SemiAnalysis' Jeremie Eliahou Ontiveros on all things datacenter / power

Summary

• Data Center Power Trends: The podcast discusses the increasing demand for on-site gas power solutions for data centers, highlighting the use of modular units and reciprocating engines to meet large-scale power needs more quickly than traditional methods.
• Investment in Natural Gas Infrastructure: There is a significant focus on the deployment of natural gas turbines and engines, with companies like GE and Caterpillar providing solutions to meet the growing power demands of data centers, particularly for AI labs and hyperscalers.
• Capex Growth in Hyperscalers: The conversation highlights the substantial capital expenditure growth expected from major hyperscalers like Meta, Amazon, and Google, with projections significantly exceeding current street estimates, driven by the need for increased data center capacity.
• Coreweave's Strategic Position: Coreweave's rapid expansion and strategic partnerships, particularly in securing power and data center capacity, are discussed as key factors in its ability to compete with larger hyperscalers despite potential risks in a down cycle.
• Oracle's AI Strategy: Oracle's aggressive investment in AI infrastructure, including long-term contracts with companies like OpenAI, is seen as a bold move to leverage its balance sheet and expand its cloud business, despite inherent risks.
• Power Efficiency and Future Data Centers: The discussion touches on the potential for improvements in power efficiency in data centers, with the possibility of future data centers being built in more remote locations as power and infrastructure challenges are addressed.
• Robotics Market Outlook: The podcast introduces a framework for understanding the robotics market, with levels of autonomy similar to those used in automotive, and discusses the potential for growth in mobile robotics and humanoids.

Transcript

You're about to listen to the yet another value podcast with your host me Andrew Walker. Today's podcast we have Jeremy from semi an semi analyst on Jer this is his second time on the podcast. He came on in February. We talked all things semis and particularly power then we dive into all thing power all things power. Now, I think it's a really interesting conversation for someone who is I I mean I mean if you've ever seen the Sammy analysis stuff deep deep in the weeds, but it's going to touch on a lot of things whether you're interested in, you know, Nvidia yolos and coreweave or whether you're interested in just kind of the demand for power, the outlook for natural gas, the out look for natural. We're going to touch on a lot of different stuff. I think you're going to be really interested in it. We even hit on a little bit of Oracle and Larry Ellison, who I I've been a little bit obsessed with since the book club on Oracle. So, we're going to get to all that in one second, but first, a word from our sponsors. This podcast is sponsored by Portrait Analytics. People ask me all the time. What's your favorite stock screen to run to look at for ideas? Is it low price to earnings, high dividend yield? What What is it? And my answer is simple. I don't run screens. I somehow doubt that there's serious alpha and sorting through stocks that are trading under 10 times price earnings on Yahoo Finance. But, portrait analytics has completely changed the game on screening. It lets you create bespoke screens to generate actually unique ideas. Let me give you an example of one that I've been using recently. I wanted to look for stocks with greater than 200 million market cap where the company has publicly discussed trading at a discount to peers and both the companies and insiders have been buying shares on the open market in the past 12 months. To me, that's an interesting screen. It's unique ideas. It's a blend of quantitative and qualitative. It's pulling things that aren't, you know, just purely numbers based that the insiders are talking about and it's building something that kind of fits with my view of the world and my view of the types of stocks that would be interesting. Portrait let me find a handful of companies that meet that criteria. The last time I did the screen, by the way, the screens run every day if you want them to, every week if you want them to. The last time I did the screen, it had four or five stocks that exactly hit that criteria. And then boom, I had a list of really interesting things that I actually might buy that I could sort through. Uh they it also showed me exactly where the company was talking about, how their valuation compared to peers, so I could see, hey, was this a one-off or are they consistently talking about it in a way I think is interesting. Anyway, I think it's completely changed the game for screening and for generating new ideas. If you're looking to up your screening game, you should check out Portrait Analytics at portraitanalytics.ai. I'll include a link in the show notes. >> All right. Hello and welcome to yet another value podcast with me. I'm happy to have on today for the second time, uh, Jeremy from Analysis. Jeremy, how's it going? >> Hey, man. Good to see you again. Thanks for having me. >> Look, you've got an open invite. your your last pod was actually I I never reveal stats, but it was one of the most popular podcasts we've put out this year. Uh YouTube stats were great. I don't know what it says that the the audio stats were even better. I don't know if that speaks to my face, your face, whatever it is, but both the stats were great, but audio in particular was off the charts. So, the people loved it. I'm super excited to have you back. I want to dive into all things semi analysis after I just remind everyone, quick disclaimer, not investing advice. We're going to talk about a ton of stocks today. See the full disclaimer at the end. Uh Jeremy, you and I were just wrapping earlier today. Obviously, I think our conversation is going to cover everything. Semis, power, and everything, but I'd love to just start like kind of in the semis and power space. What's real what's the next big trade in the semis and power space, I guess, is the headliner. >> Yeah. Yeah. Disclaimer again, not investment advice, but uh definitely I think one of the interesting things that you're seeing is the surge in on-site gas. Um and I'd say a lot of people talk about uh deploying CCGTs. uh to power data centers but there's an issue with those uh is the lead time everyone knows there's a multi-year lead time to get TCGTs maybe what people don't realize is that actually you can deploy at very large scale much faster more smaller units more modular units um so you have two types of systems you have the IRO derivatives which are termites mostly sold by GEV and CAD um and you also have uh basically huge car engines uh it's called rice engine reciprocating supplied by folks like Volto Grid, also Caterpillar and a few others. And in both cases, we're increasingly seeing um I say solid reliable data center projects. Uh projects that I believe are going to be operational in 26 27 serve some very large end users like AI labs and hyperscalers. Um and I think they're going to be deploying those technologies. Um so basically if you think well what's the market for on-site gas for data centers um in 2024 it's pretty much Elon Musk it's pretty much the Memphis site in 2025 Elon Musk again is a driving force um so in 24 you had like 300 megawatts of on-site turbines or 250 whatever 25 he's probably adding another 500 to gawatt on site but you have other sites uh another one that's well publicized is the Stargate Abene Texas site which is going to deploy 300 160 megawatts of those GEVOR derivatives, but there's more actually. Uh I can't talk about all of these sites. Uh if you want to do more, subscribe to the data center model. >> That's a fantastic pitch. And look, I was just reading some stuff to prep for this podcast. And I if you have I mean I mean Alice is huge. If you haven't subscribed though, you I my personal favorite is you guys just have the the drone shots of all of these things, you know, with the uh especially on the Oracle colllocated one where you're like it's Oracle here. It's it's just really interesting. I'll give you a really deep knowledge just on natural gas turbine. So you said it and I think me as a dumb dumb journalist the thing that worries me is like look G Vernonova who provides a lot of these turbines. Uh the stock has been a screamer. I mean it I believe the spin-off from GE happened in 2024. It's like a sixbagger since the spin. The the stock is a is up one and a half times this year. So I'm not saying like