1. Podcasts
  2. Big Technology Podcast
  3. Can AI Keep Growing Exponentially? Let’s Ask SemiAnalysis — With Dylan Patel and Jordan Nanos
Big Technology Podcast

7 Oct 2026 · 1h 2m · 21 chapters

Episode artwork

Can AI Keep Growing Exponentially? Let’s Ask SemiAnalysis — With Dylan Patel and Jordan Nanos

In short

Whether AI infrastructure can keep scaling sustainably (including AI CapEx as % of GDP, GPU cloud growth rates, and risks like supply-chain limits and security). Also covers SemiAnalysis’s ClusterMax 3.0 rankings of GPU cloud providers and NVIDIA’s role via “backstops,” plus Anthropic IPO valuation and AI business economics.

Guests

Dylan Patel, CEO/founder of SemiAnalysis (AI infrastructure research/ClusterMax). Jordan Anos, technical staff at SemiAnalysis; author of ClusterMax 3.0.

Key claims

AI CapEx is likely higher than widely cited charts (Patel: US/North America CapEx near ~$2T next year; AI capacity serving other regions makes GDP share look understated). Hyperscaler CapEx growth will decelerate as “big numbers” hit limits, but AI revenue/profit should accelerate later. Revenue requirements estimates are too low because revenue trails CapEx and depreciation timing differs; Anthropic/OpenAI are expected to reach profitability soon. NeoClouds’ main risk is demand/supply mismatch and operational constraints, not easy contract cancellation. NVIDIA backstops total ~$588B off-balance-sheet support to NeoClouds.

Notable examples

Europe/Asia inference served from US data centers; Anthropic revenue growth (sub-$10B to >$10B, then >$100B). Oracle praised for managed-cluster quality but faced pipeline delays (New Mexico) affecting OpenAI contracted capacity. Security concerns: Ilya Sutskever tweet after Hugging Face incident; SemiAnalysis claims many NeoClouds have weak security and cites accidental client visibility on a cluster.

Written by AI. May contain mistakes.

Chapters

1:12

Introduction of Guests

Introduction of Dylan Patel and Jordan Nanos from Semi-Analysis.

“Welcome to Big Technology Podcast, a show for cool-headed and nuanced conversation of the tech world and beyond.”
1:54

AI Infrastructure Scale Compared to History

Analyzing the scale of AI infrastructure relative to historical benchmarks.

“Let's just start by talking about the scale of what we're seeing in this AI infrastructure build out if we can.”
2:41

Real Numbers Behind AI Infrastructure Spending

Dylan shares insights on the actual CapEx numbers being invested in AI.

“I'd love to hear your perspective on just how big this is whether that's the right way to look at it and then what it means if we're building out something this big.”
4:43

Trends in Hyperscaler Growth

Jordan discusses the growth trends of hyperscalers and their CapEx growth rates.

“All right, Jordan, and you are the author of ClusterMax 3.0.”
7:40

Future Revenue Projections for AI

Evaluating future revenue requirements and projections for major players in AI.

“And I think the biggest thing right now is twofold.”
11:44

The Role of AI in the Economy

Discussion on AI's potential impact on the economy and revenue generation.

“They could pull the plug whenever they feel like it.”
14:00

The Future of AI Revenue Growth

Discussion on potential revenue sources for AI technologies across various industries.

“So, yes, AI revenue needs to grow a lot.”
15:10

Labor and Capital in the AI Economy

Exploration of how AI will impact labor share versus capital share in the economy.

“And so I think people lack the creativity to understand how returns can be viewed beyond the like clawed code subscription and per token pricing regime of today.”
19:40

Impact of AI on Knowledge Work

Insights on the future of knowledge work and its relationship with AI advancements.

“Dylan, you've said that, I think in the past, you've said something to the extent of knowledge work is F'd.”
20:50

Risks and Challenges for AI Companies

Examination of potential risks faced by neocloud companies in their contracts and projects.

“Jordan, you know, this is kind of in your area of expertise.”
Show all 21 chapters
24:10

The Implications of Anthropic's IPO

Discussion on the potential implications of Anthropic's IPO on the future of AI.

“So anyway, with that distinction in mind, what's the risk with project cancellation or a downturn or things like that?”
28:00

The Optimistic and Pessimistic Future of AI

A discussion on the potential future growth of AI companies like Anthropic and the challenges they may face.

“And therefore, like, if you believe Anthropic will get to 20 trillion, you know, society as a whole could be fucked.”
37:08

NVIDIA's Role in Supporting AI Infrastructure

Exploring how NVIDIA backs AI infrastructure through financial support and partnerships.

“And we're back here on Big Technology Podcast with Dylan Patel and Jordan Anos from Semi Analysis.”
42:06

NVIDIA's Compute Revenue and Supply Chain

Learn about the small portion of NVIDIA's revenue linked to their internal compute and the broader supply chain dynamics.

“And then the dynamic that we've seen is that this amount of compute is actually not a meaningful portion of NVIDIA's revenue.”
43:10

Cybersecurity Concerns in NeoClouds

Discuss the implications of Ilya Sutskever's tweet on cybersecurity within NeoClouds and the potential risks.

“Ilya says, Neo clouds have limited cybersecurity.”
44:48

The Sovereignty of AI Infrastructure

Explore how nations view AI infrastructure as a strategic asset and the implications for cloud services.

“You can get open source models to just hack NeoClouds or get access to stuff within NeoClouds.”
46:50

Ranking NeoClouds: Oracle and Google Cloud

Examine the ranking of various NeoClouds, focusing on Oracle and Google Cloud's positioning and performance.

“And so I think it's actually no surprise that federal government research and intelligence agencies are on some of these clusters sharing infrastructure with others.”
49:19

Oracle's Cloud Strategy and Execution Challenges

Analyze Oracle's cloud strategy, including their AI data center deals and the risks associated with execution delays.

“So you kind of have to do the rankings and set certain boundaries.”
55:08

DeepMind's Compute Allocation and Future

Discuss the shifting compute allocation at Google and implications for DeepMind compared to its competitors.

“Like the basic things that fail with respect to security and reliability being the two top ones are just, in our view, dead simple.”
56:00

Analyzing Google's AI Compute Strategy

Explore Google's evolving AI compute strategy and its implications for DeepMind.

“to model training and deployment in the last six months than they had two or three years ago.”
1:02:52

Sign Up for SemiAnalysis

Encouragement to subscribe to SemiAnalysis for valuable insights.

A man walking through an airport, asking his phone a question

Ask about this episode

Get the gist, explore an idea or ask about something said.

Claude

On Big Technology Podcast, what are the key points of this episode?

Searching your listening

Transcript

Automatic transcript. May contain errors.

0:00How big does AI have to get to justify the trillions in infrastructure spending? Let's talk about it with the leading authorities from Semi-Analysis right after this. This episode is brought to you by Genesis. What does it actually take to put agentic AI to work across an entire enterprise? I recently spoke with Genesis chairman and CEO Tony Bates about why orchestration is emerging as a competitive advantage in customer experience and what companies need to get right in the AI era. Then, I sat down with Adam Mitchell, head of enterprise business solutions at Voya Financial, to go inside a large-scale agentic CX transformation, including the business decisions, trade-offs, and lessons along the way.

