instrument you're in you're out in one Ethereum block it's amazing um and as you look at the credit instruments that are available to allocate into there's some pretty vanilla treasury products Um, but anything a little bit more complex than that, like we need to be really thoughtful about. Like, you know, there's ETFs that are credit like where, oh, they added a new loan, they removed

a loan, it's it's, you know, the NAV is bouncing around. Uh, there's private credit instruments where there's very limited reporting on the underlying instruments. And I think this is actually kind of like a downstream trend of tokenization that's really important is like how do we do like like and BDC's business development companies are generally the public vehicle that these private credit funds raise out of but like how do you do properly where you have control over the assets you don't have like a race to exit with other depositors you don't have you know offiscation of the actual underlying credit um these are things that like crypto can develop and and it will take a lot of time and engineering and I think Sky will lead on this but it's a minefield out there if you're trying to

deploy into credit and that's why Sky is just kind of taking the most riskoff uh you know style that you can possibly have and until things are safe like why take more risk than you absolutely need to um it feels like the market has really come to appreciate that about Scott >> so so let one quick like double click is like the BDC's and I think it's important to talk about the structuring because that's like this is what we're basically talking about Um, these are like semi-public entities which have a nav and they they have sort of like trading value in certain instances, but they're but they're semi private as well in that you have very prescriptive ways that you could subscribe and would withdraw or redeem from these entities.

B REIT is like a great example of one which is a real estate investment trust that Blackstone put together which is this like I think 6070 billion dollar uh real estate investment trust and the u you can subscribe on a monthly basis. You can you know request withdrawals on a monthly basis. Um they can gate it like there are all these types of models that um or or rules that are implemented to to keep these things semilquid. But what invariably has happened is that a lot of these entities are investing in illlquid assets. You know, if you're if you're buying an office building or if you're buying a data center with these with this with these assets that are put in, but they have the ability to withdraw or redeem on a monthly basis,

it's not like you can like sell 10% of that data center and be like, "Okay, cool. Like I need to I need to get some liquidity on that." Like you're holding on to that asset or you're fireelling that asset. And so it's this like how do you manage the illquidity mismatch which is what basically these entire you know um private credit uh entities are going through right now which is you've got you know software loans that are usually private equitybacked companies or owned by private equity and these loans are are you know still I think in in large part performing but it's just a question of the underlying business of these companies but a and so if that was, let's say, to um take over or if you

were to start to see real redemptions that forced the sale of some of these loans, you're going to have to potentially have a fire sale of some of these assets. Now in 2022 we knew that or 20 maybe it was 21 22 I I can't remember exactly when but bite went through this exact situation which is that everybody knew that there was going to be an issue with office leasing because nobody was going back to work nobody was going into offices and so you're going to have this situation where basically like the revenue that you were going to be getting from these buildings that had less than 20% occupancy is just like going to go flat and it takes a while for that to happen but they were trying to figure out a solution.

And I I think it was Calsters or Calpers stepped in and basically bought a bunch of these like positions of redemption requests into B rate um or or withdrawals from B rate and basically stopped like staved off the the total liquid liquidation cascading effect. But it's sort of like you need that white knight to come in otherwise you're going to start to see like fire sales of potentially companies where you know the loans are going to get fire sold and that requires you know like potential takeover of the company itself and so like that this is what I mean when it's like nobody's talking about the private equity for these private credit you know entity assets and the equity could get screwed here um which you know and

that's like private private equity itself is like an order of magnitude maybe two larger than than venture capital and we're talking about venture capital getting out of overseas like this is where people like this is this is where some of the reading the tea leaves or like tinfoil hat perspective is like oh this leads to that leads to this other thing I don't think it's going to get there but but this is the perspective that people have and and this is why it's an issue because there's an illquidity mismatch that is currently sorting itself out and it's going to take a while. So the the curric the quick history of of private credit is uh starts basically in the 1980s with this guy Michael Milin. Later went to jail. Don't need to really get into

that, but he he effectively created the junk bond financing market for companies that are not investment grade. What that means if you're not investment grade is you're not AAA, you're not uh I think triple B, you're triple B and or you're B minus and below. And you know that was kind of the start. Then you know when he went to jail uh effectively after that is when the old uh Milk and team went and started Apollo. They were really the first kind of large private credit fund. Um you know people would do LBOs you know in and around that time and generally how the financing worked was it was bank financing against the company and that was like the first iteration of this. Um everything was kind of just growing on a

linear basis. Deals got larger, the access to capital got better. Um, and then really the kind of most recent development was like the ETFification of this stuff. And so like there's credit ETFs that you can invest in. There are also these things called BDC's which we've covered which are you know this kind of new type of vehicle. But I think this is actually a perfect application of tokenization uh to deal with how to use structure originate syndicate and then allow for minting and redemption of these uh credit assets. Um and so I I think crypto has a lot to contribute here. It's a reason for optimism that people don't fully understand, but I think it is being acutely recognized by the market as it tries to get out of these things. So, you know, all roads lead to tokenization in one way or the