hey I'm scared to buy a stock that's gone up though. that that is always a little worried. I guess the thing with G Vanova is what I have heard is hey these wind turbines they they are complicated right but pricing is going through the roof because the demand is crazy and what I kind of worry about as a generalist is twofold. A if the demand pauses and I do want to talk a lot about AI capex, but B I believe pricing is also going up because GE and the it's kind of an oligopoly and the other people who make it are like we don't know if we can trust in this demand. So they've actually held back on building new plants, bringing new supply online. So they've got this beautiful, beautiful thing where demand is through the roof and they're not bringing supply online. Like how long can that really hold? Right? at some point somebody's gonna be like I need to bring a little bit more supply online and once that happens the floodgates break and you know you worry you're buying into a super cycle and there's two methods for it to break so I knew I threw a lot out to you but I'd love to just get your overall thoughts on that. Yeah, sure. Uh so gas for the roof, you said it. Um however, I would say there's two types of demand here. Um one is the big turbines again 500 megawatt per unit. Uh those are mostly serving the grid in a few select very large data centers like Meta for example talked about building an on-site power plants but mostly it's to it's to serve the grid. Uh so it's still related to data center growth. Uh but it's for a broader use case. Now you have another use case which which is deploying turbines on site for time tomarket purposes. And what is interesting is that the dollars that flow to these turbines are actually to some extent not net new dollars. They're taking market share for something else which is the diesel generator market. Uh so actually what you could see u and what you're starting to see if we take the outline Texas data center as an example. I mean I look at solic pictures I don't see any diesel generator. Um, and what I understand is that they're going to be deploying those turbines maybe at first for on-site power, but then as grid power comes online, uh, they're going to be using those turbines for backup. Um, I think you're seeing that same pattern with, uh, the Memphis data center where, uh, Elon brought online those smaller turbines again, which are much faster to scale up, uh, which can be manufactured at scale. They're factory made. He first uses those turbines for on-site power and then as with power is built and they have a substation and such they use it as backup. So what happens is that the dollars that typically flow to diesel generators now flow to turbines which serve initially as primary power and then as backup. >> Let me just a question on that again. I I I'm you are the data center expert I'm not but when you're building these big wind turbines and backup and you had a great piece I I was so fascinated by the piece on intermittent power issues at data centers and their implicate their impact on the grid. We can talk about that later. But when you say, "Hey, you're making I mean, you're making tens and hundreds of millions of dollars of investments into these tur into these n gas turbines, right?" And they initially serve as primary power and then they're eventually serving as secondary power. And this might blend nicely into our capex discussion at some point. But when I hear that, I say, "Oh, you have a rush right now, right?" But when you're spending hundreds of millions of dollars on something that's going to end up being your backup power once like the grid's kind of online, that doesn't seem sustainable, right? like yes it's sustainable when you need to get stuff up now but when you're planning your 2027 2028 uh data center at that point aren't you kind of looking and saying hey we're going to be connected to the grid maybe we don't need to spend hundreds of millions of dollars on these uh giant turbines and also I mean there is the discussion of as batteries come as batteries get better like maybe you need less backup sources it just strikes me as if these are eventually going to be backup sources it sounds great in the near term and obviously it sounds like diesel generators are really up a creek uh you know the the medium to longer term outlook for these strikes me as a little bit murkier. I mean data centers have been deploying backup forever diesel generators and if you think about it most of the time those are stranded expense right they never actually serve like if you normalize the cost of power uh in terms of how many hours they run per year for these diesel generators it's absolutely high probably in the thousand per megawatt hour so those are not economical purchases uh they serve when you have a blackout um and so for example in Spain a few months ago there was this horrible blackout the data center industry was to some extent like I I would say I don't know proud I don't know if that's the right word but just because they were actually able to navigate uh through that event uh >> you you've got to be very careful because you're the data center industry you're like don't worry guys we had 100% uptime through the blackout and then you're like hey the hospital down the street was offline granny was sitting in 100 degree depart I think that answered it well I I do have some more questions there but let me let's go to the overall AI context you guys published a piece go oh did you have something else please >> yeah I I want to flag something else which is that um so the thing is that uh today it's all about time to market um but as the market matures like some of these uh or or maybe think of it this way uh it's obviously it's technical so right now it's the upcycle so time to market matters everything uh but then as we get into the down cycle which for sure is going to happen at some point uh you're going to see companies turning cash saving mode when they do that they're not going to be contracting new cloud capacity or much less uh they're going to be using their existing footprint which basically means taking training GPUs to serve inference. Um and so what if you build a training data center with no backup at all because uh because uptime doesn't matter so much for training data service and then suddenly you have a down cycle and you want to save money. So your inference service um I mean you want it to be high availability. Like that's also why it matters to have backup. Like that's uh again a decade old rule for the data center industry to have backup. I I I just think it's it's a you just shift from diesel to to gas basically because it serves as an initial purpose of on gas. Yeah. Sorry. Go ahead. >> No, no. Hey, that that's fantastic to think about. And obviously when I asked the question, I wasn't framing it as a hey, these guys always need backup power. But I guess my my push to you would be, you know, right now they're doing it with the NAC gas turbines because it's you do the Nackas turbine, it serves as your primary and then you can use it as your secondary once you're plugged into the grid. If I was planning in 2028, right? Let's say I had something in 2028 and I had a hookup to the power grid in some way in this thing that's coming on 2028. Would someone be planning that 2028 power plant with the NAT gas turbines? Or would they go back and say, "Hey, obviously I need backup." Nobody's on that. But would they go back and say, "Hey, in 2028 batteries much better, so maybe I can store a lot. And even if I don't want to rely purely on battery, I mean diesel is, as you said, it's always been the backup. It is so much cheaper than gas backup. Should I just use diesel plus battery as my backup instead of buying these, you know, giant super expensive nack gas things because I don't need a primary anymore. >> So, first of all, with the second cost, I actually think it's pretty similar. Uh, yeah. Nad gas. >> Oh, is that right? I thought diesel was significantly cheaper. >> Um, I mean, if you if you if you look at the the diesel engines, they're actually very similar to the nad gas engines. So, for example, CAT has both a diesel engine and a nad gas engine. They're pretty similar, roughly same pricing. Uh now again as I said earlier you have basically two types of on-site gas power for these deployments. You either have like what is called reciprocating engines which basically giant car engine 3 megawatt 4 megawatts per unit and then you have the turbines I derivatives which are like 30 15 40 megawatt per units. You're seeing both uh today being used. I'm not exactly sure what's going