0:41You can watch both conversations now on my YouTube channel today. It's never been more challenging to get Frontier Compute. How do you know where to look, who to trust, or how much to pay? Akash is an open cloud that provides easy access, flexibility, and peace of mind. That's how companies like Razer were able to cut their inference costs by 15 times. Visit akash.network.com for a free consultation. That's akash.network.com for a free consultation. Welcome to Big Technology Podcast, a show for cool-headed and nuanced conversation of the tech world and beyond. The AI infrastructure numbers just seem to be getting bigger with no end in sight.

1:23So we're going to talk about what the physics of this entire buildout looks like and whether it can continue to sustain itself without maybe not destroying the world, but destroying the economy. Because we have some very interesting guests to speak with us about it today. We're joined today by Dylan Patel, the CEO and founder of SemiAnalysis and Jordan Annos, member of the technical staff at SemiAnalysis to talk all about everything going on within the infrastructure buildout. Dylan, Jordan, welcome to the show. Thanks, great to be here. Let's just start by talking about the scale of what we're seeing in this AI infrastructure build out if we can.

2:02Dylan, there's like a chart going around X right now that compares the AI infrastructure build out to previous build outs in the past like railroads, like highways. And the chart is showing that the percent of GDP for the AI build out today is about 3.6 % oh sorry the median is about 3.6 % the average year compared to 2.2 % for the railroads and basically the commentary is that this is far bigger right that would put it basically 50 % bigger than railroads even in terms of its share of GDP and the commentary is this is bigger than anything we've ever seen before on a scale that is you know effectively unprecedented.

2:41I'd love to hear your perspective on just how big this is whether that's the right way to look at it and then what it means if we're building out something this big.

2:50Dylan Patel: Yeah, I mean, the first thing when that chart was going around, I was like, well, people's numbers are too low. It's actually much higher than that already. Really? Okay, so share the real numbers. Yeah, yeah. So, like, I mean, if you look at, like, next year, CapEx across the U.S. will be on the order of$2 trillion, not just data centers and chips, but also all the rest of the supply chain that people are investing in. So you're close to$2 trillion of CapEx. And then you stack on top of that, GDP is like close to 30, 30 and some change trillion. So you're at like, you know, five or 6%, right?

3:28Dylan Patel: Not 3 % or 2%. So sort of my thought process is actually like, oh, wow, these numbers are lower than the actual because people are understating the amount and size of CapEx that's happening in the US and North America and the world. And I think like part of this is because, you know, with railroads, your capex is in the country that you're deploying it in. With fiber, your capex is in the country you're deploying it in. But for AI infrastructure, your capex in America is actually a lot of it's being used to serve outside of America uses. Right. So if you look at, for example, Europe, most of Europe's inference for anthropic is in America.

4:08Dylan Patel: And most of open AI's inference in Europe is in America. um and and similarly if you look at a lot of asian countries um and so there is a bit of this like this this you know capex in america is actually like much higher as a result um because one you're doing all the training here and then two you're actually serving other countries from data centers in america um and so ultimately you're sort of at this stage where capacity is much the economic impact of it is much larger than any other sort of physical infrastructure investment cycle ever before. All right, Jordan, and you are the author of ClusterMax 3.0.

4:47This is semi-analysis rankings of the GPU cloud systems, basically putting a lot of the hyperscalers and the Neo clouds in tiers, talking a little bit about their performance. And we're going to get to that in a bit. But, you know, you do look across the board in your report, and this is something that I'd like you to comment on. Basically, we're looking at where the hyperscaler quarterly year on year CapEx growth is going. And in the third quarter of 2026, the growth is at 116 % year over year, which is astonishing. The consensus estimates, so what Wall Street is expecting to come after that.

5:31They dropped down in the fourth quarter to below 100 % year-over-year growth, and then the first quarter next year in the 70%, and it continues to drop from there. And I think this is a cycle that we're used to, right? Where it's like, well, we're on a bigger base, so it can't possibly exceed expectations again, and can't possibly double again. So I'm curious to hear from your perspective and your research, do you think that we're actually going to see that trend? Like the growth is going to decelerate or are, you know, are we again falling into this trap or is wall street again falling into this trap where it's like, all right, you've, you're basically at the biggest numbers you could possibly imagine, but you still need to think bigger.

6:13Yeah. I think we're definitely going to hit limits towards the end of next year

6:17Dylan Patel: and into the Jordan, before you answer seriously, I got, I got it. I got, you know, it's like when you ask a kid, what's the biggest number they can think of. And they say like 1100, right. And now, And it's like, no, no, no, no, we're at a trillion, babe. Yeah. Yeah. Four commas club. So wait, Dylan, in this example, the kid is Wall Street. Not making any statements that people are children. Just, yeah, go ahead. Sorry. Dylan may have a habit of referring to his employees and his customers as kids sometimes, but that aside. I think I use a much more mean word.

6:59I like where this is going already. We're running out of big numbers. There are things to look at when it comes to the build out. This could be everything from industrial equipment to firms that are prepared to do EPC work, like engineering for construction, to chips, to cooling systems, to all the components that go outside of that and IT estate. Basically, there's a lot of work being done to expand the supply chain in order to be able to accommodate for all of this capex to go into it. And a lot of it's working. And so we're going to see massive growth. But, yeah, you can't just have triple digit growth and accelerate the second derivative, you know, even further from where it is right now.

7:48You run out of things. And I think the biggest thing right now is twofold. One is like it becomes really hard for the biggest neoclouds in hyperscalers to just organize and marshal the resources required to build another gigawatt on top of what they're already trying to build, for example. And it becomes pretty hard for them to raise the capital, even though there's incredible returns on a lot of these investments already. You still just need to find the capital to pour into these projects. And that's a difficult thing to do when you're when you're talking about numbers in the trillions of dollars.

8:23Dylan Patel: I would say the other thing is, like, by definition, the numbers have to shrink. You know, this year we have 10x growth, Anthropic going from sub 10 billion dollars of revenue to north of 10, north of 100 billion. So they had 10x growth. Definitionally, they're just not going to have a trillion dollars of revenue by the end of next year. So, yes, there's some decelerating growth there. um you know nvidia's revenue uh growth is still 2x year on year but it's you know it's not as much as it was at an earlier point um just because it is definitially harder to grow 2x off the same base so 3x off the same base so it's sort of like um the same applies to like other firms so it's it is it is definitially like definitely decelerating okay so the growth on spending is going to need to decelerate at some point, but the growth on revenue and profit, for that matter, is going to have to accelerate beyond.

9:16It's basically going to have to trail this big infrastructure build out and accelerate probably beyond where we've seen the dollars being spent. I mean, you would imagine to facilitate all the AI action. And there's been further conversations that have circulated about this. And Dylan, let me just throw these numbers out to you, right? So this is a Goldman Sachs report. And sorry, so Goldman Sachs, they say Meta, this is a tweet, right? Meta's committed corporate suicide after taking on insurmountable debt that the company will struggle to pay back. There's Bain. Bain says hyperscalers will need $6 trillion in revenue, not ever.

