The Exchanges

Q uh, we've done a bunch of great episodes recently with Sholto from Anthropic, Jerry from OpenAI, and then, uh, Julian from Anthropic, uh, if you're curious to learn more. Let, let's move on to the business, uh, of AI. Uh, you mentioned in the report that the business of AI Finally caught up with the hype. What caught your attention in terms of fact stats in the last 12 months?

A Yeah, a couple of them. Again, like, where we came from one or two years ago was just tons of money going into this segment, building models, a lot of usage, but not clear where the revenue would come from. I think it was maybe OpenAI was making fifty million dollars or something two years ago, and it was very Unclear how they would ever hit like billions of revenue. Um, and nowadays, I think if you sum sort of the top 20 or so, uh, major AI companies from the labs to the most popular kind of vertical applications, you know, across them, they're making tens of billions of dollars of revenue. Um, you can look at the smaller scale companies, which, you know, are growing from zero to twenty million or twenty million plus, uh, as a group, they generally grow about 60% faster on a quarterly basis than non AI companies. Um, we've all seen like the famous charts about ARR or non ARR. It's unclear, uh, but, uh, you know, very steep curves for various coding companies. Um, and perhaps most interestingly across a segment of 43,000 or so, uh, US customers, we work with ramp to show that retention of, uh, subscriptions on AI products across this customer set has really improved markedly since 2022. On 2022 is around the 50% after 12 months. Uh, and now in 25, it's hitting around 80%. Um, and the second stat in that analysis that was interesting was the total spend on AI products, uh, per c…

AI assessment note: “retention of, uh, subscriptions on AI products across this customer set has really improved”

Q riff on that theme a little bit, uh, IP rights, safety, regulatory, uh, a little bit like the sustainability, uh, thing that we were discussing earlier. It sort of feels like that, that whole world, uh, as, um, Sort of slow down in terms of, like, progress, maybe starting with regulatory. Do you think that regulatory is anywhere near catching up or providing an adequate response to what's going on?

A Yeah, I'd say like a big one on that one. I mean, clearly the Trump administration unwound a lot of the Biden era policies, uh, whether that was on, uh, diffusion, you know, trying to push a lot of state level legislation against AI. The, uh, over in Europe, like the EU AI act has had, um, delays in implementations, only three member states that have actually implemented it. And now we're finally seeing how even its authors are saying, uh, maybe we went too far, uh, particularly as we look at progress, Uh, the speed of progress in the U S and China compared to Europe. Um, you know, famously this bill in California, um, you know, rate limiting AI progress was really watered down into what eventually became SB. Um, there were, you know, many, many proposed bills, I think over a 1010% of them actually made their way into laws. Um, so it's still kind of patchworky, but like at a meta level, it looks like we traded regulation for just going faster. It was perhaps like best encompassed by, uh, by the shift between the AI safety summit and the UK, which was at Bletchley, which basically pledged like a whole network of, uh, AI safety institutes and conferences that would happen over the coming years, um, to then the subsequent event in Paris, which was called the AI action summit, completely different than AI safety summit. And JD Vance saying something along the lines of basically lik…

AI assessment note: “at a meta level, it looks like we traded regulation for just going faster.”

Q its entirety for free at stateoff.ai. So, um, we're going to riff on some of the most important topics and ideas in the report, but obviously people can go and check out the report directly for more. All right. So, uh, starting from the top, in the world of research, you mentioned that, uh, your reasoning got real. Uh, so, how far have we come in the last 12 months?

A I'd say pretty far. Um, about 12 months ago or so, we had, I think, the very early inklings of it with O.I. Preview, uh, potentially around, like, this time last year. And, uh, that was the first time you had a system that could kind of Show its reasoning, show its stepwise process to get a more complicated answer. And this has generally been the dream in AI for a long time. And, uh, and since then to now, I'd say like the progress is pretty astounding. One of the areas that the progress has kind of unveiled itself is in mathematics and other verifiable domains where you can like explicitly say, yes, the system works or doesn't work. And, you know, we saw gold medals on the International Math Olympiad by a couple of labs, including OpenAI and DeepMind. That area probably with Uh, if you asked the experts again, how long it would have taken? Probably been a decade. Then in areas a bit closer to my heart in biology and science, we've seen reasoning models, uh, kind of be used as a, as an AI co-scientist. So just as a human would be reading lots of papers, planning experiments, running the experiments, and then doing data analysis, and then reformulating their hypothesis as a result. There's examples of, uh, models doing that in lieu of a human, which is exciting because there's way, way too many papers, uh, to read. You know, AI people kind of complain that it's like 50,000 paper…

AI assessment note: “I'd say pretty far. Um, about 12 months ago or so”

Q all the, what used to be known as Thin wrapper. So the vendors that happen to be powered by those models, so the cursors, the windsurfs, and all the, whatever, legal, financial, AI startups as examples. The other big debate in the business of AI, of course, is the bubble Uh, question. What's your, what's your take? Are we in an AI bubble? Are we not in an AI bubble?