to be the share but both seem to be be getting a lot of traction. maybe slightly more on the turbine side. But anyways, in both cases, actually both the turbines and the rice engines have a roughly similar cost structure. It's expensive to be clear, but it's not like uh a surge versus diesel. So, in terms of cost, it's roughly the same. Um the difference is really about uh sight selection like basically you need to be near a pipeline. That's the the only difference. You didn't need that before. You could just build a tank on site for diesel whereas now a site selection criteria is having that gas access. >> Perfect. And I actually do want to come back to site selection clear. Well, let me back up a second. Capex, you know, you guys published this great piece on uh it looked at the five hyperscalers basically, right? Uh if I remember correctly, Oracle, Amazon, Google, Microsoft, and I think you guys had core reef thrown in there and you said, "Hey, we are forecasting capex a your cap I thought the two interesting pieces were a your capex estimates for the rest of 2025 and especially 2026 and 2027 were way above street estimates, right? So the party the party ain't stopping for capex. And then the other interesting actually I'll just pause there. What are you seeing that suggests to you that the the capex estimates for the street which by the way it's pretty bullish on capex and I think capex estimates have been going up a lot as the year has gone on. What are you seeing that's suggesting that the street is still not just low but way too low in your opinion? >> Yeah I just want to say first of all that the report you're mentioning was sent on core research. It's only for institutional clients. Uh it wasn't on the broad newsletter website. Uh, so sorry guys, it's not accessible for free. >> We can still talk about it though. We can still talk about it. >> Talk about for sure. Of course. And we're gonna do it. Um, anyways, why is capex street? Well, first of all, I ask you like you look at sellside capex estimates. Um, and you they're normally modeling growth versus 2025. So 26 27 they have some growth like maybe high single digit, maybe low double digit, but it's not a lot, right? Um and so I guess we go back to something we discussed last last time which is how do cycles typically play out? Uh is it generally slow slow steady trend or is it actually strong steady up and then down? Yeah, I think again 50 years of history back me up in saying cycles tend to be like very strong towards the upside. Anyway, this is just overall I guess theory. Now what we're seeing more on the ground uh using our data center model for example, we're seeing that construction stocks uh for self-built data center are surging. We're seeing that the the leasing rates uh pre-leasing I should say of hyper steer are extremely high today which again when you pre-release it means that the data center is going to be official next year. So that's indication of next year's capex. So most of the forward looking data center indicator uh points to very sharp growth uh in the high double digits which tracks pretty well with the numbers that you saw which is like high double digits growth for hyperscalers and way ahead of shoot. So I I it's just like that th those signals are clearly don't show capex being stabilized at high levels. They show capex going up up up and I I mean this is up up right like from 23 to 24 capex was up 50% from 24 to 25 another 50%. You guys have I I think only 35% to 40% growth in 26 and like 25 to 30% growth in 27. uh you know at that point the these companies are eating the the world right like just those five companies I think you guys are saying in 2026 over half a trillion dollars of capex so I I unless you have anything to add there I I have some questions on that but if you want to add anything I can just pause there >> yeah I mean um again like all the signals go up so we're just saying this company's like committed to to invest and if you think about like um I mean we can talk about the drivers maybe we can do that after uh there's a couple of ways to frame what's driving the markets. Uh I would just say yeah like we don't see anything today that would suggest they're slowing down and also I do want to emphasize that or report was actually sent before earnings. Um and the early indications that you saw the best one was from Meta where they basically said we're probably going to add roughly the same dollar amount on capex uh which suggest plus 40 billion year on year. So they're spending what 65 70 billion in 2025. They're basically saying we're going to be close to hundred billion dollars by 2026 soft guidance, right? Um so like that that's the first indication from management that we've had and it clearly goes towards what we have which is actually in terms of dollars amounts the growth year on year is going to be roughly similar. >> But so you've got accelerating spent and I mean these companies are spending at levels that in terms of as a percent of the economy I don't think we've seen since kind of the railroad buildouts in the late 1800s. I could be wrong. It's not like I've got my fingers on all that kind of but I mean I it's going to be over 1% of the economy really really quickly. Right. So just high level is there obviously it's not accelerating because we're going from 50 to 35% growth but I I think no one would fault me if I said hey it's not exactly diminishing when you're talking numbers this big. Are there any worries about signs of diminishing ROI yet? Um yes and no. Um I mean if you if you take a a simple view of the balance sheets of these companies that you're seeing definitely uh the sort of revenue to assets are going down uh for folks like Amazon for example and if you compare uh revenue to assets for these companies versus a core uh for example you're seeing that GPUon business model clearly has a lower asset turnover uh which is I guess is a way to say returns are going down but on the other hand but that that's not fair though like you are fine. You're going from Microsoft where they were selling capex free software licenses, right, to this heavy business like yes, it's going to be lower ROIC and lower whatever it is than the software licensing business, but it doesn't mean the the returns on capital aren't incredible. >> The the returns of capital from from the estimates that we've done, they're they're decent. There's not they're not as good as uh typical Azure business, for example, uh which was very high margin, but they're pretty good regardless. Um, and one thing it's a it's a one of the big questions. What's going to be the uh the life uh of these chips? The useful life of these chips? >> Actually, my next question. Do you want me to can do you mind if I frame it because I did a lot of work. So, look, if you talk to Bears, they will tell you Jim Chenos, and I I hate to pick on him, but he's got two prominent tweets in his last like five days on it. He says, "Hey, Microsoft's uh I I think the one that really jumped out to me was he had a tweet that said, "Hey, Meta is depreciating their entire useful life at 11 to 12 years, and a lot of that is GPUs, which they're depreciating, I believe, at 3 years." And there's lots of debates on shouldn't you be depreciating GPUs faster. And I think he kind of missed the point, but he's basically saying, "Hey, if Meta is building all these things and they're depreciating at his best guess was 11, 12 years, he thinks it might be 20 years if you do something. Are these guys dep are these guys I mean it's basically saying not accounting fraud but are they overstating the returns on investment because you should depreciate these at five instead of 10. So you're actually way overstating the investment and I I'd love to get your thoughts on the depreciating useful life that you kind of so teed me up to ask you. >> Yeah. So just top of my head the the accounting depreciation that we have today is about six years. Meta is 5.5 if I remember correctly. Coven or six four servers. Um anyways uh so the big risk is what if from six years you go to four years. Um there's a couple of ways to think about this. First of all we can look at empirical evidence. Uh the best empirical evidence is the HPC world. So high performance computing um some of that stuff is pretty public right if you think about the top 500 supercomputers some of those have been running for over six years. Uh typically what you hear from the HPC community is for five to six years useful life. So I think when you use empirical data what has happened historically u it's tough to make a case that this is this is going to be that six years