# Dylan Patel Explains the AI War While Cooking | In-Context Cooking

Latent Space · 2026-02-26

<https://addtry.com/bc70148b-e73e-40d6-8596-881430e9d86a>

Dylan Patel, CEO of SemiAnalysis, argues hyperscalers like Google, Amazon, and Meta will sacrifice all profits to build AI infrastructure, spending $180–$200 billion in capex this year alone, because the AI adoption explosion—Claude Code driving 4% of GitHub commits in one month, Anthropic adding $2.5 billion monthly revenue—makes it a Pascal's wager: spend or die. He details how Taiwan's semiconductor geopolitics create endgame scenarios, from a KMT win placating China to full invasion, with TSMC's output critical. Patel explains Nvidia's paranoid founder Jensen Huang is responding to vertical integration threats from hyperscalers by diversifying into chips like CPX and Groq, but warns moats are shallow. The real bottleneck in AI progress? Semiconductors themselves: fabs take years to build, and no one can buy enough GPUs through 2028. He also predicts a massive AI backlash from the public and financial markets, as capital consumption outpaces revenue and labor displacement accelerates.

## Questions this episode answers

### Why are hyperscalers like Google and Amazon spending hundreds of billions on AI infrastructure amid falling stock prices?

According to Dylan Patel, Google announced $180 billion and Amazon $200 billion in 2025 CapEx, yet their stocks dropped — Amazon’s by about 10%. The market hates the spending, but hyperscalers are driven by exploding AI demand: Anthropic alone added $2 billion in revenue in a single month. Companies see a ‘light at the end of the tunnel’ and face an innovator’s dilemma — if they don’t invest, they risk being disrupted.

[24:33](https://addtry.com/bc70148b-e73e-40d6-8596-881430e9d86a?t=1473000)

### What is the current primary bottleneck limiting AI progress, and why has it shifted back to semiconductors?

Dylan Patel states the bottleneck has cycled back to semiconductors because there aren’t enough fabs. While data centers and energy were recent constraints, chip fabrication has multi‑year timelines, involves the world’s most complex buildings, and uses tools costing hundreds of millions of dollars. The industry hasn’t built enough capacity, so the semiconductor shortage will persist through the end of the decade or until AI slows down.

[49:01](https://addtry.com/bc70148b-e73e-40d6-8596-881430e9d86a?t=2941000)

### What is Dylan Patel’s ‘galaxy brain’ scenario for US-Taiwan-China relations that benefits the US?

Patel’s ‘galaxy brain’ outcome is for Taiwan’s pro‑China KMT party to win elections. This would placate Beijing, lowering the risk of invasion, while TSMC would still have to enforce US export controls because it depends on American banking and equipment. The US would retain access to advanced chips, and China would gain political satisfaction without actually obtaining the semiconductor technology.

[9:24](https://addtry.com/bc70148b-e73e-40d6-8596-881430e9d86a?t=564000)

## Key moments

- **[0:00] Intro**
- **[1:46] Origins**
  - [3:06] Dylan Patel launched SemiAnalysis after Doug O'Neill urged him to go non-anonymous and charge for his semiconductor analysis.
- **[6:56] Taiwan Endgames**
  - [6:56] The most likely Taiwan endgame is a political coup that destabilizes without full invasion, letting China creep control.
- **[10:57] Export Controls**
  - [10:58] At least one-third of AI researchers in American labs are Chinese, making export controls a delicate talent issue.
  - [16:02] Google will spend $180 billion and Amazon $200 billion on AI infrastructure in 2025, quadrupling from a few years ago.
- **[18:19] AI Backlash**
  - [19:02] 'The biggest risk is actually just like the general public hates AI.' - Dylan Patel
  - [21:12] Claude Code doubled its share of GitHub commits from 2% to 4% in one month; AI-generated code may now be 10% or more.
- **[21:25] Adoption Surge**
- **[23:40] CapEx Backlash**
  - [25:30] Anthropic added $2.5 billion in monthly revenue with 40% gross margin, requiring roughly $1.5 billion in additional compute.
  - [27:14] Oracle's stock peaked within a week after announcing $300 billion-plus in AI infrastructure deals with OpenAI.
- **[30:26] Hyperscaler Dilemma**
  - [30:26] Hyperscalers face an innovator's dilemma: invest massively in AI or risk total disruption to their core businesses.
  - [31:21] Google will have zero cash flow profit in 2027, spending every dollar on AI infrastructure.
  - [33:12] Meta is taking on $40 billion in debt to fund its largest AI cluster in Louisiana, a sign that even profitable companies must leverage up.
  - [38:24] The anti-AI party will win the next US election because the public hates AI and fears job loss, predicts Dylan Patel.
- **[38:49] Chip War**
  - [40:53] Jensen Huang is one of the most paranoid founders, quickly pivoting Nvidia's strategy upon hearing of OpenAI's Cerebras deal.
  - [44:11] Nvidia must out-innovate vertically integrated rivals like Google and Amazon by a wide margin to justify its high margins.
- **[45:32] Bottleneck**
  - [45:32] The ultimate bottleneck for AI infrastructure is semiconductor fabrication capacity, not data centers or power, because fabs take years to build and have limited capacity.
- **[50:55] Tasting**
  - [52:01] French technique vs. slop: Dylan Patel argues that heavy seasoning and charred flavors beat delicate ingredient-focused cooking for fried rice.
  - [53:07] Q: Is SemiAnalysis hiring? A: Yes, SemiAnalysis is hiring globally for 60 employees, working with top AI labs and hyperscalers, including an AI usage tracking role.
  - [54:24] Dylan Patel's intensely seasoned fried rice beat the restaurant version in a blind tasting, winning over the panel with bold flavor.

## Speakers

- **Brandon Anderson** (host)
- **Dylan Patel** (guest)

## Topics

Hardware, Industry

## Mentioned

AMD (company), Amazon (company), Anthropic (company), Cerebras (company), CoreWeave (company), GitHub (company), Google (company), Groq (company), Intel (company), Meta (company), Microsoft (company), NVIDIA (company), OpenAI (company), Oracle (company), TSMC (company), ChatGPT (product), Claude Code (product), Codex (product), Devin (product), GitHub Copilot (product)

## Transcript

### Intro

**Dylan Patel** [0:00]
I'm not crying because of the, because of the re- AI researchers leaving. I'm crying 'cause of the onions, God, promise.

**Brandon Anderson** [0:05]
It's because he cut the onions.

**Dylan Patel** [0:05]
I swear to God, if Uncle Roger finds this video, I'm gonna cry. This 's about to be sweeter than Panda Express.

Wait, what do you vote? What do you vote?

**Brandon Anderson** [0:16]
Hey, guys. Welcome to In Context Cooking, where we take one dish, taste it, and try to recreate it with minimal help. Today, we have a very special guest, the founder and CEO of SemiAnalysis, Dylan Patel. Welcome, Dylan.