A Yeah. I think like with most things in, in markets, there are probably localized bubbles all over the place. And I think at a, at a high level, what's interesting in terms of vibes and who's calling bubbles and who's not like the finance crowd in New York is definitely talking about bubbles a lot more than what we're talking about in San Francisco, where they're, Their view is like, this is the golden era of AI, and a lot of things are working. We have so much more to, to do, uh, you know, compute build outs are enabling us to experiment a lot faster. Uh, you know, this huge flood of, like, talent that's built the consumer internet and cloud computing is moving into AI, and with that is bringing a lot of optimization techniques and knowledge that AI researchers didn't have when they built the first generations of ChatGPT, et cetera. But I think you, you can't ignore the fact that the, The sums of money going into this industry are truly gargantuan. Um, you know, like, five hundred billion to build, uh, Stargate, and then, uh, you know, a couple hundred billion here, a couple hundred billion there, like, pretty soon it's real money. And then the, um, and then, like, the circularity of these deals is, like, is interesting. Uh, of course, NVIDIA is at the center of this, and it has incentives to use its, its balance sheet to, sort of, spin the wheel faster. Uh, and then perhaps mo…

AI assessment note: “there are probably localized bubbles all over the place.”

Q What, what do you mean by that? What?

A Well, you know, for example, there, there were some labs like Anthropic that were built, you know, to, to really push the safety agenda. Um, because, you know, if we didn't do that, the irrational, uh, went that, you know, we could lead to the extermination of humanity. Right. Um, and I think quite recently, like Dario was interviewed by, uh, Mark Benioff. Uh, just this past week and asked about like some of these data center build outs and, uh, and, you know, he said something along the lines of, yeah, there's a lot of money going into this, a lot of costs, but at the end of the day, the only thing that matters is revenue. Like, I don't think he would have said that, you know, on the founding day of Anthropic. And, you know, it's, it's just the reality that, that the table stakes in this game have changed. And with that, you know, entrepreneurs have to update their priors and, um, and, you know, change their strategy a little bit. And so we document some of this in like sort of the blooper section of the report, uh, which is, uh, which is just like how, how much of a sort of pendulum, um, swinging we've, we've noticed in, um, in corporate priorities at AI labs, um, as a result of the extreme financialization of the sector.

AI assessment note: “how much of a sort of pendulum swinging we've noticed in corporate priorities”

Q 1.5 billion still being a drop in the bucket for a company like Anthropic. Interestingly, does that create a moat over time, meaning that you have to be large enough to be able to afford that kind of money, uh, that you're gonna pay to data rights if you want to do pre-training? And does it make it harder to start a company that needs to do pre-training from scratch?

A Uh, in one sense, yes. In another sense, uh, if you can exploit the Knowledge of these frontier models, particularly from open source, and then generate synthetic data could be a way to get to capable models faster. And also, I think, I mean, you'll have many guests that go deep on this, but, um, But even the, the nature of pre-training and what, uh, information is included in the corpus and at what point it's kind of like data mixtures as people call it has been evolving over time. So I think we're just getting smarter about how to do pre-training rather than shoving everything we have into a bucket and like seeing what happens. And so as a result of that, you might not necessarily have to spend the exact same amount of money to get a capable system. And, you know, some of this kind of came out from the deep seek paper.

AI assessment note: “In one sense, yes. In another sense, uh, if you can exploit the Knowledge”

Q What about, uh, data rights? That was another part of that just general kind of like policy universe that was very, uh, sensitive and controversial. There's been some, some evolution in the last year, right?