or five to five 5.5 years whatever is not an accurate measure right so that's the first thing I would say is it so it it kind of makes sense based on the signals we've seen historically now you could also argue that uh we're running these systems at at max power these days uh we're just yeah we're just doing more crazy stuff I guess with uh these GPUs uh obviously we're trying to sweat every single watt out of these GPUs. Like the whole point is trying to utilize them as much as possible. Um, as such, you could imagine maybe they're going to be um they're going to be they're going to be dying faster. Um, it's uh it's honestly it's anyone's guess at this point. Like I just I just cannot know what's going to happen in the future. Again, just looking at historicals, five to six years seems okay. But then I do think it's a risk. Um and if it's five instead of six years or four years instead of six years then I think what is going to happen we had a post on Oracle showing that um the very large open AI type contracts actually if you look at how they work the margin that they get out of it is very high uh we estimated about 40% EBIT margin uh just on a project basis to be clear so that doesn't include all the structural cost and such but on a project basis this is very high very high margin project but now of course if actually the useful life is four years They're going to book a massive write down in a few years and they're going to have maybe one quarter with like minus10 billion dollars of of loss into accounting. So, >> you know, just what strikes me is like you said, hey, useful life is kind of six years. If you go from six to four years, like maybe you overinvested a little bit, you know, but it's not the end of the world. I think the differentiation is if he's right and they're doing useful life of 12 years or 20 years like that is the crux of the bear tweet and yeah if you did a useful life of 12 or 20 and then you write it down to four well then it was a disaster right but what you're saying if it's at six and it's four I don't think anyone's sitting here saying hey you know this giant hundred billion dollar data center build it's going to be completely useless within five years so I I'm kind of with you now there's there's still the questions on return on investment. Let let me just ask that you know the returns you you said the returns on capital are good. How do you guys measure the returns on capital here? Because to me like you do have the issue you know there's the famous thing with bank financials. The scariest thing in a bank is a fast growing bank because if it makes a dollar of loans today and then $500 of loans two years from now next year and then $50,000 of loans three years from now, their metrics are going to look great, but that first dollar loan might be terrible. And you kind of don't know it till the till you level out and then you say, "Oh my god, like we've just been fooled because all the new loans aren't paying off." These guys are accelerating capex so fast. How are they measuring their their current return on spend, right? because they're spending $350 billion this year. That's not even coming online for 12 months. So, how do they know that they're getting this great spend? And it's not just, oh yeah, the first 10 billion was great, but the next 350 billion was just crazy. >> Yeah. Um, yeah. I mean, it's the it's basically the co business model, right? Like the bears are looking at that and saying this is the worst business model ever. uh and the analysts that we've done which we actually posted a report was pretty uh pretty positive on corre uh to be clear we don't provide financial advice but we just looked at the trends and everything to us looked actually much more uh viable that many people were suggesting um and the way we do it we just build a project by project analysis right um and so actually we built a comprehensive model about a year ago called the AI cloud TCO model um which we actually built for someone building a GPU cloud literally. Um, and so we we feel like very good about all the estimates we have here. And what we did is we just estimated all of the different capex related costs, all of the different opex. We went extremely deep in the rabbit hole to model like every single line item. Um, and look what we're seeing is there there's a few things that impact your returns. Of course, like uh you want to have low data center cost. Of course, cost of capital matters a lot as well. uh because many people in the neo cloud industry uh have a very high cost of capital right like uh a year ago it wasn't unusual to see like 15 to 20% cost of debt uh cost of equity probably also in the 20% plus I think it's going to get down because now these companies are getting more and more mature especially core weave people are much more comfortable lending to core at much better rates um but yeah any anyway so there are a lot of these assumptions that are baked into the business model and if you actually are able to optimize on your opex on your cost of middle and you can also optimize on your capex to some extent which is something we like that Oracle is doing with networking. they have a really good networking configuration to serve very large clusters uh enable enabling them to have a high lower capex than others that can actually all of these optimizations taken together can take a business that for random neocloud it may be I'm making it up five to 10% ARR uh to a 25 to 30% AR business for anyway the point being to answer your question they think they're thinking about this on a project basis uh just like we're doing at least that's my understanding I might be wrong uh but I I think the best way to think about those is because there's so much upfront capex. You just want to know like what are like your likely returns and that's also why you see a lot of these big hyperscalers signing much longer contracts than the average market rates. Uh I think from core disclosures uh from leaks that happen in by the information and rotors and whatnot on OpenAI it's pretty clear that they tend to sign four to fiveyear contracts. Uh, and when you sign a five-year contract, basically you have a guarantee AR. Uh, unless something goes wrong, of course, or any I fail and whatnot. Um, just on Core Reef, just you said, hey, you publish a report that was positive. You know, I wish I I really looked at that IPO for a while and I was like, man, this IPO, you know, you get an IPO that's just really on launches low float. Like, we we've seen this a few times with ARM and uh especially with ARM. And I really looked at it and then the stock was up 3x a month later. I was like, gosh damn it, Andrew. You >> you swole. But I I do want to ask you, and we're not making financial B. I I'm not even talking stock price here, but you know, Corweave, you heard lots of bears, especially at the IPO time. And even now, you know, a lot of people will say, hey, the stock price is completely inflated by a thin float. There's lots of, you know, pod monkeys trading around with borrow rates and everything. And just on the business, you know, it jumps out to me that this business was really spun up in large part because Microsoft was so desperate for capacity that they entered a huge deal with them. Microsoft open and Microsoft has said, "Hey, we're not doing that that deal again. We're trying to do everything on our own." Right? Look at this company. Their capex spend even at their level is a pittance compared to uh compared to their larger peers. Their larger peers obviously have other business models where they can subsidize. they can go build a new g a new data center knowing hey most of it's going to be taken by our core internal processes and then we can fill the books with third parties if necessary. Coreweave doesn't have any of that. So I I just love to ask again not the stock but just the core business kind of why why do you think it makes sense and why are you guys still kind of positive on the business versus kind these giants that are playing in a similar field with other advantages? >> Yeah. Um so basically if you think about the market for very large contracts open AI hyperscalers uh nanthropic uh these guys it's basically a commodity because what these clouds are providing is bare metal infrastructure so what you have to do is building a data center putting some machines in there assembling the machine uh through networking but there's no software layer on top of it there's no like technical mode technical differentiation or there's to some extent but really much lower compared to what we we're used to see from AWS and Azure and others. Um