other. Um, in terms of other crypto items to cover this week, um, there's a big acquisition by BBNK, by Mastercard, which is great. And then Tempo launches their blockchain as well, which was >> we should talk Tempo and we should talk, you know, basically Aantic Payments. I mean, a lot of people were [ __ ] on tempo. I think >> that's cope. Um, >> that's like a permission permissionless like >> yeah anyways, so like you don't need to worry about it. Um, >> and I think what they're doing is, you know, they're doing interesting things.

The MPP protocol, which is kind of their version of X42 for aentic commerce, um, cheap, fast, uh, quick finality staple coin transfers. I think all that is great in terms of the tech side. I think really the the motion that they're making on the go to market side is what's really impactful. Getting people to to use crypto and not hate it uh I think is a huge contribution to the industry. So hats off to Paradigm and Tempo. Um but what what's your take? >> Yeah, I think the the best way to think about this is like many people probably heard of X 402. Um and I can't remember the exact product. Basically, it sets up this like frame like a text stack versus text stack of Stripe versus Coinbase.

And you've got basically like MP on the top, X42 on the top. You've got like whatever the wallet manager is for Coinbase, you know, as one of the wallet and then privy and then you've got like um base versus tempo in a lot of ways. Um now there's also a question as to what's the asset. Um, I'm sure the USDC is going to be the prim primary asset within Tempo. Um, but obviously Coinbase has, you know, some some, you know, bias there as well. Um, and some benefit as well. And so, you know, does Tempo start to push another asset? Who knows? I think there's a lot of questions, but it basically sets up this like Stripe versus Coinbase situation, which I think is super interesting as we look to

Stripe potentially going public in the near term or or a year or so. Um, I think there's there's a lot of really smart people who think agented payments is going to be kind of the future of crypto. Um, I I I personally think I I just like need to see more of why this is a huge advantage and I think that there's um there's potential use cases in like crossber remittance which I think is is fundamental. there like uh I saw privy just announced something with Gusto today where there's like contractor payments done via privy uh which crossber remittance where it's like you don't have to pay this like crazy insane feed that takes out like Western Union

and a lot of the correspondent banking international wiring like that there is uh basically all you need to do is pair that with like uh local currency FX function and you've got basic like payments anywhere to anyone um and an instant settlement which is huge Um, and so I think that there's definitely crossber remittance use cases, but this agentic kind of like what are agents that are going to be using stable coins for that they wouldn't be able to paired with something like a ramp and ramp it last week, uh, I think it was last week, maybe the week prior, announced that they're doing virtual cards or virtual accounts for agents. And so it it's going to be this really interesting

battle where and obviously there's going to be agents that probably use both or each or or one or the other. Like many different models will will kind of proliferate. But you need to be able to send money back and forth if you're an agent that's transacting on behalf of someone or transacting on behalf of the agent itself. That's an obvious point. But how do you do that? And why do you choose one versus the other? And what you permissions do you have? I think these are all really, really big questions that haven't been answered. And we know that Asians aren't humans. They're not going to be able to like sign up for a Wells Fargo account or get a credit card in their name, but they can have someone permission an account or permission a card to them that gives them access to

send and receive. And one of the elements that I I think has been really interesting that nobody has talked about. Um, but I I think people have started to glean is this whole like stable coin yield part of clarity where you've got people on the banking side saying or community banking side saying the second that clarity goes through like we're going to see deposit flight and we're going to lose assets. We're not going to be able to make loans. It's going to destroy the banking business model. Like no. That's just like complete [ __ ] And and the reason why is well why aren't we seeing that now? We already have 3.5% 3.8% yield. It was up to 5% on Coinbase and this year you have JP Morgan hitting

all-time highs in terms of profit. You don't have any deposit flight whatsoever. And and so like if that's not happening now, why would it happen after basically what's happening right now gets solidified into law. Um, I think what they're actually worried about, which is JP Morgan and Bank of America being the two largest credit card networks in in the in the world, in the country, um, they're worried about if stable coins can pass back yield in some way, shape, or form. Does this erode the strong network effects that they have with their merchants and card holders? And everybody should remember that credit cards are effectively the business model is merchants pay 2 and a half to 3%. consumers get paid one and a half to 2%. via, you know, points, you

know, cash back, whatever it is. And so being able to have that 2% cash back, I think, is one of the reasons why I I just I think it'll be exceedingly difficult, if not impossible, for, you know, basically a stable coin or a payment mechanism to break the strangle hold of in-person sale, point of sale transactions with credit cards. just like with you know Apple Pay and and Apple Cards like it is now and and um RFID it's just like it is entrenched so hard into the consumer uh that it is going to be hard to rip out. Agents is a new it's it's a green field opportunity and we don't know how agents are going to prefer to pay or or how they're going to be structured enough to choose