A Yeah. Major changes. I think it looks a little bit like the, the sort of on-demand commerce war of, You know, the, the Uber style of do something that's a bit like dodgy for a long time and get to scale and then get, once you're at scale, you're kind of too big to, to kill. And so, and so similarly in AI, like a lot of companies took slightly dodgy practices, uh, to acquire training data and then got to scale. And they were subject to many lawsuits in the last year or two, uh, particularly in, in the media sector, whether that's, um, you know, music or video, uh, and, and books. And then there was a biggest, uh, settlement that happened in the last few months with Anthropic that agreed to pay out one and a half billion. Uh, and this is settled out of court, so it can't be used as precedents, but, but generally shows the, the rough price, uh, tag that's affiliate that's associated with, uh, with human works, uh, in the context of AI training. And then separately, there's been, you know, dozens, if not a hundred organizations that have, uh, agreed content licensing deals with, um, various model companies as I think the The power shift has, has really happened.

AI assessment note: “Anthropic that agreed to pay out one and a half billion”

Q From last year to this year, there's been a little bit of a vibes shift. Is that how you feel as well?

A Yeah, I think it's, I think it's accurate. I mean, there's been vibes in different, In different kind of flavors. Like one is like, you know, when you're trying a model, oftentimes builders will talk about the vibe, like what it feels like when you're talking to it. That's like one category of vibes. But then the other major one, which we highlight, particularly in the politics section is just like the sort of pendulum swift switch from existential risk. If we get this wrong and if we scale it too fast, like we can potentially, um, you know, generate the extinction of the human race, basically. Um, to like switching to the complete other side of like, we need to scale these products and like these, these models are very useful. We probably have like a technical overhang right now of just more products that we could build with the tech that we have. And there's a big race of trying to make money on this. And so what I found pretty surprising is just how powerful this fiber shift can occur where you're like the same individuals who are, you know, in front of like government leaders, 12 months ago, kind of warning about this or the same ones that are buying billboards in Paris or in New York saying like, please use my app. Um, so, yeah, it's, it's, it's tough to, like, it's tough for me to understand, like, how, um, how that can occur, and how that same individual can, can maintai…

AI assessment note: “Yeah, I think it's, I think it's accurate.”

Q Broadcom is experiencing a resurrection of some sort as well, right?

A Yeah, yeah, exactly. Yeah, it has. Uh, I think it's maybe a decade ago they bought a company that Now is kind of the internal team doing this custom ASICs for, uh, Google's TPU, and, uh, you know, more recently they announced a deal with OpenAI also to do, uh, a custom chip. And at a high level, what's interesting with the rise of Broadcom is basically GPUs have been the, the, the dominant chip set for a long time as the, uh, kind of nature of the neural network or other kind of AI system that you're running on the hardware was still changing very rapidly. But as soon as you get to a point where there's some convergence on an architecture that's looks like it's stable and is revenue generating and developers are coming to, uh, sort of work on it and confirm that it is like the thing, then you can flip towards doing a custom chip that's built to extract the most value out of that architecture. And so the rise of Broadcom basically tells you like there's strong forces that are saying that the transformer is the thing, but At the end of the day, like, we also look at how would your dollar be best used as an investor if you wanted to bet on chip companies. And, uh, and in the graph, in the report, we look at sort of six of the major contenders, uh, to NVIDIA and basically said, you know, if you bought NVIDIA stock on the day of, uh, the announcement of all the, like, private, uh, r…

AI assessment note: “Yeah, yeah, exactly. Yeah, it has.”

Q You were just talking about Lama. I think somewhere in the report, you have this, uh, really funny sentence you call Zuckerberg the de facto messiah of open source. What do you see around Lama? You know, what would you think? Uh, you know, why are they doing this? Uh, and what's, what's Impact it's, it's having.

A Yeah, I mean the, it's probably one of the best like ROI trades of like a public company in like a long time. Um, and basically like the chart shows that from the launch of Metaverse, like companies saying we're going to invest, I don't know how many bajillion dollars in Metaverse to basically we're going to stop doing Metaverse when the stock price was just bleeding. And I think various shareholders had written letters to tell them to stop, et cetera. That was like negative six hundred billion or something like this in market cap for meta. And then once they shifted from we're not gonna do metaverse anymore to we're gonna do open source AI and you go through like one, two, three, Lama. Um, that like trough to peak is like right now probably a plus 1.2 1000000001.2 trillion, sorry. So he turned a negative six hundred billion to like a positive 1.3 trillion dollar. Uh, market cap appreciation, which you could argue, like, who cares if anybody's using this stuff? Like, that in itself is, like, amazing shareholder value creation.

AI assessment note: “it's probably one of the best like ROI trades of like a public company”

Q Do you see some movement around, uh, hey, we may have over-regulated and is that powerful enough that, uh, you know, things may evolve or is just like the inertia is just way too big?