so barriers to entry are much lower. Um and if you think about it, okay, it's a commodity. How can you win? Um there are two two ways to win. One is long-term is uh having the best cost structure. But actually that's not what matters today because what matters today is speed. That's really the thing that matters because uh that that's what these big end users want. That's what OpenAI wants. That's what Antropic want. They want the clusters as fast as possible. So basically you got to optimize everything for speed. Um and this is where you look at core strategy and they've actually done some stuff uh which enabled them to beat everyone else's speed. I mean they contracted over like 2 gawatt of power in roughly two years. Uh they had to take on of course material financial risk and such to do that. Um but for example they've been working with those crypto miners which no one was considering back then. Cory was among the first to sign with a minor and these guys they have the power right there. They have the substation. Everything is in there. Um so you just need to trust that the minor uh has good enough contractors to actually build a data center but the time to market is unrivaled right so if you think about the challenges to build a green field data centers well cor said no I just want to do I just want to go brownfield uh and they have a couple of others like Kissa Tech Fusion a couple of other partners that um build brownfield data centers at an extreme speed um and in terms of cost honestly it's probably not the not the best cost structure uh and I think they're trying to optimize that increasingly as of today because now they scale and they have gained sort of that um that trust from many partners. Um that's why they acquired core scientific is because that way they're more vertical. They can own this the infrastructure and such. Uh but they really scaled at an impressive pace uh in 2023 and 2024. Um and so if you think of the I want to add something if you think about the time to build a data center you ask most people that have been in the industry for a while they're going to tell you those are two three four year projects right that's the time it takes to build a data center and if you look at the listed companies think digital realy and others uh that's what you see on their statements is that when they start building it becomes real and generates revenue whatever two years after 18 months sometimes slightly more sometimes slightly less in core if you If you look at their deal with core scientific, things are moving faster, right? They're they're doing that stuff uh in less than a year in some cases. Um and so that speed again speed is really what enabled them to gain those contracts. Um and so really I think they took some pretty innovative ways to think about infrastructure uh that are much more optimized for AI whereas everyone else was still in the cloud mindset. And look, I'll I'll give you another example which is what Oracle did in he um >> can we pause on Oracle because I I actually do want to come back to Oracle. I'm very Do you mind if I ask one more on Cory? Just, >> you know, I I heard a lot of interesting things there, but if I was a bear, I would say, hey, what Jeremy just described was they took on risks that no one else was willing to do, and they really emphasize speed. And that is in an up cycle, that is everything you want, right? But one thing Jeremy said is there is always a down cycle. Like when whenever back in 2022 when people were just first starting to talk about everyone said, "Hey, even Nvidia, there's a down cycle." and all the Nvidia uh like longtimes, they might have missed the big move because they were like things are starting to look good. That's when you sell because there's a down cycle coming and they missed it. If coreweave optimized for they're basically optimized for an upcycle like aren't they going to get crushed the moment a down cycle comes and then all of a sudden they're committed to hey all they've done is accelerate and take as much as they can and they're delivering it as fast as possible and then a down cycle even if it's a blip for six months come and they're just they're just completely stuffed. Am I crazy to think that >> you're not crazy and actually I think that applies also to Oracle which we can talk about in a bit but yes 100% it's a it's a lot of risk and the the simplest way to frame the risk is uh your longest GPU contract is going to be five years uh your data center deal is going to be 15 years uh so if you cannot replace uh those GPUs and find a new contract you're stuck with 10 years of paying rent uh to a data center operator without any revenue so that's the simplest way to frame risk which I agree 100% exists. Um so you can think of it as a bet on on the underlying companies right uh are these companies likely to need capacity in the future? Uh that I think that's the the way the bet is framed, right? And so if you think about I'm exposed to open AI so both Cor and Oracle uh the way you think about it is open AI is not going to fail. OpenAI is gonna need GPUs forever because they they have a business model that structurally requires GPUs on the different side and also they they always have big big training requirements. So that's the way you can frame it is I'm I'm betting on the AI industry to be a long-term consumer of GPUs, a long-term consumer of power and as I secure power, as I build data, uh I can uh I can renew those contracts over time. I I actually have some qu more questions power, but let's go to Oracle. You wrote a piece on Oracle, and I I'm sure you didn't see, but my last book club, I I read Larry Ellison's biography from 2002. And as I was reading it, in the back of my mind, I mean, Oracle stock has been on a heck of a run this year, right? And as I was reading it in the back of my mind, I was kind of thinking, hey, Larry Ellison managed somehow a as a late 70s, early 80s, I can't remember if he's in his 70s or 80s. Somehow he managed to catch the AI wave. Now, he wasn't like crazy early, but in 2023ish, he saw where it was going. He made a big bet and boom, this man who's called so many trends calls this one again. And Oracle's really benefiting right now. So I I I just want to throw that background out and ask just like what is Oracle's AI strategy and why is it working out for them? >> It's corre path. Uh Oracle strategy is I'm going to use my balance sheet uh to get the largest contracts. Um it's it's using their investment grade signature to uh get call it normal contracts. Um let me explain. uh basically corre one of their issues was that because they're not uh investment grade and they're not the reliable hypers scare and such uh they struggle to get capacity from reliable uh data center operators but sorry when I say reliable I mean experienced guys again the digital realies all of these people that have been building forever that have a lot of land that can deliver power um so definitely these guys they love working with hyperscalers list so with Oracle fits in the hyperscaler category so Oracle can easily get capacity from folks like digital and others Uh so that's one advantage that they have which all hyperscalers share to be clear. But what Oracle did in the site was basically going corre I'm going to do a massive bet on this company. Uh which by then like if you look on paper who is Cruso Cruso is a crypto miner uh that never built a data center for like uh uptime related like traditional data server purposes. They built mines and I do think they had amazing engineering teams and such. They're a great company, but back then, if you look at it on paper, it's betting on Cruso was the same as betting on other crypto miners uh today, right? Uh and Oracle took that bet. They signed this contract. Uh and if you think about it's the same situation. They're they're getting a fiveyear deal with OpenAI, but the deal they signed with Cruso, I think the leak numbers were like 15 years, right? 