9:53This is per year by 2031 to break even. And there's a Columbia professor Sten Van Newberg, who said that America will need to spend roughly 3.5 trillion annually on AI services by 2032, 8.8 % of GDP to justify today's massive data center investment. What do you think of these sort of assessments of where the spending and where the revenue needs to be to pay off the investments? And do they have a chance at making good on those investments?

10:27Dylan Patel: Yes. I mean, I think everyone's numbers for revenue requirements are quite low, where this build out is going to go. I think revenue will be much higher than that by 2031, for example. I mean, there's a couple of points here, right? One is how much are people spending on capex and then revenue? And the revenue always trails to capex. These infrastructure investments are six years on an accounting basis, at least depreciable assets. Although there's quite a bit of evidence that they're actually going to last longer than that. But six years is let's just continue to use six years. So if I spend a trillion dollars today on AI data center and GPUs, the GPUs are lasting six years, the data centers are lasting 15.

11:09Dylan Patel: But that means definitionally in terms of revenue, either I could assume it's a straight line, right? So then over the six years, I need to call it 1.5 trillion of revenue. So$250 billion of revenue a year for that trillion dollars of capex or it's going to be something sloped where the first year is much less than the last year is much higher to get the payback. As far as like, you know, like statements on Meta, it's like this is this is just kind of silly because Meta at any point can just stop buying new stuff and their cash flow can easily pay for all the stuff that they've signed. Right. They just don't have to, you know, they don't have to build a hundred billion dollars of data centers next year.

11:48Dylan Patel: They could pull the plug whenever they feel like it. There's obviously contracts that they've committed to, but there's also a lot of stuff they haven't committed to yet. And so as far as like the solvency of meta, like they're completely solvent, they're completely fine. And then the question is like, what revenue do they make on top of that? And as far as like AI's impact in the in the US economy? Yeah, I mean, if AI revenue is only$3 trillion in 2031, then yeah, we'll have big problems. But, you know, I'm like of the belief that we can get to like 600, 700 million across open an anthropic alone by the by December of next year.

12:26Dylan Patel: I'm really in that you can get to, you know, one and a half trillion, maybe even by 2020 end of 2027 with just those two companies. So that's end of 27. So then the payback is clear, right? You know, where they're getting that from. Ultimately, the capacity that people are building out is matched to what they think business is coming to. And so there's always going to be a mismatch if we're in the building phase of how much infrastructure are you spending on versus what's the AI revenue. Because that AI revenue for today is really based on the infrastructure you spent in prior years. And right now, CapEx is not going up at 10x a year, but revenue just went up 10x year on year for the AI labs.

13:09Dylan Patel: And so we've seen Anthropic is now profitable in terms of compute cost minus revenue or revenue minus compute cost, both training and inference combined. And OpenAI gets there within the next couple of quarters, maybe two or three quarters. So based on at least what we expect in the tokenomics model. So I think those companies are already breaking even. And if you look at like Amazon, for example, the biggest builder of AI infrastructure, all of their AI infrastructure investments are profitable today. Now, you know, the customers that for them could go insolvent, those customers being open AI, Anthropic, et cetera.

13:49Dylan Patel: But right now they are solvent on those, you know, making very strong gross margins. In fact, AI has made their gross margin go up in the most recent quarter from bedrock and AI revenue from bedrock. So, yes, AI revenue needs to grow a lot. But I would posit that being able to automate all the stuff that we've been able to and being able to build all the stuff that the world has been able to is going to generate meaningful revenue and surplus for the consumer ultimately. Yeah, I think we need to stretch it a little bit in our minds to understand what we are going to measure as AI revenue going into the future.

14:25if these companies really do have the best models in the world that are capable of effectively all knowledge work, you can imagine them going into a whole bunch of different industries beyond coding and having returns on their investments similar to what they've seen in coding. So you can look at things like robotics and self-driving vehicles. You can look at drug discovery. You can look at just health and life sciences generally. We can take this across all of these industries and say, if they want to bring a product to market like a cancer drug or like a self-driving vehicle, you can just imagine how much revenue they can potentially return based on the investments they have today.

14:59And if they have the AI, the system to be able to produce the knowledge work required to build that product incredibly well, I think they're the only ones that can really compete on the frontier of being able to do these things. And so I think people lack the creativity to understand how returns can be viewed beyond the like clawed code subscription and per token pricing regime of today. Yeah, there's this chart that looks at how much we're spending as a share of GDP. This is just 2025. So a share of GDP, health is 18%. That's the top category. Food is 9.1%. AI services is 8.8%. It's the third category in 2025.

15:38I imagine it's going to lap or it will pass food in 2026. So guys, I'm just curious, where's this money coming from? Like, you know, let's say OpenAI and Anthropic combined get to$700 billion in revenue next year. Is the money coming from these companies labor pools? Is it just additive because they can sell more? Where is the money that these companies are, the customers of these companies, where's that money coming from that they can funnel it into the AI companies?

16:12Dylan Patel: It's a few factors, right? First off is obviously companies are going to be able to accelerate what they're building until they can invest more. Increasing share of R &D dollars will go towards AI. Increasing share of cost of goods sold dollars will go towards AI. But also you'll have a significant amount of money sort of going, coming from, you know, labor as well, right? Labor. And I don't think labor shrinks per se. I think the rest of the economy grows so fast, at least in the short term, that labor sort of stays, you know, flattish. And so, like, you know, in the past, we've had different cycles across the economy.

16:51Dylan Patel: But one longstanding one has been labor share of GDP has fallen over the decades. And it's been a steady slow line since the 70s. And so a lot of the, like, strife that people say, oh, America's economy sucks. And yet, oh, look, the economy is bigger. It's actually, well, like, labor share of economy has shrank. And that's why. And in reality, a lot more of the labor, the value is accruing to capital. What AI does is it's going to supercharge that trend, long term trend of labor share versus capital share, because AI tokens are generally basically just capital, you know, in the form of chips and data centers that are sitting there and creating labor.

17:28Dylan Patel: And so that labor share is going to fall even faster. And so, you know, I would hope and for society, that's my belief, you know, and maybe it's like a irrational belief, but I believe that we will just grow the economy so fast that labor in terms of dollars will not shrink. and it might even grow slowly. It's just that the rest of the economy grows so fast. And that's sort of where it comes from, right? So all these companies that do offer various services, their services are going to get better. They're going to be able to charge more. They're going to be able to sell more. And they're going to be able to do it with fewer people.

18:04Dylan Patel: Or another way to look at it is they're going to have the same number of people and their business just going to get better. And there's a number of industries where this is possible, right? You know, obviously, coding is an area where people have done it quite a bit. Obviously, general white collar work is getting rapidly automated, but also companies are just able to do a shitload more than they were able to do in the past. But there's a lot of industries like pharma and trading and chip design and such. Right. Like chip design is a great example. Right. Chips. The number of American R &D engineers working in chip design has been basically flat for 20 years.

18:42Dylan Patel: And yet the economy for it has exploded. Now, obviously, manufacturing isn't, we're not counting manufacturing, we're just counting the designers. But the designers are like, you know, two of the most valuable companies in the world, right? NVIDIA, Broadcom, as well as like Google and all these other companies that are making their own chips and many others, right? These designers have not really grown that much in terms of number of people. It's been flat as an industry for US. And yet the value that's accrued has exploded. And why is that? There have been a lot of tools. Now, what does AI-a-stitched chip design now do?