**Dylan Patel** [0:28]
Hello. Thanks for having me.

**Brandon Anderson** [0:29]
Yeah. Thank you for being here. I guess the question to start off is how would you rate yourself on a scale of one to 10, of one being awful-

**Dylan Patel** [0:36]
10. 10, 10.

**Brandon Anderson** [0:36]
10? Okay.

**Dylan Patel** [0:37]
10. Based on yesterday.

**Brandon Anderson** [0:37]
So you're an amazing cook. So you're amazing.

**Dylan Patel** [0:38]
Looks, looks, looks, looks.

**Brandon Anderson** [0:40]
As a cook, how would you rate yourself on a scale of one to 10?

**Dylan Patel** [0:43]
Uh, probably like a five or six.

**Brandon Anderson** [0:45]
Okay.

**Dylan Patel** [0:45]
Yeah

**Brandon Anderson** [0:45]
Five or six is not bad. I feel like you have-

**Dylan Patel** [0:47]
Maybe three. Maybe three. Let me, let me revise. Let me revise. Yeah, yeah.

**Brandon Anderson** [0:49]
So low expectations, and then over delivering. You're gonna over deliver. That's good.

**Dylan Patel** [0:52]
The presentation is great, just not the plate.

**Brandon Anderson** [0:54]
Okay. Okay. Honestly, like, we'll work with three or five, whatever. But yeah, I guess in front of us, we'll... We have a bunch of ingredients. Do you have an idea of what we're gonna make? I feel like it's kind of obvious just looking off of it.

**Dylan Patel** [1:05]
Mm-hmm. Eggs-

**Brandon Anderson** [1:05]
Um-

**Dylan Patel** [1:06]
Eggs, rice, chicken. This is, like-

**Brandon Anderson** [1:09]
Very, like-

**Dylan Patel** [1:09]
This is, like, very, like-

**Brandon Anderson** [1:10]
... typical-

**Dylan Patel** [1:10]
Yeah

**Brandon Anderson** [1:11]
... refrigerator stuff, leftovers

**Dylan Patel** [1:11]
It's anything, anything. But then the right side, you know, like, the peas really throw me off.

**Brandon Anderson** [1:15]
Yeah.

**Dylan Patel** [1:15]
The ginger, though. So I th- and soy sauce. Okay, so this is, like, fried rice.

**Brandon Anderson** [1:18]
Okay. Yeah.

**Dylan Patel** [1:19]
Yeah.

**Brandon Anderson** [1:19]
So today we have chicken fried rice, which we'll be recreating.

**Dylan Patel** [1:24]
Nice.

**Brandon Anderson** [1:24]
Um, this is restaurant chicken fried rice, so we'll try to recreate it as close as possible-

**Dylan Patel** [1:29]
Mm-hmm

**Brandon Anderson** [1:29]
... based off tasting it. So here's a spoon for you. Cheers.

**Dylan Patel** [1:31]
Cheers.

**Brandon Anderson** [1:34]
Mm. That was... This is very good.

**Dylan Patel** [1:36]
I'm shocked it's still warm.

**Brandon Anderson** [1:38]
Yeah.

**Dylan Patel** [1:38]
I assume this has been sitting here for hours.

**Brandon Anderson** [1:39]
Yeah. We nuked it a lot, and then kept it as is. So yeah, just to get started, do you wanna introduce yourself? I know that, um, you went to Minnesota briefly after, and was it you were a beekeeper for two years?

### Origins

**Brandon Anderson** [1:51]
You did a lot of these kinda, like, side quests.

**Dylan Patel** [1:53]
This is the most random bee stings.

**Brandon Anderson** [1:53]
You have, like, a lot of lore. But now you're, you know, the leading voice on, um, chips and everything that... You know, whether it be hedge funds or even people in AI.

**Dylan Patel** [2:01]
So I did live in Minnesota briefly after college. Um, I'm from rural Georgia. Um, yeah, we, uh... I, I did beekeep for, like, a, a year and a half, basically. I feel like I've just kinda sort of went through a lot of life, uh, just next step, next step, next step.

Doesn't seem like there was a clear immediate path.

**Brandon Anderson** [2:18]
Yeah.

**Dylan Patel** [2:18]
Uh, looking back, I can spin a narrative. Like, oh, obviously- ... I'm gonna be doing this-

**Brandon Anderson** [2:22]
Yeah

**Dylan Patel** [2:22]
... because my interest when I was eight was this, and my interest when I was 12 was this.

**Brandon Anderson** [2:26]
Mm-hmm.

**Dylan Patel** [2:26]
But, uh, you know, like, moderating forums related to chips.

**Brandon Anderson** [2:29]
Yeah.

**Dylan Patel** [2:29]
But, you know, I thought it was just, like, a serendipitous thing, you know?

**Brandon Anderson** [2:31]
Yeah.

**Dylan Patel** [2:31]
Um, but then eventually sort of everything culminated-

**Brandon Anderson** [2:35]
Mm-hmm

**Dylan Patel** [2:35]
... and, like, blogging, and doing consulting, and doing research, being interested in AI and data science, being interested in chips, and then it all sort of, like, culminated in, like, oh, my God, everything blew up all together. And, and so I guess right time, right moment.

**Brandon Anderson** [2:47]
Yeah.

**Dylan Patel** [2:47]
Um, maybe some-

**Brandon Anderson** [2:48]
Like the AI boom and-

**Dylan Patel** [2:48]
Maybe some foresight to have your passion be the thing that everyone cares about now.

**Brandon Anderson** [2:52]
Yeah. No, I think it turned out very great. You know, doing amazing. And rumor has it you started your Substack because you read Doug's and thought you could do a lot better, and he was like, "Hey, you should start a Substack."

Is that true, or is there more behind that story?

**Dylan Patel** [3:06]
Yeah, so what, what happened was I, um, I had an anonymous blog-

**Brandon Anderson** [3:10]
Mm-hmm

**Dylan Patel** [3:10]
... on the internet for many years.

**Brandon Anderson** [3:12]
Yeah.

**Dylan Patel** [3:12]
Uh, I was moderating Reddit and all these things anonymously-

**Brandon Anderson** [3:15]
Mm-hmm

**Dylan Patel** [3:15]
... for, a- around, like, hardware, Nvidia, Intel, AMD, this kinda stuff. And I was posting all this stuff, and I had an anonymous Twitter in the sil- in Silicon Twitter, right-

**Brandon Anderson** [3:23]
Yeah

**Dylan Patel** [3:23]
... which people on, like, Teapot and, like, tech Twitter don't understand.

**Brandon Anderson** [3:26]
Yeah.

**Dylan Patel** [3:26]
Um, and so I was doing this anonymously. Doug, Doug started posting, and I was like, "Wow, this is interesting, but, like, I think I could do a lot better." And then he's like, "Dude, why are you, like, why are you, like, posting on WordPress," right?

**Brandon Anderson** [3:38]
Mm-hmm.

**Dylan Patel** [3:38]
Like, "Do it, do it non-anonymously. Do it on Substack and start charging for it." So then, but then I was like, too, I was like... I was like, "I'm not gonna charge for this. I'm so great." And then, and then one day I was like, "You know what?