15-ear deal, uh it's probably 15 to20 billion dollars over like those 15 years. Um, so if uh if the contract fails uh after five years and they cannot renew it, they're stuck with like 10 years of paying a billion dollars a year or more uh to do so. So it's it's the same thing, right? It's taking massive massive financial risk uh betting on the success of these companies and if something were to happen to Open AI or to the overall AI geni industry growth, uh you could you could frame a future where Oracle isn't very happy. I I guess I I mean I guess it's easier because even in 2023 this is a three to400 billion EV company. So they're not they are making a big bet but Cororeweave was existential like this works or it the company zeros and it's kind of interesting Oracle made that bet and it was not purely existential right like Oracle would exist if this bet had gone in flames but at the same time you know coreweave as many people pointed out they took the optionality bet right this works and we're 100x this doesn't work and we're zero oracle I mean they had both sides of that bet right if it went up they would they'd benefit as they are benefiting you know the stock is up what probably 50% this year, but if it didn't work, they were on the hook for all those payments. And then they're sticking around and they're saying, "Hey, okay, we uh we're the largest data center people in the world. We've got all this excess capacity and we're going to have to lease it for a song." Um, anything else? >> Yeah. Yeah. I was going to say like I guess you could frame a difference with Oracle, which is that they already have the cloud business. There's always the hope that uh by signing a gigantic deal with OpenAI, you also incentivize OpenAI to use your uh CPU based infrastructure. Um which uh I'm I'm sure the spending that OpenAI does on Azure for traditional cloud services. I'm sure that's fairly high again like to manage like 700 million weekly users. Uh it's not just about models. It's also about like store data storage and traditional sort of CPU front end processing and whatnot. Um so that's also I guess the hope that Oracle has is by signing those GPU deals they can also grow uh their Oracle cloud infrastructure business uh which was like 10x or more smaller than other rival hyperscalers. So I think Oracle like if you look at the history they made this very bold bet on cloud in 2016 or 2017 um and it turned out okay but not not that not that good right like they're still they were still lagging like way behind other hyperscalers growing decently but not like triple digits at a pace that was enough to catch up with the others. Uh so it makes sense also from a strategic perspective to just uh increase the size of the cloud business and hopefully upsell some services. And what we've said in the report is that we don't see a lot of evidence that this is happening uh today. Um and and we don't think it fundamentally has to happen uh because very easy to sort of connect GPU infrastructure with another cloud provider. Uh so we we're yet to see a meaningful benefit of collocating uh or having on the same cloud GPUs and CPUs. Uh but it's always a possibility, right? So they have this option. I don't know if it's going to happen or not. I don't think so. it could happen and if it does of course very positive for their purposes >> and I think the thing that's coolest about Oracle is just like high level as someone who's kind of not a specialist in this but Larry Ellson again he's in his let's just call it 70s I can't remember his exact age but you know Oracle is late to the cloud computing party I mean you just said they started in 2016 2017 I mean you know AWS starts becoming the driver of Amazon in what 2012 Microsoft with the zero like they are late and they miss the boat and they're basically You can tell me if I'm wrong. Darren also ran there. And Larry Ellison in his 70s says, "Hey, we missed the boat in cloud computing. Hypers scaling starts taking off in 2023." And look, there were plenty of people at the time who were saying, "Hey, this is a bubble. Too much capacity is already getting built. You know, where are the returns?" All this sort of stuff. And Ellison says, and I say Allison, Oracle, but I think Oracle is Ellison. They they instantly go all in. And when I read that Oracle autobiography, it fits totally with his personality. But, you know, he goes all in and he hits it again in the 70s. It's just like crazy to have that mental flexibility and that that forecast of the future. I want to ask you a few things about power, but any any last thing on Oracle or anything? >> No, I think the way you frame it is is great. Like, yeah, definitely that was a big bold bet um paying paying off pretty nicely now. >> Power. So, it still strikes me that and we had this discussion in our first conversation, but it still strikes me that for the most part all these data centers are going up. And I don't want to say only domestically, but the US is the hot spot for data centers. And you can tell me if I'm wrong. There are big data centers in Asia. Stargate 2 is in some Saudi Arabia, I think. I can't remember, but for the most part, they're going up in the US. And they're going up, you know, not not in New York City, but they're going up pretty close to uh, you know, urban areas that like obviously connection agency, all that sort of stuff is a star. Tell me if I was wrong on any of that, but when can we start seeing data centers getting built in crazy places, right? Like uh I think there was the Elon Musk tweet about putting a data center up in space. I I doubt we're getting one in space, but you could see the appeal, right? Like you don't have to worry about power as much because you don't have any heating components if any heating worries if you're in space. I'll I'll take one. Alaska, right? We don't have to go build in Antarctica or somewhere with But there's infrastructure in Alaska. It's really cold up there. There's a lot of access. We talked about natural gas. There's a lot. When do we start seeing a big data center get built out in Alaska where land is cheaper, labor is cheaper? I I'm asking I I see you laughing because it's clearly a silly question, but why is it a silly question? >> No, no, I think I don't think it's actually silly. I think it could happen. Uh generally, the problem that you see in more remote locations is labor. Uh can you actually get thousands of people on site? Are the logistics uh good enough to be able to handle uh a project of that scale? that can be one of the big issues. But overall, I I think actually I think what you said in the beginning is inaccurate because I do think we're already seeing very large data centers pretty far away from population centers. Um, West Texas I think is already growing. Look, Abene, I mean, you could say Abalene is is a city uh it's not too far away from Dallas, but I think we're going to see a few massive uh data centers in West Texas uh in the next two, three years uh in more remote locations. I mean, you could also say like Ellenale, North Dakota. Uh, sorry. Apply digital cash, but that's really far away from everything. Uh, >> Ellen's the applied digital site, right? >> Yeah, correct. Correct. >> Uh, the logistics to go there are pretty pretty insane. It's pretty pretty tough to go, but I mean still possible. Um, so I I do think we're seeing already a move away from those metros. A lot of these data centers are being built in areas that are yeah, just more remote, right? But even there like and it's still in do the domestic US right and yeah getting out to I haven't looked at the map but I'm sure it involves a pretty long car ride and a flight into a very small airport but you know Alaska or I I don't think I said Alaska because my first thought was hey super up north Canada right but super up north Canada is there's nothing built out there right so you would have to build out the the gas pipeline and you'd have to build out a fiber connection like Alaska has all that right now. I'm not saying it has it in then in what you need for a 500 billion Stargate or something, but it's got a lot of that. And if it's got a lot of that, you can lay some more fiber and stuff. I I'm just surprised like an Alaska. I know like there were a few people talking about the uh the Nordic regions and I know there's some smaller ones out there, but again, I'm an outsider, but I'm surprised that you haven't seen like something huge getting built in something where uh a little more out the way. you start saying, "Hey, you know, we're we're spending $6 billion on this. Let's pay the data scientists an extra $5,000 to fly out there when we need them or something, you know." >> Yeah. I mean, you might have seen an announcement, I think two or three weeks ago, from Cruso saying they're going to build a massive vid in Wyoming. Um, it's I mean, still, I guess, it's close from Cheyenne, but still pretty like far away, I would say. Um, but yeah, overall, I I I definitely expect to see data going to more uh remote locations. It's already happening today. I think it's going to happen more. I haven't looked at Alaska specifically, so I can't I can't really tell you. >> I was just throwing up that >> actually. One thing we can touch on is there was a question on Twitter I saw that was asking