19:11Dylan Patel: Well, now all of the chips that are across the economy get better and better. Your phones are going to get better. Your AirPods are going to get better. The chips in them. Chips for robots are going to get better. Chips for obviously AI are going to get better. Chips for all sorts of things, Bluetooth and GPS and all these other things. Sensors for gas leaks or whatever that Bosch makes. These are all just going to get better because AI can help you design chips. You do it with the same number of people. Revenue grows. Margins grow. Everything looks fantastic, right? And that's sort of like, there's many industries where this can get replicated, in my view.

19:45Dylan, you've said that, I think in the past, you've said something to the extent of knowledge work is F'd. So do you no longer believe that? I think most knowledge work, okay, like, let's separate like my belief in like Looney Tunes

20:00Dylan Patel: 2035 versus like, you know, 2028. I think for the next X number of years, as long as AI is very powerful, but it is not better than humans in most every way the most high agency people are going to be able to do the coolest stuff they're going to be able to make way bigger impacts on the world now in the case where AI is just better than everyone that could change drastically then knowledge work is screwed and actually like, you know, what matters is like people being able to like play games and have fun, people being able to act, people being able to like, you know, do the things that humans do that humans only appreciate in each other.

20:46Dylan Patel: If the AI doesn't kill us all. Right. But that's sort of like, you know, we won't go there today. Jordan, you know, this is kind of in your area of expertise. I want to get started talking a little bit about the risks to like what might derail this story. and you know we'd already talked about the fact that there are some contracts that a company like Meta could you know may not have to go through with you're obviously looking a lot at the businesses of the neoclouds which are dependent on companies like Meta actually making good on the on the promises that they've they've put forth to build with these companies is this sort of an underappreciated risk that like everything looks like fairly fine right now if you speak to one of these neoclouds because they'll tell you that they have the best customers in the world uh they have the metas and meta will make good on the on the um on the promise and they're raising debt based off of this uh but in reality the risk might be that a company like meta might just decide yeah we're not going to do that and that the contracts might not be as binding as uh you the neoclouds tell us they are okay so i want to be clear about one thing dylan was talking about there in terms of termination versus what's going on with meta and the neoclouds which is that generally speaking these hyperscalers or you know large frontier labs who are signing up with contracts for the neoclouds do not have arbitrary cancellation rights on like a monthly basis where they can just pull the plug and the contract's done uh generally speaking they're signing five years six year term links and they're off taking the full capacity and they have a set But hourly rate and total contract value, that's very, very attractive to the NeoCloud providers because those guys have gone out and secured land and power and built a site and even in some cases purchased the chips prospectively before they've actually gone and done that contract.

22:33What Dylan's referring to with the potential to slow down or pause or cancel projects is more with respect to the self-build sites, which is to say like Meta has massive campuses that are going to serve as more of a base load of capacity going into the future, which they can either choose to buy chips for or just leave as an empty shell for years if they want to slow down and not go as hard as they can. And generally speaking, construction costs, while very expensive, are a small percentage of the total cost of a new chip project. Like, in other words, the concrete and the steel and the walls is quite, you know, it's important and big and, you know, very real challenge to get it done.

23:14But on a dollar basis, like the HBM and the optics and the GPUs themselves are the bigger portion of the bill. And so what it's looking like right now is that all of these big companies are really needing to contend with the massive growth that they're expected to pursue by their demand from their customer base. Either that's Meta or OpenAI or Anthropic who have demand on their consumer-facing applications like a ChatGPT or CloudCode or Codex or even Muse from Meta, things like that. In order to serve that demand, they need GPUs and it's profitable to serve the demand. So they need to ramp up and be prepared to roll this thing out to the billions of monthly active users on the meta, different, you know, surfaces that they serve and approaching a billion of monthly active users for ChatGPT and hundreds of millions for Cloud Code and things like that.

24:10So anyway, with that distinction in mind, what's the risk with project cancellation or a downturn or things like that? I mean, in order for us to forecast some sort of change in the demand supply curve there, you'd really need to see demand slow. There are hard constraints on how many chips you can produce, how many data centers you can build, and how fast this stuff is going to come online. And as of right now, all of the modeling we can do basically shows that demand outstrips supply. And so is demand 10 % more than supply right now? Is it 50 %? Is it infinite? I mean, we don't get to answer that question because we are not able to catch up to demand with the supply of GPUs that you can bring online right now.

24:59And so when that question gets answered, if it gets answered, if we find limits to the demand for knowledge work, to the demand for intelligence from the economy, then we'll be able to figure out who needs to slow at what time, where, and how that plays out. Yeah, but so far, I guess those limits haven't been found. So, all right. Before we get into ClusterMax, one last topic I want to cover. Big IPO on the way, which is, of course, the Anthropic IPO. Dylan, you had a very interesting tweet about it. I'm just going to read it. You said, you can't invest in the Anthropic IPO at$2 trillion because it could be zero and I'm fucked, or it could be$20 trillion, but at$20 trillion, we're all fucked.

25:40Can you explain that and tell us a little bit about what you mean by that?

25:45Dylan Patel: To be clear, it's entirely a shitpost. I was like listening. You know what? I get it. But I think there's some truth to what you're saying at the high end. So talk to us about it. I'm generally extremely bullish. Anthropic, I think. Anthropic will make you, you know, be a very successful company in the world. Anthropic could even be the first company to be a$10 trillion valuation company in the world. I'm that fucking bullish Anthropic. But at the same time, for them to be a$10 or$20 trillion valuation company, especially 20, first of all, their multiples on revenue are going to have to collapse quite a bit.

26:24Dylan Patel: So they used to raise at 30x, and now their IPO is going to be, let's call it they're at$100 billion of revenue, and they're going to maybe be at 20x. you know if they get to 10 trillion dollars i bet you they would need to be at like a trillion dollars of revenue um or something crazy right like that right um which could happen at some point in 28 um or 29 um you know it depends of course like the stock market can also go crazy and like make multiples go crazy but you know in the world where they're doing you know where they're worth 20 trillion you know i don't think their revenue i don't think they're even at a 10x sales multiple now they're probably at like a 5x or even less um in which case they're you know if they're at$20 trillion of revenue, they're at, are they at like two, three,$4 trillion, $20 trillion valuation, two, three,$4 trillion revenue.

27:11Dylan Patel: And at that point, you know, what is the state of AI? If on a$150 billion global economy or trillion dollar economy, they're generating multiple trillion dollars of revenue and by far the largest company in the world, like AI's capabilities are freaking insane. In which case, like, you know, all of this worry of like whether it's AI kill us all or AI so good that everyone loses their jobs. And because distribution of that earnings is going to be heavily leveraged to the people who own anthropic stock or, um, constituent supply chains of it and not distributed well, then people are going to have disrest and like overthrow governments because all of the growth is accumulating to a very small number of people.

27:56Dylan Patel: So like these things could like tear the fabric of society apart. And therefore, like, if you believe Anthropic will get to 20 trillion, you know, society as a whole could be fucked. Jordan, you have similar worries? I think, yeah, I think I think there's a lot to do to get to that point. Like, I actually don't think I'm necessarily as worried as Dylan. I'm optimistic, by the way. I just painted the pessimistic picture. I'll go optimistic. Yeah, I mean, like the the rosy optimistic view is kind of what I described earlier, which is like if they're getting to 20 trillion dollars of revenue, This is not done on a per token metered basis for people using cloud code as a product.