different payment methods. What we do know is that if they're able to earn 2% yield 2% back on on credit cards, uh, if they have this ramp, you know, virtual card or this, you know, maybe it's a Stripe virtual card, who knows? But I think Stripe is kind of hedging their bets. They're like, "Okay, if it's stable coins, we got Tempo. If it's credit cards, we got Stripe's core business, but we know agents are going to utilize us. We just don't know how." And and and maybe Tempo is across both. I don't I don't know. But um I I do think that that is more of the nuance of why the banking industry is so scared of clarity. Um and it it it I think it has to do with the credit card networks much more than it does the deposits. Um and so that's just a little bit of like why

Tempo launching this week is super important because it it emphasizes the fact like hey we got to solve this problem soon like because status quo right now is that if you have a third party you can earn yield. So if you're holding USDC on Coinbase, you can earn yield in the form of rewards. And so the status quo is like, you know, banks are going to be able to not like they're they're they're users are going to be able to earn the same yield that they are today if the status quo basically stays the same because genius is already enacted. So, like the banks need that to change and and so I I think it's just an interesting setup um with everything going on and that's why not only is it

like a big launch by by Tempo and and the Stripe team and everybody around that, but it is also a big launch in the context of everything else that's going on in the background. Honestly, I hate these banks. um they were the ones who were like deplatforming people and like doing land acknowledgements, you know, and like just the full DEI circus. These people should not be in control of who gets fixed income yields in the future, especially if something like AGI happens. You really want community banks and JP Morgan deciding how much yield you get. Just a terrible outcome for all of mankind. And the nice thing about the SEC rulem and the CFTC rulem is that that does take a lot of the market structure into their purview. And then we already have genius. And so I do kind

of I I am sympathetic to Brian Armstrong's perspective of like unless it's a perfect deal, why pass it? There will be a symbolic victory of course. Um but I think more than any market structure, we need to build very compelling products and things like Hyperlquid and Sky and Syrup um are just incredible. You know, the proof is ultimately in the pudding with these things and people will buy them uh if we give them a reason to. I don't think market structure really stands in the way of a lot of stuff especially when this rule making has already happened. I think one of the interesting points that we should talk through um okay so we talked a lot about agentic payments and agents but like what do we think about the intersection of crypto and AI or just like what's going on in AI right now not outside of the like funding

landscape but the veteran landscape just like what what is happening in the AI space that's interesting um or you know the intersection of crypto and AI >> um I I honestly think the intersection of crypto and AI is is too early to really forecast at the moment. But like the end state is fairly clear like we turn over the governance of something like Sky or Hyperlquid to an AI and we just tell it go off build more products make more money and distribute it to the token holders. Like that is the purest form of capitalism where you strip out all the costs with AI and then you have all of the crypto revenue directed towards the tokens. And like that is a concept that does not exist today in traditional equity markets. And so it's easy to see the end state. Um the agentic commerce stuff I do think is

cope right now. Uh it just feels like a lot of people who really are jazzed up about a bunch of words being smashed together. Um and being able to raise money on that. And so like I I do think the future of agentic commerce is far out. Um, but ultimately the things that are so powerful about AI will be even more powerful in crypto because crypto is like the the last real free capitalistic playing field that you can apply these tools to. And so we start with that as the framework. I think there's just so so much that you can do. Um, I think right now the probably the clearest use case is something like USDA use wrapped in stable coins. You can pass through depreciation. You can pass through the rental income. Um, like that stuff is so awesome and it addresses a

need that's immediately apparent. Financing of these things and it allows people to give us more respect. Like what we've learned with Sky is like, you know, you want to try this thing? No, not really. Do you want $100 million? It's like, oh yeah, you guys are awesome. Like how does it how does this all work? >> Money ultimately talks >> and and I think we should stick to at least for now. Well, and that that's kind of the angle that I was going to approach it from as well, which is and I I mean there there's all kind of people talking about the numbers and I I don't know how much of them are verified or how much of them are just like let's take the last two weeks and and annualize that, but the the idea generally is that both uh Anthropic and OpenAI are somewhere in the 21 billion

annualized revenue run rate um right now. And I think the um but what is the biggest uh decider as to what the revenue run rate is uh is basically compute. And as they launch and as they go live with more compute, you actually have an increase in the amount of revenue that they're able to charge. It increases the contact window. It increases basically like the the amount of users that they can have on simultaneously. like the the capabilities of what people are willing to pay for probably actually increase.