A I think the nourish is too big. I mean, Michael has had this narrative for a while, but, and like France was one of the bigger, um, kind of defenders of open source, et cetera, but I don't think it's going to change much. Um, I think the UK had a cool opportunity because, I mean, I think Brexit was a terrible idea for the record, but, um, the value of that at least is like they can define their own policies, which they did well during COVID with approving their vaccine really fast. It was not a great vaccine, but like the approval process was good. Uh, and so the theory they could use that, uh, to, to be, like, competitive in AI, um, but again, I think the country has, like, all sorts of other problems that they're working through, um, which are probably, like, higher priority than, like, solving for AGI and things like this, um, and then other countries just seem like they're complete disarray and, like, have basically no growth, so it's, it's, it looks depressing to me, to be honest, and then it doesn't help that only to You know, every other week or something, there's somebody that's like, I'm so happy to announce I'm leaving London for New York or something. I don't see that many people going the other way.

AI assessment note: “I think the nourish is too big.”

Q All right, let's switch to the physical reality that this whole stack sits on. So infrastructure, data centers, uh, energy. You, you, you mentioned in the deck that, uh, power Uh, has become the new bottleneck. What is your sense of the state of play in the energy procurement game?

A The biggest step for me is one gigawatt of a data center for AI basically costs fifty billion dollars in capex. Um, and, uh, on an annual running basis, it costs between, like, another eight to nine, uh, eight to nine to maybe even eleven billion dollars to run. And so, uh, when you have just, you know, casually a 10 gigawatt data center, uh, that's like a lot of money. Um, and, um, and so one of the problems is, like, where does this energy come from? Uh, you know, traditionally it would be from, I don't know, coal, uh, or natural gas, um, essentially solar, or ideally at some point in the future nuclear. And what we're seeing is, right now, Companies are trying to do deals with anybody who has any capacity, so we kind of call some deals with, uh, future, uh, nuclear, uh, you know, reactor companies, then that would take maybe a decade or two decades to deliver, um, you know, famously.

AI assessment note: “Companies are trying to do deals with anybody who has any capacity”

Q You know, maybe thinking through applications with your investor hat on, what is it that you're Looking for, excited about these days?

A Yeah, um, I'd say, I'm trying to find, like, the, the, the classic, like, what is not consensus now? Like, what are people not believing? What has been tried, and like, they're like, ah, no, it's not gonna work anymore, but for some reason you think, like, it will in, like, you know, two, three years, and obviously if you're wrong after three, four years, you're basically dead and wrong timing, et cetera. Um, so taking that lens, like, I was excited about, like, AI in biotech five, six years ago, made a few investments in there. Defense was interesting in Europe. It played out a little bit more in the U.S. a few years before, but, you know, pre-Ukraine, there was really not that much, and now there's, like, more and more.

AI assessment note: “taking that lens, like, I was excited about, like, AI in biotech”

Q And it's interesting that they decided to be very strong open source contributors. We had Clem from Hugging Face on this podcast a couple weeks ago, a few weeks ago, and we were talking about this, and, uh, is it sort of, sort of unclear why that happened? There must be some kind of very smart geopolitical reason.

A He doesn't know, I probably don't know. He's the king of open source, but, Uh, yeah, it's, it's, it's interesting. And, and Western companies are using them. Uh, I think the, you know, the data as of like a week or two ago from his platform, Hugging Faces, like there's probably over half a billion, uh, Lama derivative models that have been downloaded. I think Quen is growing rapidly as well. Um, so that's like one part of this, this debate. The other one is, of course, this like constant, um, fight between OpenAI, Anthropic, you know, Google DeepMind, GDM, uh, as being the main contenders. I think on the one side, you could say, like, the gaps seem to have narrowed in capabilities, like, when you're looking at these various benchmarks, and they sort of look increasingly esoteric to the non-expert receiver, but the gaps seem to be pretty small. Um, this is outside of, like, vibes and what they feel like, and I think there are consumer preferences for one or the other, but, uh, but while the gaps have gotten small, and, you know, Meta has this, like, ginormous systems are giving away for free, which, um, which are interesting in their own right, You know, a sort of satirical take on all this is like, you could have basically looked at the state of the, um, the landscape, like, who's number one, who's number two, who's number three, 12 months ago, and then just deleted Twitter, li…

AI assessment note: “He doesn't know, I probably don't know.”