28:35And I don't think, you know, you go and see cloud code go 20x better from where it is right now in terms of like quality. And that drives the revenue. I think that they get into other areas of business. And I think I would trust Anthropic quite a bit to bring, like I said earlier, a cancer drug to market or something like that. And we've seen similar stuff happen recently with GLP ones just be such a meaningful thing for so many people in the economy. And as a result, you know, companies like Lilly have have had massive returns on that incredible research that led to those incredible drugs that help people.

29:10And so I think I would be more optimistic for the future where theoretically Anthropics at a 20 trillion dollar valuation. um but i do think that there are a lot of uh going to be a lot of bumps along the road along the way um which is to say going back to your to your question previously like how can they possibly serve all of this demand and train these models and uh you know have intelligence diffused to all of these different parts of the economy so that people can realize the benefits and i think the way you do that is to build compute and it so depends on you know the nitty-gritty details of construction and installing GPUs and running them reliably and all of that.

29:53Like, to me, the center of the world is the Neo clouds on that road from a$2 trillion IPO to a$20 trillion valuation in the public markets. Yeah, Anthropic, kind of an interesting business case, right? They're spending all this money on infrastructure. Even in the S1, you saw big losses accompanying their revenue. Although, Dylan, as you mentioned, it looks like what they've done is they've turned the tide a bit and they're making more money right now than they're spending on compute. Talk a little bit, because I think the belief here is that a company like Anthropic is a low margin business or at least much less than software, but that doesn't actually seem to be the case.

30:36And you guys have run the numbers.

30:38Dylan Patel: I would say like if you look at like Salesforce, which is probably like one of the greatest SaaS companies out there, you know I mean I think like at least like consistently grown to become one of the biggest SaaS companies their gross margins are like 75 percent and Anthropic on their inference is doing 75 percent gross margins a little bit higher even you know especially on the frontier models so it's sort of like well and it took it took Salesforce quite some time to get there right they were not at 75 percent when they IPO'd you know they took quite a bit of time actually they were but then And then in the late 2000s, I think they were not quite there.

31:19Dylan Patel: It took Anthropic quite a bit less time to get to that strong levels of gross margin. And if you look at many other SaaS companies, they're not even that high. And then the other thing is their customer acquisition cost is insane. If you look at a SaaS company, it's like, fine, my COGS is really low, but then my customer acquisition cost can be as high as like 30%. And Anthropic has near no customer acquisition cost. Right now, obviously, they have a humongous R &D spend, but their R &D spend has grown slower than their revenue. And so now they've reached profitability. And there's no reason why that trend doesn't continue, right?

31:57Dylan Patel: And they continue to grow more profitable. So at least operating margin-wise. And then scalability is the other thing, right? Like the beautiful thing about a lot of SaaS companies is even if your customer acquisition cost is so high, you know, as long as your margins are high enough to pay for it initially, eventually you don't have to reacquire that customer. Now you've got this super profitable business. But with like the AI labs, their business seems somewhat sticky. Yes, there's all this like messaging of like, oh, you know, Meta's cutting spend or Microsoft cutting spend or what have you.

32:26Dylan Patel: But ultimately, the revenue has only been a straight line up on the chart and almost every customer has only grown on revenue for them. and switching cost even when some other Chinese model comes out or, okay, it comes out with a better model, doesn't really shift the spend share that much for existing customers. You end up with this pretty sticky customer potentially who is growing revenue at some pace and there's no customer acquisition cost and the margins are high. So it's arguably a better business than software is an AI lab. If you can do it successfully. But how much is the risk, the fact that they're at this pacing the frontier um movement movement is coming out i think that's just signaling i think it's talk about that you don't think that's gonna that's true at all um i think pacing the frontier is a responsible thing to do if you believe that we could kill ourselves um i don't necessarily think that um they will pace the frontier too much though i think like there's there's like very little pacing they can do given the race that they're in um and people have made skits about this but like you know you know dario can't slow down because he doesn't trust sam and sam's not going to slow down and sam can't slow down much because same with dario obviously they're going to slow down a little bit when they have these massive security incidents um and neither of them can slow down because china is not too far behind so it's like sort of like very difficult that's reassuring all right um look we're going to take a quick break and then when we come back We're going to talk a little bit about the state of the AI clouds, the ClusterMax, of course, and then we'll continue on with some more questions about the technology and the business.

34:09So we'll do that when we're back right after this. This episode is brought to you by AppPoint. Everyone's racing to roll out AI right now. Copilots, chatbots, agents doing real work. But here's the part nobody loves talking about. All that AI runs on your data. And most teams have no single way to see it, secure it, and prove it's under control. That's exactly what AvePoint does. For 25 years, they've been the trusted layer beneath the world's most demanding data, now extended across your entire AI estate. Your data, your cloud, and the agents acting on your behalf. It's how more than 28 ,000 organizations deploy AI with confidence.

34:50So innovation scales without scaling risk. Avpoint, the unifying trust layer for AI. Head to avpoint.com to see how enterprises deploy AI with confidence. Learn more at avpt.co slash bigtechnologypodcast. That's avpt.co slash bigtechnologypodcast. You know those pieces in your closet that you find yourself wearing on repeat? The sweater that goes with everything, the perfect pair of pants, or the tee you somehow always reach for first? Those are the pieces that make getting dressed easy, and that's what Quince does best. Quince focuses on high-quality wardrobe staples made with premium materials like 100 % Mongolian cashmere, organic cotton, and merino wool.

35:36The styles are timeless, versatile, and designed to become the dependable pieces you reach for day after day. I am kind of obsessed with my black zip-down sweatshirt from Quince. It's great for planes and malls, it's easy to tie around my waist whenever it's warm out and then when the breeze comes in, back on it goes. Find your next fall favorites at Quince. Download the Quince app for app-exclusive offers or go to quince.com slash big tech. Get free shipping on your order and 365 day returns. Now available in Canada and the UK too. That's quince.com slash big tech. Violent threats against executives are growing at an alarming rate.

36:16When a company experiences backlash, executives are the first people blamed. When their address and other personal details are sitting online, this backlash can lead right to their front door. The team at Ironwell knows this better than anyone. They've protected some of the most targeted executives and individuals on the planet for almost two decades. As a white glove enterprise security service, they protect your people with continuous personal data removal, proactive prevention tools, and human-led emergency support. So when someone goes looking for your executives, Ironwell ensures they hit a dead end.

36:47So here's what to do. Go to ironwall.com slash big technology, fill in the quick form and request your free risk assessment. Their team will show you just how exposed your executives are and how to lock it down before a threat reaches their front door. That's ironwall.com slash big technology. Learn how to stop online threats before they become real world attacks. The links in the show notes.