Screw this. I'm gonna start charging for it," 'cause Doug told me to many, many times.

**Brandon Anderson** [3:50]
Yeah.

**Dylan Patel** [3:51]
Um, and then a few years go by, and then Doug, Doug joins the company. Um, so it, it was really a great moment that, uh-

**Brandon Anderson** [3:56]
Yeah, full circle

**Dylan Patel** [3:57]
... he, he told me to, uh, do it because otherwise it might still just be, like, a, a niche anonymous blog where I'm still just doing random consulting rather than actually, uh, you know, a company with 60 people.

**Brandon Anderson** [4:06]
Gotcha, okay. So at that time you were still doing consulting, right, and you just had a separate blog that was more-

**Dylan Patel** [4:11]
Well, it was consulting related to the blog and-

**Brandon Anderson** [4:12]
Okay

**Dylan Patel** [4:12]
... the industry.

**Brandon Anderson** [4:13]
Gotcha. Okay. Yeah, that's amazing. I guess now we could kind of look at the ingredients and maybe try a little. So here we have ginger, garlic, carrots, cheese.

**Dylan Patel** [4:22]
We're gonna, we're just gonna eat the whole clove of garlic.

**Brandon Anderson** [4:24]
Or yeah, we could all take a piece. And then this I think is... And we also have tasting spoons if anyone tries. Um, I think this is sugar. Yeah.

**Dylan Patel** [4:35]
Mm.

**Brandon Anderson** [4:35]
Like baby ginger.

**Dylan Patel** [4:35]
So sweet.

**Brandon Anderson** [4:36]
Okay, let's see.

**Dylan Patel** [4:37]
I really hope this is the salt.

**Brandon Anderson** [4:38]
Okay, it's sugar. Yeah. So this is sugar. And we'll know 'cause everything will be laid out for you exactly like this, um, so we'll have to eyeball it, and we'll be going fast. This, I don't think we need to taste.

I'm pretty sure this is-

**Dylan Patel** [4:49]
MSG?

**Brandon Anderson** [4:49]
... baking powder.

**Dylan Patel** [4:50]
Oh.

**Brandon Anderson** [4:51]
Yeah, so I don't think we need to-

**Dylan Patel** [4:51]
What kind of, what kind of fried rice-

**Brandon Anderson** [4:53]
I'm just playing

**Dylan Patel** [4:53]
... has MSG?

**Brandon Anderson** [4:54]
Yeah.

**Dylan Patel** [4:54]
Oi.

**Brandon Anderson** [4:55]
Yeah, that's true.

**Dylan Patel** [4:55]
Let's go to the-

**Brandon Anderson** [4:56]
That's true. We're, we're kinda-

**Dylan Patel** [4:57]
... let's see the YouTuber

**Brandon Anderson** [4:57]
... Uncle Roger. Yes. But yeah, we have soy sauce here, baking powder, sugar, onion of course. I think this is corn starch.

**Dylan Patel** [5:03]
If, if he discovers this, he's gonna roast the fuck out of us.

**Brandon Anderson** [5:05]
Yeah, he will, but hopefully, you know, he'll give us some grace.

**Dylan Patel** [5:09]
Yes.

**Brandon Anderson** [5:09]
But okay, and then eggs, salt. Yeah.

**Dylan Patel** [5:12]
Wait, is that the tactic? Intentionally don't put MSG in your rice.

**Brandon Anderson** [5:15]
Yeah.

**Dylan Patel** [5:15]
To rage bait Uncle Roger.

**Brandon Anderson** [5:17]
Yeah, this is also short grain rice, which is also a red flag, but we just have it. And this is day old. So this is the important ingredient, to have it day old, not fresh.

**Dylan Patel** [5:25]
Why does that matter?

**Brandon Anderson** [5:26]
Um, it's a little drier, and it'll make the fried rice just, like, the grains be a little more separated compared to being a mushy mess. But yeah, I think we're good. Are you ready to show off your cooking skills?

**Dylan Patel** [5:36]
Sure.

**Brandon Anderson** [5:37]
Great. Let's get started.

Dylan, are you ready to make some fried rice?

**Dylan Patel** [5:43]
Absolutely.

**Brandon Anderson** [5:44]
Okay. So the first thing, let's grab the chicken, and then we just need to marinate it, so, or velvet it rather. So take the corn starch, it's one of the white powders that we saw before, and the baking powder.

And then add, like, a spoonful, maybe half a spoon of corn starch, um, to the chicken, and then just a hint of the baking powder

Um, and so yeah, once we add that, then-

**Dylan Patel** [6:11]
Like a pinch or what we talking?

**Brandon Anderson** [6:13]
I just did like, yeah, a pinch or like a very small amount with my spoon. And then we can add a little bit of salt and soy sauce. I think this is salt. Make sure it's salt and not sugar.

**Dylan Patel** [6:25]
Yeah.

**Brandon Anderson** [6:25]
Okay. There's a little bit of salt and then soy sauce. And so what we wanna do is we wanna mix it kind of vigorously so it just looks like-

**Dylan Patel** [6:34]
You said salt and soy sauce only?

**Brandon Anderson** [6:36]
Yeah. Um, just like a little bit of each to give it some flavor 'cause we're gonna marinate it. And yeah, we just wanna mix it vigorously so that it looks like a slop of chicken. But I guess one question on my mind that you've also discussed about is endgame scenarios with Arrakis, or in this case Taiwan, to end TSMC.

### Taiwan Endgames

**Brandon Anderson** [6:56]
I guess, like have you thought about what that would look like?

**Dylan Patel** [7:00]
Bro wants to make fried rice and talk- ... about China politics. Let's go.

**Brandon Anderson** [7:04]
Indeed.

**Dylan Patel** [7:05]
Um-

**Brandon Anderson** [7:05]
Very on topic.

**Host** [7:06]
That's vaguely racist.

**Brandon Anderson** [7:09]
Just- just vaguely. Just vaguely.

**Dylan Patel** [7:11]
Vaguely racist, yeah. Um, okay. So- so endgame scenarios are-

**Brandon Anderson** [7:14]
Yeah

**Dylan Patel** [7:14]
... are kind of insane, right?

**Brandon Anderson** [7:16]
Mm-hmm.

**Dylan Patel** [7:16]
Like, um, there- there's- there's a variety of things that one could, uh, that could happen.

**Brandon Anderson** [7:20]
Yeah.

**Dylan Patel** [7:21]
Um-

**Brandon Anderson** [7:21]
Such as?

**Dylan Patel** [7:23]
So- so i- i- in some cases, right, like it's like, you know, status quo is the best, right?

**Brandon Anderson** [7:28]
Mm-hmm.

**Dylan Patel** [7:28]
Um, you know, n- no war happens. I'm done with the, uh, chicken by the way.

**Brandon Anderson** [7:32]
Okay, great.

**Dylan Patel** [7:32]
No war happens.

**Brandon Anderson** [7:33]
And just start cutting the onion to like small dices, but continue.

**Dylan Patel** [7:37]
We have no war, no invasion, there's no blockade.

**Brandon Anderson** [7:39]
Yeah.

**Dylan Patel** [7:39]
Sort of status quo. China continues to, China continues to, you know, industrialize itself-

**Brandon Anderson** [7:45]
Mm

**Dylan Patel** [7:45]
... with the industry that Taiwan has, but there's no, there's no major event that occurs.

**Brandon Anderson** [7:49]
Mm-hmm.

**Dylan Patel** [7:50]
You know, s- that's sort of option one. Option two, which is what a lot of people seem to like think is the best, which is Ch- Taiwan actually like more stringently claims it's independent.