It's like Fermy's paradox, but you know, it's like as you have more like there's an there's an increasing potential of revenue uh as you unlock more compute. The the other thing is um there was an interesting take I saw from uh um Dylan Patel um which is like why do H100 rental prices continue to go up? >> Yep. which is like everybody expected, Michael Bur included, you know, the the classic short seller and you've got Jim Chanos who's like calling for everything to fall off a cliff, >> washed.

>> He's predicted six of the last two recessions. Um I think the the big the big thing is um they all expected basically like the the GB So the GB200s are out, the GB300s are are out, but they're kind of like still in training mode, I would say. So they're scaling up. Uh and and this is the shift, you know, from basically air cooled to liquid cooled. >> The capabilities of a hopper relative to a a liquid cooled GB300 is like I think 6.6 or depending on how you measure it, it's like 5 to 6x in terms of token throughput per energy. And what um so what the expectation was that you know all the hoppers that are deployed are

basically going to fall off a cliff in terms of the rental prices. Well, what has happened is the exact opposite. in the last few months which is it's actually increased about 40 or 50% per hour and Dylan had a great take on this which is basically like um when hoppers came out you had GPT4 so there there's two ways to measure it it's sort of like the universe of of throughput and you're able to like say okay well the throughput is much less so therefore and the the GB300s are are you renting at that price so therefore the hopper should be renting at whatever that dec whatever that percentage price is the other way to look at it is there's the utility function which is how much can you currently utilize like what's the revenue potential relative to the amount

of compute that's available and you know the H100's came out in 2020 or started to come out in 2020 um and so back then it was GPT4 we're now on GPT 5.4 and the ability to run GPT 5.4 for is actually completely feasible on on hoppers now, whereas in in previous iterations of of the model, you actually had to have the state-of-the-art just to be able to run it. And so what you're able to do with GPT4 and the revenue potential at that point in time was was orders of magnitude lower than what you currently have now. Maybe it was a billion, two billion, three billion, he estimated at that point in time. And so you like the rental price at that point in time was relative to what it was. Now you can run

chatbt 5.4 for on hoppers. And so and and the revenue potential is hundreds of billions. And so you're you're able to say, well, because the revenue potential is hundreds of billions, I'm willing to pay for something that's already turned on, that's already live, that I can literally plug in and can hit ready, set, go, and I'll continue to train, you know, chat 6 on these black wells that, you know, are currently getting deployed or currently. And so it's it's this like as the revenue potential of AI increases and and everybody expects that it will the expectation that these currently deployed outofdate GPUs are going to fall off a cliff I think is is a flawed you know it's flawed relative to what

we're seeing in the market right now. Um so that's number one. Number two is um as you add more compute, these foundational models, OpenAI, Anthropic, XAI will continue to scale their revenue. And the expectation based off of what Daario has said in terms of the amount of gigawatts that he is putting into the ground, which I think is four and a half or five by the end of this year, um basically triples the amount of revenue that they have currently. So it goes from 20 to let's say 60 or or maybe it's 80, I can't remember. But the so like going from 20 billion to 60 billion is just the GPUs turning on. Uh so like the AI potential I think and the AI trade for these foundational models is

is just it it is still on and this is where it's leading in the direction of like oh well if that's worth that then you know the the other one that's competitive to that should be worth some percentage of that and that's why the venture capital ecosystem is going nuts. Just like when you know in 2018 everyone was like oh well if Ethereum is worth a h 100red billion then all of these L2s and all of these alternative L1's should be worth you know billions [laughter] like that that same logic is being applied again uh unfortunately but I and so I think there's just like a lot of things that the market is telling us and market is actually telling us a lot of the the market you know pundits were wrong uh but once again because this whole Iran conflict is overshadowing

everything Nobody's really paying attention to this. Like nobody's really talking about the fact that like capex should be worth multiples of what it is currently worth if it's already deployed into the ground. And the the fact that it is continuing to launch and hit these, you know, launches this year, I think means that the AI potential revenue is going to increase, which hyperscalers have gotten hit over the head because they said $600 billion basically all their operating cash flow goes into the drown this year for for AI buildout. Um or or 100. Yeah, 600. And so there's some multiple that they have to make off of that. Well, I think that that trade is basically still intact. Um, but the Mag 7 is down, you know, 15 20% since

the highs. So, it's just like weird stuff going on. But I think AI is the the the revenue potential is is going to be much larger basically than than most people expect. Um, and it's going to grow a lot faster. Well, um, I think we've covered crypto, we've covered macro, we've covered AI, uh, we've covered some regulation stuff. I think that's it for this week. Um, thank you everyone and we will see you the next time we do this, which is >> Well, we're we're definitely gonna be off next week. Um, >> yeah, you may just get an hour of me talking into the camera, [laughter] Andrew Rant. All right, we'll leave it

there. Thank you.