37:12And we're back here on Big Technology Podcast with Dylan Patel and Jordan Anos from Semi Analysis. Semi-Analysis, I'm sure you're familiar. It's the Bible for anything AI infrastructure. These guys are the best. Very grateful to have your time, both of you. And, you know, definitely recommend you at least check out the blog. You can also get Semi-Analysis on Subsec. It's newsletter.semianalysis.com. It's the number one technology newsletter and well-deserved. So congratulations, guys. In the newsletter, you can find the ClusterMax 3.0. It's actually, even if you're not technical, I was talking with Jordan about this, it's a fun read.

37:54It goes into the state of the build out and the subhead tells you, you know, it already kicks off in the right way in the gory detail, the reliability, performance, support and pricing and security of the GPU cloud providers. There's a lot in there. There's a lot of areas we can go in this, Jordan and Dylan, but let's begin with the role of NVIDIA here. The report, you know, though it spends most of the time on the technical side of things, does go into the fact that NVIDIA is playing a major role in keeping these neo clouds. So the company's providing infrastructure, AI infrastructure, maybe to some of the frontier labs, but really across the board and making this whole thing work.

38:35And NVIDIA has been backstopping a lot of their businesses with, as you write, revenue floors, landlord guarantees, and leases that it plans to transfer to third parties, helping the Neo Clubs and AI Labs obtain investment-grade financing without a hyperscaler being involved. Jordan, to you on this one, you know, there's the criticism, sort of like the Edzatron criticism of this industry is that NVIDIA is basically funding the demand for all these neoclouds. And without that backstop, you know, this whole thing would basically evaporate and make NVIDIA look a bit smaller and make these neoclouds, you know, non-viable.

39:24What's your thought on that after going so deep into this? Yeah, well, I mean, both in ClusterMax and in another article we did recently called NVIDIA's Backstop Universe, we go into a little bit of the dynamics of what backstops are and how they help the NeoClouds raise the money they need to deploy the GPUs, which then customers are renting pretty strongly at this point. And the number that we have is that currently we've got tracking of$588 billion of off-balance sheet backstops that NVIDIA is supporting right now, which is one of yet another unbelievable number in the list that we've been going through on this podcast so far.

40:00I mean, but NVIDIA's revenue is approaching 400, about to cross$500 billion. So we're talking about really big numbers there. For context, that's like 2.5 % of the M2 money supply in the US. So obviously, there's massive, massive amounts of money going on in this universe, as we refer to it. And I mean, the only thing I can say about this is that it just seems really pragmatic from NVIDIA. They really want a wide diversity of customers. They don't want to be dependent on the hyperscalers for all of their revenue. They also don't want their customers to be dependent on getting their chips from a select set of three or four different hyperscalers.

40:41And so they have every incentive to have a nice, healthy market with a diversity of options and lots of selection for all of the customers to go work with. And in order to deliver on that, people have more problems than just raising the money required to buy the GPUs. Well, first of all, they need the guarantee that if they raise the money, those creditors are going to see returns. And that requires some sort of investment grade offtaker, which NVIDIA can serve in these cases, in many of these cases. But the way it's worked so far is that NVIDIA signs up as that offtaker. And then when it comes time to actually take those GPUs for themselves for their own internal research work, somebody else comes along and wants them more at a higher price.

41:28And so then they sell them to those guys. They don't take them for their internal research and they roll on to the next one. And for everybody who says that NVIDIA is doing this for some nefarious purpose or to keep some bubble going, I'd encourage you to look at the quality of NVIDIA's research. They publish more models on Hugging Face than anybody else. They have incredible CI across the suite of software that they support on their GPUs. And they trade models like Nemotron. We're not the biggest fan of these models or the software. We think there should be a diverse ecosystem. But the idea that NVIDIA does not use their own GPUs for research that serves the purposes of their company is itself wrong.

42:06And then the dynamic that we've seen is that this amount of compute is actually not a meaningful portion of NVIDIA's revenue. It's a very small percentage of NVIDIA's total revenue that is going back for their own internal research workloads. So what we're seeing is just that they are supporting NeoCloud's buildup the capacity for the demand from the customers, which is so incredibly strong right now and hard to serve. And then maybe the last thing to say about this is that it's not just the chips. We track a whole wide range of things that they're backstopping. This is everything from the construction costs, the memory suppliers, the people building fabs, the people building industrial equipment.

42:45I mean, it's not just like I need help raising money for my GPUs. It's I need help raising money for all of the other things in the supply chain, which is just the long pole in the tent for that specific project. Yep. Let's talk about security because, you know, the first image in this report is a quote tweet of an Ilya Syskever post, which was made right after the Hugging Face incident. Ilya says, Neo clouds have limited cybersecurity. Next time agents successfully go rogue, they'll try taking over a Neo cloud to run more copies. This is bad. Thus, Neo cloud should greatly strengthen their cybersecurity and every company with strong cyber models should help with that.

43:27This is a rare tweet from Ilya. Dylan, what do you make of this? Is this sounds to me like if he's out saying this, right? Remember, he's building this stuff himself. I'd say super intelligence that we might have a problem here.

43:39Dylan Patel: Yeah, so we posted an article titled Most NeoClouds Suck at Security. By the way, the art for that is incredible. We'll put it on the screen for those who are watching. It's like this guy getting stabbed by everybody and it's like it's SOC2 compliant. But yeah, so we posted that report and then that, you know, like maybe eight or eight hours later or whatever, where Ilya posted his tweet because they are very much using various clouds. They're obviously super secretive and care a lot about the secrecy of what they're working on, but also they are big believers in superintelligence. I guess safe superintelligence, SSI, they were the first.

44:29Dylan Patel: Trump actually copied Ilya. Ilya's first with another one, right? First to reasoning, first to LLM scaling, and now first to super intelligence branding. But Ilya sort of tweeted that out and it sort of echoed what we wrote, right? Which is like, look, most of these clouds are terrible at security. You can get open source models to just hack NeoClouds or get access to stuff within NeoClouds. You know, GLM 5.3 especially is like really good at this. So open source Chinese models can just hack dozens of the Neo clouds that exist out there and you can go look at what other people are doing. And so we've seen stuff like this.

45:08Dylan Patel: At other clouds, there was a cloud that you know, we didn't even intend to but we just accidentally ended up like seeing all of the other clients on that cluster seeing what stuff they had in storage, see what stuff they had in slurm, etc. And there were like, intelligence agencies, right? Like, oh, so we immediately We closed out, like reported it to them, were like, yo. Which intelligence agencies? We cannot say. I don't think that's a valid thing to say. National, like federal, well, security agencies and governments. They were not American. They were not American, but we said it was for a country that has a top 10 GDP.

45:44It's not like some tiny country. I'm losing my mind here that that was possible.

45:50Dylan Patel: Yeah, and it's like if we were malicious, and by the way, there were probably a dozen other customers, and maybe one of them was malicious, we could have just seen what they were training. And they were training models and running models for themselves, presumably, because this was a GPU infrastructure. So what else would you do with a bunch of Blackwell GPUs? Yeah, we throw around this word sovereign AI and it gets entangled with the concept of a neocloud many times. And people, even when they're raising money, they use this word a lot because they view compute as infrastructure. And I think that's exactly right.

46:23This is just like the telecommunications build out where this infrastructure is viewed by the sovereign nations that are deploying it and are involved in this as an asset that the country would like to control and have sovereignty over. And they believe that, similar to what you're saying at the beginning of the show, they want to be a net token importer or exporter. And that is driven by the infrastructure they build, which means data centers with GPUs inside of it. And so I think it's actually no surprise that federal government research and intelligence agencies are on some of these clusters sharing infrastructure with others.