**Brandon Anderson** [8:00]
Yeah.

**Dylan Patel** [8:00]
There, at least a lot of like Westerners, and- and then that seems like the w- like actually a poor option, um, just because that- that, uh, potentially causes China to, uh, move- move much more, um, aggressively on Taiwan.

**Brandon Anderson** [8:12]
Mm.

**Dylan Patel** [8:12]
And sort of like there's like another set of options, which is like, okay, well, what if, what if China moves clo- or Taiwan moves closer to China, um, in a political way, right? And so like that's an example of that is like, hey, like they elect the KMT.

**Brandon Anderson** [8:24]
Mm-hmm.

**Dylan Patel** [8:24]
Um, and the KMT ends up, uh, winning. So there's two parties in Taiwan, the DPP, KMT.

**Brandon Anderson** [8:30]
Yeah.

**Dylan Patel** [8:31]
Um, if the D- D- DPP wins, they're sort, they cont- they have won for the last decade, and they're sort of more pro-US-

**Brandon Anderson** [8:37]
Mm

**Dylan Patel** [8:37]
... more anti-China, more ch- more Taiwan independence.

**Brandon Anderson** [8:40]
Okay.

**Dylan Patel** [8:40]
Um, and then there's sort of like the last scenario, which is just a full on invasion.

**Brandon Anderson** [8:44]
Yeah. That makes-

**Dylan Patel** [8:44]
Um, or- or at least a- a- a political coup or takeover of some sort. And so there's like a variety of options possible. Um, what I think is most likely, um, is that there is some sort of political coup or action that-

**Brandon Anderson** [9:00]
Mm

**Dylan Patel** [9:01]
... um, destabilizes Taiwan in some way.

**Brandon Anderson** [9:03]
Mm-hmm.

**Dylan Patel** [9:03]
Um, but doesn't act- doesn't necessarily lead to an invas- a- a full scale, um, invasion. And so sort of this is the best of the both worlds for at least China, right? Like they don't have to actually enter an invasion.

**Brandon Anderson** [9:15]
Yeah.

**Dylan Patel** [9:15]
Uh, but they get to continually creep more and more on Taiwan without actually having to deal with the repercussions of a war and subsequent blockades and so on and so forth.

**Brandon Anderson** [9:24]
Mm.

**Dylan Patel** [9:24]
The like kind of galaxy brain thing for an American to want is for actually the Taiwanese, um, government, uh, the pro-US party to lose, right?

**Brandon Anderson** [9:33]
Oh, interesting.

**Dylan Patel** [9:34]
Um, so you want KMT to win.

**Brandon Anderson** [9:36]
Yeah.

**Dylan Patel** [9:37]
Um, and- and if KMT wins, then that means China will be placated more.

**Brandon Anderson** [9:40]
Yeah.

**Dylan Patel** [9:40]
You don't have T- Taiwan fully move into China's orbit.

**Brandon Anderson** [9:44]
Mm-hmm.

**Dylan Patel** [9:45]
Um, but you do have the government sort of placated in China.

**Brandon Anderson** [9:49]
Yeah.

**Dylan Patel** [9:49]
Um, at the same time, uh, even if, even if the KMT wins, it's not like TSMC starts disobeying American export restrictions because-

**Brandon Anderson** [9:55]
Mm

**Dylan Patel** [9:55]
... the way American export restrictions are upheld is that Taiwan utilizes American banking systems, American equipment industries.

**Brandon Anderson** [10:02]
Yeah.

**Dylan Patel** [10:02]
And so they'll still have to uphold any US export restrictions. So it's sort of like, you know, China's placated by the fact that they have a friendly government in power, and yet China doesn't actually have any of the chips, and the US continues to get to access the chips.

**Brandon Anderson** [10:14]
I guess for export controls as well, so I know that right now there's like stances. For example, Dario, right, is very anti, uh, China having access to things. Um, and I think supposedly one of the top researchers, Xun Yu, also left because of that.

There could be maybe a lot of-

**Dylan Patel** [10:31]
He left, he left, uh, Anthropic?

**Brandon Anderson** [10:32]
He left one of the top labs.

**Dylan Patel** [10:34]
Got it. Got it.

**Brandon Anderson** [10:34]
I guess, do you have opinions on like if the US takes-

**Dylan Patel** [10:37]
How much onion should I c- ch- dice by the way?

**Brandon Anderson** [10:38]
You can just cut or dice like half.

**Dylan Patel** [10:40]
Okay, okay.

**Brandon Anderson** [10:41]
And then after you do that, just start cutting the carrots to like a similar size. Do you have any thoughts on, you know, the danger of maybe a lot of Chinese talent fleeing due to the very staunchly anti-Chinese stance that a lot of these AI labs or top figures may take, or do you think this is a, like a non-issue?

### Export Controls

**Dylan Patel** [10:58]
You know, um, I'm n- I'm not crying because of the, because of the re- AI researchers leaving. I'm crying- ... 'cause of the onion, I promise.

**Brandon Anderson** [11:04]
You just cut the onion.

**Dylan Patel** [11:04]
No, but it would be, it'd be a travesty, right?

**Brandon Anderson** [11:06]
Yeah.

**Dylan Patel** [11:06]
If like a lot of like Chinese researchers left-

**Brandon Anderson** [11:08]
Yeah

**Dylan Patel** [11:08]
... um, American labs. You know what, I think it's probably like- like half or-

**Brandon Anderson** [11:12]
Mm

**Dylan Patel** [11:13]
... at least a third of researchers at labs are Chinese.

**Brandon Anderson** [11:16]
Mm.

**Dylan Patel** [11:17]
Um, so obviously there's like a level of like, you know, why- why are you anta- don't antagonize too much.

**Brandon Anderson** [11:22]
Yeah.

**Dylan Patel** [11:22]
Um, at the same time, there is a level of like, you know, this is, this is the greatest, uh, technology to ever, uh, fall into humanity's hands.

**Brandon Anderson** [11:32]
Mm-hmm.

**Dylan Patel** [11:33]
Um, you know, obviously we think we're the good guys, and by we, I mean Americans. Um, and so Americans think that they should control it, and maybe- maybe Anthropic thinks they're the good guys, not Americans as a whole.