about the perm bassin and so I would I would say it's just a thought. I haven't looked at Alaska again, but the issues with the permassin is that they don't have a grid infrastructure. >> Yes. Ideally, you want to build grid infrastructure because uh it's lower electricity cost uh it's higher uptime. A grid is actually the best power you can have on site. Uh ideally you want to have a grid connection. Uh fully islanded data centers are pretty expensive. It works out again as we discussed before for a fast time to market but then this equipment serves as backup and you want to have primary power coming from the grid. Perassin doesn't have a large electric transmission infrastructure. Uh there's a project going on that's going to be like five years out or something. Uh so maybe that's one of the reason why Alaska isn't considered today. I never >> let me ask a general question on power, right? It strikes me that everything like power is the limiting factor. It is the component. Obviously, time to build these things, but p it's been a race for power and you've seen this in the stocks of all the power players have gone off and uh I just want to ask I I think we touched on this in the first p, but I want to follow up again like if I look at the history of things, power efficiency tends to improve over time, particularly when it's a limiting factor. And I would point you to, you know, the biggest improvements in uh car fuel efficiency come after big oil spikes, right? Uh airline fuel efficiency. I believe jets today are are depending on your source, 50 to 75% more fuel efficient than they were 30 or 40 years ago. Homes today, they take a lot, they consume a lot more power because we use a heck of a lot more power, but they are much more power efficient. just want to ask like at one point do right now we're in the the rush right but could you see a world where hey there's still a rush for uh GPU capacity but two years from now they say all right demand is still growing but it's not growing exponentially let's optimize on power and then all of a sudden you have a world where a lot of these data centers not that they're stranded assets but you know you were building one gig data centers two gig data centers you say hey now that we optimize on power it turns out we need 50% less power and all these data centers are sitting here saying, "Oh my gosh, like there's just no need for us because even though demand's growing, like we just took our cost down because it's the it's the bottleneck and bottle physical bottlenecks tend to get solved over time." Does that question make sense? >> Yeah. Yeah, of course. Makes a ton of sense. I mean, it's always a big fear that we're doing a massive over supply and that the holiday demand is going to go away. Uh, which is completely possible to be fair. It's a scenario. Uh, I mean, we can look at the the cloud computing world, I guess, maybe as a proxy where you've seen electricity cost as a share of revenue for these huge cloud providers go down over time. There seems to be a moment where it doesn't really go down that much at some point. Uh and if like one thing you can do is like just just look at Azure revenue growth and look at what they've published in their ESG reports for example and you see that it actually tracks pretty well. Uh they're growing revenue a lot. They're growing electricity consumption a lot. Um and and basically my point being um the even if the unit of compute itself is improving uh so the CPUs in this case especially the GPUs are getting more and more powerful um like this is kind of the the Jav effect I guess which is you're just going to sell more with those GPUs. Um but like generally speaking, okay, let me put it this way. Generally, when you look at the GPUs over time, um uh if you look at Nvidia's road map, of course, pricing goes up for their GPUs, uh but power cost also go up for the GPUs. And actually, you see power and price tracking correlating pretty well. Um and that's because if you look at Blackwell, like the reason Blackwell is more powerful, there's a couple of things, but one obvious is that it's actually two comput D. It's not one compute D. And so when you have two compute dies, you you don't double the power, but it's actually close to it. U then you have system level improvements and whatnot. But the point being at the hardware level, you actually have a pretty good correlation between price and power of the system. Can I can I just slightly push back on that? I mean, it does strike me that for the past, let's say, 50 years, like since the dawn of the computer industry, power has never really been a constraint, right? And by power, I I'm using electricity. I'm basically saying electricity cost like the cost of power. I'm sure people would have loved to be get more power into the GPUs and everything, but you could build it with the assumption that basically you were kind of treating your electricity and power costs as free, right? And obviously they were not free, but they were so low fragment. Now power is actually the bottleneck, right? Like I think we would have a lot more data centers and a lot if people could just get access to the power and electricity, we would have a lot more capacity right now. uh you know as we go into a world where power has been the bottleneck for since mid 2023 for two years do people start like does Nvidia does Nvidia's next project does it what they're releasing 2026 does it optimize for this power constrained world so actually GPUs go up but maybe a little bit less but power usage goes way down does that makes it is there any road map there I mean if you look at Nvidia's road map what they're doing is as I said price and power track pretty well but actually the throughput of the chip goes up a lot. So in terms of throughput per power, energy efficiency goes up a ton to be clear. Um like if you think about like the system level improvements, it's all about having more GPUs working together and what not to deliver output roughly similar power unlock. So yes, Nvidia is very clearly uh trying to push power efficiency. Uh now the reason they're doing it is mostly because because you can use those uh added tokens to generate more intelligence and so that's where we go back to the debate of like Japan's paradox which is what are you going to do with your extra compute power? Are you gonna use it to save on costs or to increase intelligence? Um, and the path that the leading AI labs are choosing is to increase intelligence because I think what they're saying today is that uh the best way to monetize LLMs is uh to have the single best models. Very simply, Anthropic has a single best model at code and you're seeing everyone use anthropic models. Um and then if you s sort of go down the stack and start looking at the market for cheaper models that have pretty good price to for price per intelligence actually building those models is much cheaper uh because you can use techniques like distillation which is you use uh one model to generate whatever uh data synthetic data and such to dist distill intelligence into a smaller model. Uh all the AI labs are doing that. Um, and so this is where you get into much more competition, right? Like if you think about the market for these mid-level models, like there's a lot of competition from the Chinese, the uh the open source firms, the big labs, but really the money maker is uh is the frontier model. And so that's why we think because of this specific structure, there is an incentive to always use the extra computing power that you get from Nvidia to increase the intelligence of your model. And so you can frame it this way like in terms of physical constraints uh yes price per GPU goes up uh it tries so power goes up as well like that's pretty linear but then the intelligence you get the comput power you get from Nvidia goes this way much higher and the intelligence you get out of the mall also go way way up more in exponential curve I guess does that make sense any push >> it does make sense it does make total sense there there are some follow on questions I want to follow there but I do I'm aware of time and I want to ask one as completely unrelated subject. You guys had a report on robotics that I thought was very interesting. So I'd love to just pause here and uh you can give overall thoughts on robotics and then I had some specific questions. >> Yeah. Yeah. I mean basically the the idea behind the robotics piece was um we just wanted to sort of provide a framework to help people understand robotics markets and so we added this classifications of levels of autonomy which we've seen on automotive we've seen on >> I