47:00You would like them to have an air gap data center. But if there's exactly one facility in the entire country with GB300, NVL72, I mean, it's a seller's market. Goodness. All right. Well, there goes my sleep tonight. So, you know, you do in this report also rank, you know, a lot of these hyperscalers into tiers and the neoclouds into tiers. At the top tier is CoreWeave and Nebius. I'm more interested in the second tier, and that is Oracle and Google Cloud. Mostly, Jordan, because, you know, I have like in the document that I put together for, you know, our conversation today, I have like a lot of like topics and sentences that I've like pulled out of the report with questions.

47:49And then with Oracle, I just have Oracle and then a bunch of question marks, because that's kind of how I feel about that company. I don't know how to feel about them. Obviously, their debt has been rated, you know, not well by the rating agencies. They're making this all in bet. They rate pretty high. And they actually, in your report, have been credited with pioneering some techniques. But I would love to get your readout on Oracle and whether this has been a smart bet for them and whether they're going to rise in glory or sort of fall in flames like a lot of people anticipate.

48:24Dylan Patel: I think the key thing about ClusterMax is it is not stock recommendations, right? It is not anything to do with, hey, this Neocloud is going to build a gigawatt next year or 200 megawatts, which is far more relevant for the stock than how good are they at operating a cluster. Has nothing to do with, hey, this data center that they said they're going to deliver with this much capacity next year got delayed by three months. That's not what ClusterMax measures. ClusterMax is explicitly measuring sort of how good is the cluster for someone to use if you rent the GPUs? How reliable is it? How secure is it?

49:01Dylan Patel: How's the storage? How's the network? How's the lifecycle? How's the slurm on Kubernetes? How are all these things? In which case, you know, it's just sort of, we try to be more objective. Obviously making a rating pyramid is like, you know, it is something you kind of have to do. It would be nice if you could just make a long report and no one cared if there were no rankings. I mean, that's what would happen. So you kind of have to do the rankings and set certain boundaries. And you can always litigate the hell out of why and what is in what tier. but like it is an objective fact that we think Oracle is one of the top NeoClouds in terms of quality of service.

49:36Dylan Patel: Now that doesn't mean on the other hand, hey, at SemiAnalysis in April, we published to our clients that their New Mexico data center was going to be delayed because they weren't going to get the pipelines. And this was a big issue for OpenAI because OpenAI has contracted a huge data center with them in New Mexico. But if these pipelines don't get built, this data center can't go up. And then later, you know, we said it again in May, in June, and then most recently, Oracle announced a force measure on it and their stock tanked because of it. Right. So, so we're disentangling two things, right?

50:09Dylan Patel: Yes. Oracle is an amazing cloud at managed for their managed cluster, one of the top Neo clouds, but they also at the same time may have execution issues, whether it's regulatory in the case of this New Mexico data center where they can't get the pipeline there or other issues, you know, on their balance sheet, right? And so that is disentangling, like, what is good or bad for the stock versus what are we trying to test here? The goal in first ClusterMax was create something that people renting GPUs can use. And it turns out now, like, there's a lot of other uses for it that people try to ham fist in, whether it's like, you know, hey, we're going to write debt for this NeoCloud.

50:44Dylan Patel: If they upgrade rankings from bronze to silver, their rate will actually go down. things like this or like hey I'm going to invest in stocks that are at the top of ClusterMax so obviously these are not the decision these are not what ClusterMax is designed for it's a data point that you should use if you're doing anything with these companies but as far as Oracle as a company there's sort of like two things to go here right like one is they signed a ton of AI data center deals with OpenAI was it last year? Yeah, it was last year. Holy crap. It was only last year. It feels like eons ago. I think it might've been exactly a year ago.

51:26Dylan Patel: But the interesting thing is they signed these massive deals and there's a number of risks embedded with them. One is they signed them with a fixed margin essentially. So now that the price of GPUs has gone up a lot for clouds, they can't take advantage of this higher margin. So they're kind of hemmed in there. Number two, Two, they raised a bunch of debt and then the market started getting freaked out about how much debt companies can have. And so they spooked out, even though Oracle is completely solvent. And then number three, they have some execution issues. You know, I mentioned the regulatory one.

51:56Dylan Patel: There's some other data centers that have some delays and such there as well. But the main one being this, you know, such as in Wisconsin. But the main one being this New Mexico site that is delayed potentially multiple years because of this pipeline issue. And so, you know, these things can have very negative impacts on the stock. But I think the company is still fine. Now, you know, they're not Nebius. Nebius gets to turn around and they have the shortest average contract length of the top clouds in ClusterMax. And so Nebius has been able to turn around and really jack up pricing and take advantage of this.

52:28Dylan Patel: Right. Whereas CoreWeave and Oracle, despite having also good cluster management and reliability and security and all this stuff, were not able to take as much advantage because they locked themselves into long term contracts primarily. Oh, that is fascinating. All right, Jordan, let me ask you a technology question then. You know, as Dylan put out there, it is a pyramid. And there's like these top tier groups, like we talked about, CoreWeave and Nebius, Oracle, Google Cloud is up there. But in the not recommended section, there's a lot of logos. There's underperforming and unavailable. I think there's more something like more logos in the unavailable than you know the top four tiers combined what do you think that says about the fact that I mean what do you think that says about this technology that there's so many that are out there that just aren't meeting the bar for you look as people who are trying to represent the interests of buyers I think it just makes is sad.

53:31It's sad to see that a year and a half, almost since the first version of ClusterMax was published, actually approaching two years now, and since we coined the term NeoCloud at SemiAnalysis, that a lot of people have just not made progress towards the things that we were writing about. And the things we were writing about is just exactly what we talk to the customers of these neoclouds about what they want. And so I think it's just like the dynamic of it being a seller's market means that the priority for these companies is how can they raise as much money as they can, how can they build data centers as fast as they can, and how can they deploy chips as fast as they can, rather than working on the true quality of the underlying service.

54:16And that's a little bit of the dynamic that you see represented there, where people can have these great businesses without actually worrying too much about delivering a great service that their customers really enjoy and recommend to other people. The other thing is that I think there's like this constant dynamic that we'll continue to see over time, which is that there can only be so many leaders because there's only so many people who truly innovate and then everybody else copies them. And when you have a market that by our tracking is actually 323 providers as of this morning, which we track in our AI Cloud TCO model for people who are subscribers to Semi-Analysis, they have this full list.

54:52There's only a certain amount of them that are even relevant, like large enough to put on the list and then a certain number of those that can pass the bar for what we believe is what we need to see to recommend them. And, you know, in some ways, we think that the bar for being recommended is quite low. Like the basic things that fail with respect to security and reliability being the two top ones are just, in our view, dead simple. and it's just sad to see that others aren't paying attention enough or caring enough to actually implement this stuff. Okay, two quick ones before we go. Just a more broad question.