Um, but whatever the, whatever the moral justification is, like, you know, whether it's, "Hey, we're the good guys and AI's gonna be super powerful," or, "And we're the only ones who can steward it," or, "Hey, AI would be a great weapon, so let's make sure we're the only one with that weapon."

Um- There's certainly some level of control that needs to be st- had.

**Brandon Anderson** [11:59]
Mm-hmm.

**Dylan Patel** [11:59]
Right? Now, the question is, like, where do you... Where does that control, like, start and stop?

**Brandon Anderson** [12:04]
Mm-hmm.

**Dylan Patel** [12:04]
Right? Um, 'cause one could say, "Okay, well, let's just control the chips," um, uh, or, "Let's just control the AI," but then it's like, okay, well then you let them buy the chips, and they're able to, they're able to do everything they want to do anyways.

**Brandon Anderson** [12:16]
Yeah.

**Dylan Patel** [12:16]
Right? And this is, this has existed across many technologies, right? Um, you know, China, China has great engineering. You only remove one piece of the puzzle, they're able to re-engineer that last, uh, bit of the puzzle. You know, that's, that's one context, right?

And so, so that, that is, like, some people's argument, right?

**Brandon Anderson** [12:31]
Mm-hmm.

**Dylan Patel** [12:31]
So for example, if you look at Nvidia or you look at, like, David Sacks, I think their argument is like, you know, "Hey, um, you know, we should not let them have the models, but we should let them have the chips and everything else because then, uh, they're still relying on American ecosystem, American talent, American, um, you know, technology-

**Brandon Anderson** [12:50]
Yeah

**Dylan Patel** [12:50]
... um, American platforms."

**Brandon Anderson** [12:51]
Or natural resources. Mm-hmm.

**Dylan Patel** [12:51]
The other argument is like, "Hey," like, "look, this stuff, um, you know, if, if they don't have access to our models, but they have access to our chips, that alone gives them," um... Shoot.

**Brandon Anderson** [13:03]
Yeah.

**Dylan Patel** [13:04]
"That alone gives them all the control and capabilities they need, uh, to, to basically, um, be on par or just slightly behind."

**Brandon Anderson** [13:13]
Mm.

**Dylan Patel** [13:13]
Right? And so we see that in the current regime today, right?

**Brandon Anderson** [13:16]
Yeah.

**Dylan Patel** [13:16]
China has effectively, uh, great access to chips, um, not as much... Uh, not completely unfiltered access, right? They either have to rent it-

**Brandon Anderson** [13:24]
Mm-hmm

**Dylan Patel** [13:24]
... or they have to smuggle it.

**Brandon Anderson** [13:26]
Mm.

**Dylan Patel** [13:26]
Or there's some chips that are allowed. Um, and, and China's not that far behind, right? You look at, like, Kimi K- K2-5, uh, agent swarms, and it's like, well, this is, like, not worse than Codex by a marginal amount.

Maybe 5.3, right?

**Brandon Anderson** [13:38]
Yeah.

**Dylan Patel** [13:39]
But 5.2, it's, like, not worse than Codex by an ama- like, a, a large amount.

**Brandon Anderson** [13:43]
Mm-hmm.

**Dylan Patel** [13:43]
Um, and so there is, there is, like, an argument to be made that, like, hey, current regime ha- does not have the US leading in AI by enough, right?

**Brandon Anderson** [13:52]
Mm. Mm-hmm, mm-hmm.

**Dylan Patel** [13:52]
Um, and, and that's sort of the argument that, like, a Dario would make, I think, is that, look, like, current regime of export controls, China is, is still way caught up. Uh, China's still, um, you know, not that far behind.

**Brandon Anderson** [14:05]
Yeah.

**Dylan Patel** [14:05]
Um, and it would be disastrous if they are, um, in, in, in, in that... in their eyes, right?

**Brandon Anderson** [14:10]
Mm-hmm.

**Dylan Patel** [14:11]
And so the question is, how does one, how does one, um, right that r- circle, right? So one could either, A, antagonize China more-

**Brandon Anderson** [14:19]
Yeah

**Dylan Patel** [14:19]
... ban more things. Um, does that risk a Taiwan invasion, or does that, um, does that lead to something, say, say catastrophic? Does that, does that alienate researchers in, in America? Um, that's one argument. Another one is like, well, look, this is, like, we've got, like, two years, right?

You know, AI27 bros. Right? Um, and I'm not fully there, right, on terms of AI27, but I'm pretty bullish on AI generally.

**Brandon Anderson** [14:39]
Yeah.

**Dylan Patel** [14:39]
And so then the, the other argument is like, look, we only have a few years until the, um, capabilities of AI accelerate GDP growth, and it's, it's... By the way, am I cooking som- am I cutting something else?

**Brandon Anderson** [14:50]
Um, yeah. Okay.

**Dylan Patel** [14:52]
Oh, you did matchsticks, huh?

**Brandon Anderson** [14:53]
So you, you can start cutting. We need to mince the garlic and then the ginger. Um, but-

**Dylan Patel** [14:58]
All right. I'm gonna, I'm gonna-

**Brandon Anderson** [14:58]
... the garlic and the ginger has to be a lot finer.

**Dylan Patel** [15:01]
Okay.

**Brandon Anderson** [15:01]
Um.

**Dylan Patel** [15:02]
Um...

**Brandon Anderson** [15:03]
But yeah, continue.

**Dylan Patel** [15:04]
I felt, I felt a little awkward just- ... talking and not doing anything.

**Brandon Anderson** [15:08]
Yeah. Okay.

**Dylan Patel** [15:08]
And I feel like...

**Brandon Anderson** [15:09]
So we're just cutting all the vegetables to be ready.

**Dylan Patel** [15:11]
Yeah, yeah.

**Brandon Anderson** [15:11]
Yeah.

**Dylan Patel** [15:11]
Sounds good. Sounds good. The one stance that, like, a lot of, like, um, folks have is that, you know, yeah, like, let's, let's be more anti-China, um, 'cause there's only a few years left-

**Brandon Anderson** [15:24]
Mm-hmm

**Dylan Patel** [15:24]
... and so, like, you know, until- until, like, super powerful AI systems, and those super powerful AI systems will make next generation AI systems, right?

**Brandon Anderson** [15:31]
Mm. Mm-hmm.

**Dylan Patel** [15:31]
And that's, like, that's, like, a lot of the, um, argument that, like, let's say the Darios of the world would make.

**Brandon Anderson** [15:38]
Yeah.

**Dylan Patel** [15:38]
Um, and I'm sympathetic to that argument to some degree as well, right?

**Brandon Anderson** [15:40]
Okay.

**Dylan Patel** [15:41]
Um, if we start looking at, like, hey, um, this year Google's spending two hun- uh, Amazon's spending $200 billion, Google's spending $180 billion on, on AI infrastructure primarily, right?