loved it because you instantly knew right you had level zero to I think it went up yours went up to level four versus level five for driving but as soon as I said I was like oh I can equate it to vehicles I I really like that framing. Yeah, we and what we find out in our research and maybe some people are going to nitpick on some things because you cannot always we cannot make a level that's gonna make everyone happy like some some stuff uh there there's always some nuance is the point but overall what we found is that um it's pretty we can frame it in a way that's easy to understand. Uh and so if you think about our levels like level level zero is like the rigid robotic arm, level one is like the slightly more flexible pick and place arm. Level two is adds mobility. So it's like a robog. Level three is sort of a weak humanoid. And level four is like a strong humanoid. And typically again that's oversimplification. And I hope robotics guy aren't going to kill me when they listen to that. But it's kind kind of the simple way to think about is that there there's these different capabilities that add up over time. Uh and are actually somewhat uh yeah there's there's a sort of a simple way to frame it. I guess is the point easy easy to to understand for people. Um and and basically what we wanted to do with that piece is how people understand where we are. And so if if we think about level four, which is those super humanoids, we're pretty far away still. We're still in the research phase, but we're actually already seeing level two. People don't really talk about that, but all of these sort of uh mobile quadripets like the robog. Uh you're seeing that's uh actually in early production phases. And so that might be actually a trend that could uh uh be interesting in the next one one to two years >> which is the the overall question I had on robotics was how quickly do you think we accelerate in robotics right because I mean AI I you know maybe people are starting to get disappointed with the chat GPT5 versus chat GPT4 but if you were sitting here four years ago and talking about where we are with AI I I think most people would have their mind blown you know chat GPT wasn't even out at that point I think folks who how quickly does it robotics start accelerating because it does strike me like Tesla controversial the robotics are out on the field but you're starting to see more of the robotics get out into the field I saw a uh there was like a fight le a robotics fight league like you're starting to see more and once you start to see more it tends to accelerate especially with the AI did so if you and I were talking here in like four years do you think we're seeing level three out there if we're talking in 15 years like how quickly do you think this is accelerating >> yeah I'd say currently we don't have any evidence that it's going to happen as fast as we've seen with uh with LLMs and one of the big reason >> business are just harder than business are just harder than anything else. Yeah, >> correct. Right. And in terms of data like that there's always the data issue uh which is you yeah text we have the internet data like trillions of tokens. It's pretty hard to generate high quality data for robotics. So that's one of the big bottlenecks in increasing capabilities. So this is being uh currently solved and I do think progress is starting to accelerate but I just struggle to see a world where uh it gets as fast as we've seen with LMS right and so if you think about it like LMS as leaders love to say like today you have some models that are nearly as smart as whatever like advanced university students um and three two years ago it was like a dumb dumb kid no fast kids I like when you say it's nearly as smart as an advanced university student because I've given chaty pet or whatever research projects and I've had it uh spit out just like brilliant insights, right? It's digested scientific papers in half a second and spit it out in language that yeah I like to say like hey I'm someone who hasn't taken a science class since high school. Please put it out to me in a language I can understand and they do that. So you get these brilliant and then I also like yesterday it was going again how many bees are in blueberry and it says I would bet my life there are three bees in blueberry and you're just like man on one hand it can understand science so advanced university student on the other hand it might be as dumb as my kid it's it's so funny. Yeah, I know for sure. But I'm definitely with you like I'm a huge user of like deep research and that kind of tools. Um, so I think it's incredible in terms of robotics anyways to go back there. There are a few additional challenges. So I think it's going to accelerate but not to the same extent. Um, so I don't know. I'm not I'm not the main robotics analyst. That's something analysis. I don't want to make a crazy prediction. Uh, but let's let's say level four in the next few years. Seems unlikely to be right now. Wait, >> it's just it's funny when you you look at this and this gets more into dystopian future. Maybe we need to have a sci-fi writer on, but like you know, if they're as smart as uh as smart as university students and you're already hearing lots of I I think it's overblown, but you're already hearing lots of consulting firms like struggle and taking consulting jobs and then you're saying, "Hey, if if level four is here, I mean, how long till it starts taking low wage like kind of manual labor and you're like, Jeremy, what are we going to be doing in six years?" you know, like we're not we can't use our hands and it's smarter than us. What are we going to be doing? >> Yeah. I mean, there's this great analogy. I think it's interesting where you think about accountants uh like I don't know before Excel, before spreadsheets. Um and like I've talked to people that told me like back then like people were thinking spreadsheets were going to kill the accounting uh jobs, right? And actually you've seen accountants like go up a lot. It's just uh they're doing new types of work. They don't have to do all the manual calculations and such. they can let the machine do that and they can use their brain in some other ways. So yeah, I guess that's just like the framework I would say >> which I mean the one people point to now is law firms, right? And you say hey this is going to take the work of a lot of junior analysts but at the top like we have no lack of legal law and legal cases are accelerating but you do wonder like or there's the famous what is it the uh they tried to ban the sewing machine because it was going to replace all the the seamstress jobs in Britain. Uh it's just it's a little scary, right? Like I tend to be pretty optimistic and say, "Hey, there's going to be jobs and it's going to create really new interesting ones." But then when you start getting to, hey, you've never had something that can replace human thinking before and you still needed a human while it could automate basic tax. You still needed a human to like, you know, go turn the flip the French fries over and then hand them out to uh the the hand them out to the customer. And if you've got on one hand robots take the French fry machine and on the other hand computers think better than uh McKenzie consultant it's like what is left for us to do it is >> yeah I don't know I guess maybe more more relationships like uh this the the value of social relationship probably goes up over time because at some point you trust humans more than machines for certain certain things. Uh so maybe yeah maybe that's the future. You got to be sure to have to be well connected to have a good network. That's what it's a good thing we're so handsome. That That's the one thing robots can't take. You and I can hop on and we could go be fashion models. Robots can't take that for now. Uh Jeremy, this has been great. Uh look, the first one was a hit. I think this is going to be hit. Maybe we'll have to have you on like before the end of the year or something to do like outlooks since 2026 or something. Talk about how high that capex spend is going in 2026. >> 10 trillion dollars. No, >> 10 trillion. Wow. That I said it wasn't accelerating. We're getting real acceleration. This has been great. I'll include a link to semi analysis in the show notes and uh we'll go from there. >> All right, man. Always a pleasure. Thanks for having me. >> A quick disclaimer, nothing on this podcast should be considered investment advice. Guests or the hosts may have positions in any of the stocks mentioned during this podcast. Please do your own work and consult a financial adviser. Thanks.