55:32You know, Google, obviously, they've announced they have this Gemini Argon coming out, Gemini 4 Argon. Is it your view that, you know, there have been some people who have called upon Google to just be, you know, basically a NeoCloud and rent out the TPUs from Google Cloud and the computing. Do you see them going that way? Or do you think that they're still going to use like a large percentage of their compute for model training and inference?

55:58Dylan Patel: I mean, definitionally, they have allocated less compute to model training and deployment in the last six months than they had two or three years ago. And that percentage allocation has fallen over time. This is sort of, you know, one of the things that we made a joke about, like, maybe not even joke, but just a statement that Thomas Kurian won and Dennis lost, right? Now, I guess Karai. Google is producing significantly more compute, you know, year on year. But the amount that's going to DeepMind is not rising as fast. It's not like, oh, DeepMind had a third of Google's compute. Now it still has a third of Google's compute.

56:39Dylan Patel: It's actually falling as a percentage, which is really interesting because now if I look at just DeepMind versus Anthropic and OpenAI, DeepMind has the least compute of the three by the end of this year. So it's really a big deal. And so Google, whether or not there's a NeoCloud, what the word NeoCloud even means, does selling tokens of other models count as NeoCloud? In a sense, that's what the new cloud business is, is selling AI model tokens, right? So maybe that's a NeoCloud, right? The definition is quite - Is money from Blackstone and going and doing new construction projects to build your GPUs to deploy directly for customers outside of your core platform?

57:21Dylan Patel: Or selling chips, right? Is that NeoCloud? No, I mean, they're selling a lot of chips to Anthropic. So it's physically selling the chips rather than just renting them out. So I would say like Google's business model is evolving and they want to make money at every layer of the stack. But by definition, helping Anthropic this much with TPUs and helping them this much with these other things is hurting Google DeepMind's business. They need to find a niche, man. Like for Gemini, I mean, nobody wants the third or fourth or fifth best coding model. So they need to find a niche that drives Gemini token sales if they want DeepMind to continue competing on the frontier of research.

58:01because right now it's, I mean, even with this Gemini 4 Argon release, it's just not there on coding compared to the top two or three.

58:08Dylan Patel: And the risk here is what happens next year when OpenIon and Anthropic both attempt to add five to 10 gigawatts of additional compute for their labs and, you know, for OpenIon and Anthropic each. And then Google only adds a few gigawatts. Now their compute deficit, it was Google's way ahead in compute and they were kind of behind slash equivalent, mostly just a little bit behind. And now they've fallen further behind, but their compute is the same. So what happens when they have less compute? Are they, do they somehow magically catch up? It's possible, right? They have the talent, they have the people, they need to reorganize and really focus up.

58:46Dylan Patel: But also like more and more, the best people just keep getting poached to Anthropic, to OpenAI, and it's meta. Discovery loop. Discovery loop, yeah. Discovery loop, right. Okay, I want to ask one more before we go, if that's okay. You know, I think, Dylan, you've said in the past that the people who are getting most of the value out of the AI models that are created are the end users, right? The people that are using the clods, for instance, at the base. And of course, Anthropic makes that model available via API, and to go into all these various uses, you know, it's going to have to do more licensing, and who knows if they'll be able to do products for all this.

59:23I'm curious if you see a scenario where given the big numbers that we talked about, about in the past, whether they keep their frontier models at home for longer, much longer, and use them to build their own products. Like we know they have the wet lab, we know they have Cloud Code and Cowork, and then they don't make them available because they want to get more margin out of the infrastructure that they've bought and built.

59:47Dylan Patel: Yeah. So, I mean, I think we've already seen this, right? Anthropic had Mythos ready in February. Their model safety cards noted that they have two other models since Mythos that have released. And publicly, we still don't have full Mythos. We have a neutered version of it called Fable. And it took many, many months before Fable even came out. They had the Glasswing, which was like Mythos, but for only some subsectors. and it was neutered in some areas, right? Not all areas. And so it's clear that Anthropic has not released new models. OpenAI likewise, Astra was actually done a couple months ago and it only came out like, you know, it took a couple months before they released Astra.

1:00:28Dylan Patel: And their new model Bell, they even showed a blog where they showed the performance of it in math and how it, you know, solved all this stuff and they have not released it. So clearly they are just using it internally, not externally. And so, you know, the labs have already moved along this line, right, where it's like, oh, yeah, the gap between them and Chinese open source is still six months. They're kind of keeping that. But in reality, they kept, you know, months of progress internal only because the moment you make it external, people can distill it, you know, and all these other things. So I think, you know, Anthropic will likely release a new model in the coming weeks, you know, for their IPO.

1:01:10Dylan Patel: I think OpenAI will likely have to release a new model in the coming weeks as well, just because it's such a competitive environment. But at the same time, they're keeping a lot of their stuff internal only. Yeah, we were discussing internally some of the capitalistic incentives. If the incentives are actually going to motivate people to release more models and get them in the hands of consumers because of the incredible gross margins on the per-token API pricing, or if the incentive is to keep it internally so that you can pursue the end user applications, which are even more high value than selling the access of the model per token.

1:01:47It's going to be interesting to see how it plays out. But in one of those scenarios, the capitalist approach will be to pace the frontier for yourself, right? And what's your gut on where it's going to go? Which one of those scenarios you think plays out? I think Anthropic has publicly stated that they are pacing the frontier by keeping themselves restricted. And that's the only way that their blog posts have manifest in reality. Nobody else is slowing down. They're only slowing themselves down. I think that's true. I think that's what's happening. And I think the incentives to both improve the status of AI by solving things that are, you know, like cancer that are just unequivocally good, that everybody can agree is a good thing would be a positive incentive.

1:02:37And then I think there's also plenty of money to be made if you do something like that. So my bet is that we continue to see the minimum viable models released by both OpenAI and Anthropic to keep a lead over Chinese and the rest of the open source ecosystem. The website is semianalysis.com. Sign up for the blog, as Dylan says. Sign up for the newsletter, as I have. it's a wealth of insight and analysis and I think you'll love it if you're not there already and this conversation has just you know been the manifestation of that so Dylan great to see you again Jordan really nice to meet you you guys are welcome anytime and we appreciate your time today thanks for having us Alex all right good stuff thanks everybody for listening and watching and we'll see you next time on big technology podcast

1:03:29You

1:03:34You

From the publisher

Dylan Patel is founder and CEO of SemiAnalysis, Jordan Nanos is a member of its technical staff. Patel and Nanos join Big Technology to discuss whether AI’s massive infrastructure buildout can justify the trillions of dollars being spent on compute. Tune in to hear how big the boom could get, where the revenue needed to support it will come from, and what it could mean for knowledge work and the broader economy. We also cover Nvidia’s role in financing AI infrastructure, the economics and security of neoclouds, Anthropic’s potential IPO, Oracle’s AI bet, and whether frontier labs are keeping their best models to themselves. Hit play for a deep look at the financial and technical machinery powering the AI boom, and what could happen if it keeps accelerating.

---

Enjoying Big Technology Podcast? Please rate us five stars ⭐⭐⭐⭐⭐ in your podcast app of choice.

Want a discount for Big Technology on Substack + Discord? Here’s 25% off for the first year: https://www.bigtechnology.com/subscribe?coupon=0843016b
Learn more about your ad choices. Visit megaphone.fm/adchoices

More from Big Technology Podcast

All episodes