**Brandon Anderson** [15:52]
Yeah.

**Dylan Patel** [15:53]
Um, this is, you know, 4X what they were doing just not too long ago.

**Brandon Anderson** [15:57]
Mm-hmm.

**Dylan Patel** [15:57]
Um, and if they get returns on that of any degree, then we're talking about trillions of dollars of economic value-

**Brandon Anderson** [16:02]
Yeah

**Dylan Patel** [16:02]
... being added, um, in, in just the next c- uh, handful of years.

**Brandon Anderson** [16:07]
Yeah.

**Dylan Patel** [16:07]
And so, you know, the, the, the risk here is that, like, okay, um, whatever AI is capable of doing, it's obviously adding hugely to, to the economy, and you wanna do anything and everything you can to slow China down.

We've never had an explos- explosive growth like this.

**Brandon Anderson** [16:22]
Mm-hmm.

**Dylan Patel** [16:22]
We're on the cusp of it, right, right, right now. We only have, like, you know, um, you know, the AI industry maybe does 50 bill of revenue.

**Brandon Anderson** [16:28]
Mm-hmm.

**Dylan Patel** [16:29]
Um, you know, acro- And, uh... But, like, we're seeing it explode, right? Anthropic's, you know, adding two, three billion of revenue a month now-

**Brandon Anderson** [16:35]
Mm-hmm

**Dylan Patel** [16:35]
... versus they were just adding a few hundred million of revenue a month earlier. Um, so clearly we're in the takeoff period.

**Brandon Anderson** [16:41]
Yeah.

**Dylan Patel** [16:41]
Um, and so, so the argument there is that, like, le- let's just limit them completely-

**Brandon Anderson** [16:45]
Mm-hmm

**Dylan Patel** [16:46]
... um, so that they don't end up with, um, all the, all these super powerful AI systems. I, I, I'm, I'm a little bit, like, you know, it's, it's hard to rationalize e- you know, e- every specific, like, argument just because, like, you know, I could, I could, I could argue or I could steelman, uh, any of the arguments, right?

**Brandon Anderson** [17:03]
Yeah.

**Dylan Patel** [17:03]
Actually, okay, no, maybe, maybe, maybe China is... um, should be sold our chips because at the end of the day, they're still relying on our chips now for their AI systems.

**Brandon Anderson** [17:12]
Mm-hmm.

**Dylan Patel** [17:13]
Um, and, and then they have less incentive to invade Taiwan.

**Brandon Anderson** [17:15]
Yeah.

**Dylan Patel** [17:15]
And if they invade Taiwan, then, you know, all of a sudden, um, the whole party stops and we can't do anything. And when you think about what's the capability of, uh, China versus the US to, um, to, to have a, a vertical supply chain in-

**Brandon Anderson** [17:31]
Mm

**Dylan Patel** [17:31]
... um, chips or AI or in really anything-

**Brandon Anderson** [17:34]
Mm-hmm

**Dylan Patel** [17:34]
... uh, the... China, China has by far the most advanced, uh, supply chain in semiconductors if you just look at China itself.

**Brandon Anderson** [17:40]
Mm.

**Dylan Patel** [17:40]
Right? In which case, like, if the, the lifeblood of AI is compute to some extent, uh, to a large extent, then China would win if we didn't have Taiwan, right?

**Brandon Anderson** [17:49]
Yeah.

**Dylan Patel** [17:50]
And maybe that timescale would be way longer because Taiwan is so far ahead in the production capacity and China doesn't have the equipment ecosystem like the rest of the world does. Um, but at the end of the day, that's, that's exactly, like, sort of the argument that, that one would make is, is, hey, like- You know, if you push China too far, now they might invade, um, they might invade Taiwan, and that ends up with, with this catastrophic scenario.

**Brandon Anderson** [18:14]
Yeah. Double-clicking on something you mentioned already, with these big AI labs and even hyperscalers somewhat overextending on, um, future spend with these data centers, would there ever be something that would prompt you to get a little concerned? 'Cause it seems right now we're in a very, um, acceleration type of moment for AI where still adoption for what, even like Claude Code and a lot of these great tools, um, isn't as mainstream.

### AI Backlash

**Brandon Anderson** [18:38]
But are there things that if you see happening in the next, what, six months or year that would kind of concern you in terms of where the future lies of these companies maybe overextending a little too much, and whether concerns of being in a bubble may actually have some merit to it?

**Dylan Patel** [18:56]
I mean, I think, I think sort of not answering your question, but talking about something else I want to talk about-

**Brandon Anderson** [18:59]
Yeah

**Dylan Patel** [18:59]
... which I love to do.

**Brandon Anderson** [19:01]
That's fair.

**Dylan Patel** [19:02]
Is, is sort of like you mentioned, like these doomsdayers and these believers, I think the biggest risk is actually just like the general public hates AI.

**Brandon Anderson** [19:09]
Mm.

**Dylan Patel** [19:09]
Right? Um, you know, I think, I think if you go literally anywhere, the general public absolutely hates AI so much. They, they have literally no, um... They don't, yeah, they, I mean, like, you know, you go to your random artist and they like hate AI.

You go to your random like person in rural America, they're like, "Screw AI, it's like, you know, taking all the water." Um, you know, completely like nonsense arguments-

**Brandon Anderson** [19:33]
Yeah

**Dylan Patel** [19:33]
... but like it doesn't matter. Um, so when you look across the ecosystem, you've got that problem.

**Brandon Anderson** [19:40]
Mm-hmm.

**Dylan Patel** [19:40]
Um, and so like, and then you've got the doomsdayers, right? There's also the general public, which like doesn't quite understand AI or maybe they do, um, you know, y- and, and, and they're just like so worried about AI, like taking, taking our jobs and like all sorts of other things, right?

Um, and so-

**Brandon Anderson** [19:55]
Taking our jobs

**Dylan Patel** [19:56]
... I think, I think that's, that's another aspect of this that's like quite interesting is, um, you know, general public hates it. Um-

**Brandon Anderson** [20:03]
Yeah

**Dylan Patel** [20:03]
... the doomsdayers, you know, look, I'm, I'm, I'm, I'm a, I'm a bit of a too, uh, live in the moment person to like think, you know, hey, what exactly does alignment mean? And, and does AI kill us all?

Um, you know, uh, uh, obviously there's huge risk to that.

**Brandon Anderson** [20:17]
Mm-hmm.

**Dylan Patel** [20:17]
Uh, that's not what I'm an expert in, so I don't, I don't really care to opine too much. Um, but, um, a- a- as far as like, you know, is, is it a bubble? Are people, are people, are we, are we doing too much?

Like what's going on?

**Brandon Anderson** [20:32]
Mm.

**Dylan Patel** [20:33]
Um, is there a bubble? Are we doing too much? Um, you know, the question is, um, you know, if, if AI model progress slows down, then of course we're in a bubble.

**Brandon Anderson** [20:43]
Yeah.

**Dylan Patel** [20:43]
That's, that's obvious. Um, but you know, the, the, the way that progress is accelerating, uh, so fast, right? Um, you know, you, you, you see it m- uh, month on month on month, right? I mean, you know, the new models are coming out every month, every week almost it feels like nowadays.

**Brandon Anderson** [20:58]
Yeah.

**Dylan Patel** [20:58]
Um, or new th- capabilities, right? Just think about like, like this year so far, right?

**Brandon Anderson** [21:03]
Mm.

**Dylan Patel** [21:03]
Um, you know, obviously Claude 4.5 came out and Claude Code came out, uh, last year.

**Brandon Anderson** [21:08]
Mm-hmm.

**Dylan Patel** [21:09]
But adoption really upticked in the beginning of this year.

**Brandon Anderson** [21:12]
Mm.

**Dylan Patel** [21:12]
Um, and so we saw a huge uptick, uh, just in this mo- uh, just in January, it went from, uh, 4% of, um, or 2% of, uh, commits on GitHub to 4% of GitHub commits were done by Claude Code.

**Brandon Anderson** [21:25]
Yeah.

**Dylan Patel** [21:25]
Right? That does not mean, you know, a lot of people use Claude Code without having Claude commit for them. So you're like, you're talking about like, or, and then they use Codex, they use Cognition, they use, you know, or sorry, Devin, they use all these other platforms.

### Adoption Surge

**Dylan Patel** [21:35]
They use GitHub Copilot, they use Codex. You know, we're probably at 10% or so of total code is being committed, um, or written by AI, if not more.

**Brandon Anderson** [21:42]
Mm-hmm.

**Dylan Patel** [21:43]
Um, but at least for Claude Code itself, it went from 2 to 4% in one month.

**Brandon Anderson** [21:46]
Mm-hmm.

**Dylan Patel** [21:46]
Right? Um, this, this acceleration is, is like, I think it's the thing that we've all been waiting for.

**Brandon Anderson** [21:53]
Yeah.

**Dylan Patel** [21:53]
Um, 'cause like fine people are using ChatGPT, that's great. Fine people are using like Image Gen, that's great. These were not like things that add trillions of dollars to the economy. These are social networks. These are like, you know, help help me on my, uh, homework.

These are like chill things.

**Brandon Anderson** [22:07]
Yeah.

**Dylan Patel** [22:07]
Um, but now we're, we're in the stage where it's like, no, no, no, like these are trillions of dollars of, uh, economic value that could be added. Um, and if you think about worldwide software developer wages, $2 trillion in wages, you end up with, uh, a pretty incredible amount of, um, spend that could happen.

Um, I feel like it's my worst interview ever because I'm, I'm, I'm like so focused on chopping garlic- ... and not cutting off my fingers. Um, I wonder if people are gonna like trash me for my lack of claw, like grip.

'Cause like I've tried to do it, but then I've failed, and I've done like more dangerous cutting grips.

**Brandon Anderson** [22:42]
Mm-hmm.

**Dylan Patel** [22:44]
I, I do wonder how, how, how vicious is, is the audience?

**Brandon Anderson** [22:47]
We'll see.

**Dylan Patel** [22:48]
Um-

**Brandon Anderson** [22:49]
But yeah, you're talking about Claude Code adoption and GitHub commits being significantly more, especially this, um, past like couple months, and like the, uh, value that Claude Code will have towards the economy. Do you wanna elaborate more on that, or was that a finished-

**Dylan Patel** [23:05]
Yeah, yeah. I mean, I think, I think like there's, there's quite a bit of, um... I'm definitely claw gripping now, you know. This is- I gotta make up for it now that I'm cognizant. Um,

so I think, I think, um, the adoption of AI has been so accelerated over the last month.

**Brandon Anderson** [23:24]
Yeah.

**Dylan Patel** [23:24]
You know, just, just think about everything that's happened in the last month. We've had Claude Code happen, we had Claude Bot, we had, uh, Malt Book. Uh, now we have, uh, you know, we had Kimmy K 2.5 Swarms.

We had Codex, um, 5.3, which is, you know, a significant step up as well, um, and seems better in some specific areas. And, and these are, these are, there's so many areas, right?

### CapEx Backlash

**Brandon Anderson** [23:44]
Mm-hmm.

**Dylan Patel** [23:44]
And it's like, um, at least internally, like at my company, we've, we've completely, um, flipped over, right? Like about, about a third of the company's engineers, about a third of the company's hedge fund people, and about a third of the company is like passionate individuals.

**Brandon Anderson** [23:55]
Mm-hmm.

**Dylan Patel** [23:56]
Um, you know, and so the ex hedge fund people, they are, they're all in on Claude Code now too, right?

**Brandon Anderson** [24:00]
Mm.

**Dylan Patel** [24:00]
They scrape data, they do pro- they do financial modeling, they do pro forma financial modeling with Claude Code as the assistant.

**Brandon Anderson** [24:07]
Mm.

**Dylan Patel** [24:07]
Um, and so there's a variety of like sort of like, um- You know, I think, I think, like, i- we, we've hit, like, sort of escape velocity in all these things. And so, you know, over the last month or over the last two weeks we've had, um, you know, the hyperscalers report earnings.

**Brandon Anderson** [24:23]
Mm-hmm.

**Dylan Patel** [24:24]
Um, and everyone's stocks have gone down, right?

**Brandon Anderson** [24:26]
Mm-hmm.

**Dylan Patel** [24:27]
Um, Google announced $180 billion of CapEx, the stock went down. And then Amazon announced $200 billion of CapEx, and their stock went down, like, I wanna say, like, 10%. So the market hates it, but they don't realize, like, you know, these CapEx decisions are because they see the light at the end of the tubble- tunnel, if you will, right?

**Brandon Anderson** [24:44]
Yeah.

**Dylan Patel** [24:44]
Um, the amount of, the amount of adoption is just insane. So now that we've had these companies, like, report earnings, and they've disclosed what their plans are for the year, and they're much higher than almost anyone predicted-

**Brandon Anderson** [24:54]
Mm-hmm

**Dylan Patel** [24:54]
... um, you've, you've, you've got, you've got-

You've got the market just hating it. Um-

**Brandon Anderson** [25:00]
Yeah.

**Dylan Patel** [25:02]
And, and, and, and, and so that, that, that brings, like, this interesting conundrum which is that, like, okay, the market is mad they're spending all this money on compute CapEx-

**Brandon Anderson** [25:10]
Yeah

**Dylan Patel** [25:10]
... but these companies know much better than, than, than you, right? Like, a- and by you, I mean the investor.

**Brandon Anderson** [25:15]
Yeah.

**Dylan Patel** [25:15]
Right? Um, in reality, they're spending this much because they see insane amounts of demand, right?

**Brandon Anderson** [25:21]
Mm-hmm.

**Dylan Patel** [25:21]
Um, you know, An- Anthropic doesn't just add $2 billion of revenue in one month, uh, you know, wi- without having, you know, huge demand.

**Brandon Anderson** [25:30]
Mm-hmm.

**Dylan Patel** [25:30]
Um, and they're doing it at positive margins, right?

**Brandon Anderson** [25:32]
Yeah.

**Dylan Patel** [25:33]
Um, and they're doing it, you know, and, and, and when they go to everyone they're like, "Look, guys, we need more compute. We need more compute. We need more compute." Um, and so, so it, it's, it's ... You know, after three years straight of everyone, of the AI labs saying, "We need more compute," the hyperscalers are now seeing it's not just we need more compute 'cause we wanna train bigger models and we wanna do more research.

It's actually we need more compute because we need w- we need to serve our users, right?

**Brandon Anderson** [25:54]
Mm-hmm.

**Dylan Patel** [25:55]
Um, we need to add hundreds of millions of dollars or billions, actually billions of dollars of compute, right? If Anthropic add $2.5 billion of revenue, um, and their gross margin is 40%, they added, like, $1.5 billion of compute in one month.

**Brandon Anderson** [26:07]
Mm-hmm.

**Dylan Patel** [26:07]
Right? To, just to serve that. You extrapolate that line out a little bit and you're like, "Holy crap, you know, they actually need hundreds of billions of dollars of compute. Um, okay, fine. Well, we, we need to build this all," right?

'Cause it's a front run of, you know, you build it and then they can rent it.

**Brandon Anderson** [26:21]
Yeah.

**Dylan Patel** [26:21]
Um-

**Brandon Anderson** [26:23]
So the bet here is that growth will continue to accelerate, and the amount of money that Anthropic and OpenAI makes will just continue to go up?

**Dylan Patel** [26:31]
Yeah, exactly. And, and, and, and I don't n- I, you know, like, look, the party can stop at some point. Like, you know, that's for sure.

**Brandon Anderson** [26:37]
Mm-hmm.

**Dylan Patel** [26:37]
Um, at any point ... You know, I think my favorite thing was, uh, I, I twee- I tweeted about, like, people were like, "Oh, who expected this CapEx?" And then I was like, "Well, we did," right? And then someone replies, "The whale watcher told you that you're gonna see whales.

Wow, surprising." I was like, wow, that's a pretty good, uh, uh, reply to me, you know? But, like, anyways, like, it's like-

**Brandon Anderson** [26:56]
Yeah

**Dylan Patel** [26:57]
... you know, uh, you know, obviously, obviously I'm the whale watcher here. Um-

**Brandon Anderson** [27:00]
Mm-hmm

**Dylan Patel** [27:01]
... you know, we see the CapEx coming.

**Brandon Anderson** [27:03]
Yeah.

**Dylan Patel** [27:03]
So the market doesn't like it, but it's clearly obvious that it's needed.

**Brandon Anderson** [27:06]
Mm-hmm.

**Dylan Patel** [27:07]
Um, and, and, and now the discussion is sorta like, you know, I think a couple years ago I said the hyperscalers would have no free cashflow.

**Brandon Anderson** [27:14]
Yeah.

**Dylan Patel** [27:14]
Right? I.e., they would not have any, uh ... They would not be generating profits and buying back their stock in a short amount of time.

**Brandon Anderson** [27:21]
You could also transfer your onions and some veg to the bowl, white bowl on your left, I believe, or right.

**Dylan Patel** [27:27]
Yes.

**Brandon Anderson** [27:27]
Um, so on my cutting board I just have garlic, ginger, and the green part of the scallions. So leave the green part for a garnish for later, so just keep it on.

Um, I guess at what point do you think the market would be, um, satisfied or okay with this CapEx?

**Dylan Patel** [27:46]
I think, I think the market is gonna get really mad at the hyperscalers. Um, they haven't really yet.

**Brandon Anderson** [27:51]
Mm-hmm.

**Dylan Patel** [27:51]
Um, but we saw the, we saw the s- signs of it at the beginning, uh, at, about mid last year, right?

**Brandon Anderson** [27:56]
Okay.

**Dylan Patel** [27:56]
Um, for example, Oracle peaked when they announced ... within a week after announcing, um-

**Brandon Anderson** [28:01]
The OpenAI deal

**Dylan Patel** [28:01]
... that they were gonna do, you know, $300 plus billion of deals with, uh, OpenAI.

**Brandon Anderson** [28:06]
Mm-hmm.

**Dylan Patel** [28:06]
Um, and the market really, uh, peaked, uh, around then.

**Brandon Anderson** [28:12]
Mm-hmm.

**Dylan Patel** [28:14]
Um, and, and, and then, like, since then they've gone down. And, and, like, other darlings that were, like, doing AI infrastructure, like CoreWeave, uh, have also peaked, right?

**Brandon Anderson** [28:24]
Mm-hmm.

**Dylan Patel** [28:24]
So now we've got this, like, interesting, um, conundrum where now the hyperscalers are starting to say how much CapEx they're gonna do.

**Brandon Anderson** [28:33]
Yeah.

**Dylan Patel** [28:33]
Um ...

And, and, and, and so when we, when we think about, like, hey, what's gonna, what's gonna end up happening is, you know, these hyperscalers are gonna keep spending, right?

**Brandon Anderson** [28:43]
Mm-hmm.

**Dylan Patel** [28:43]
What is their biggest advantage, right? It's, it's, it's that they can build the most infrastructure in the world.

**Brandon Anderson** [28:48]
Mm-hmm.

**Dylan Patel** [28:48]
They've built the organization to build, uh, infrastructure faster than anyone else. Um, wow, this, like, got s- ki- cooks super fast.

**Brandon Anderson** [28:57]
Show the camera.

**Dylan Patel** [29:00]
It's just eggs, bro. Um, I just don't generally use an, uh, induction, right?

**Brandon Anderson** [29:08]
Yeah, they heat up very fast, so just as a heads-up.

**Dylan Patel** [29:11]
Yeah. Um, so, so, so the hyperscalers are, you know, have been the most profitable companies to ever exist in the humani- in humanity.

**Brandon Anderson** [29:17]
Mm-hmm.

**Dylan Patel** [29:17]
Right? Whether it's Meta through ads, uh, whether it's, um, Google through Search, uh, whether it's Amazon through AWS and, uh, amazon.com. Um, so on and so forth, right? Uh, Microsoft through, you know, Windows plus Office 365 plus, uh, Azure, right?

Um, they've, they've all been the most profitable companies. Am I, am I gonna continue with the onions and such?

**Brandon Anderson** [29:39]
Yeah. Or so ... Are the eggs cooked?

**Dylan Patel** [29:40]
Yeah.

**Brandon Anderson** [29:41]
Okay. You could put the eggs on the plate.

**Dylan Patel** [29:44]
Yeah, already done.

**Brandon Anderson** [29:44]
Okay. And then now add some oil and then, yeah, add all the veggies.

**Dylan Patel** [29:48]
Yeah, yeah.

**Brandon Anderson** [29:48]
But make sure not to add too many onions, 'cause there'll probably be more onions in carrots and other thing. Just try to have, like, an even balance, um, of vegetables.

**Dylan Patel** [29:57]
Yeah.

**Brandon Anderson** [29:59]
Yeah. And then, yeah, so just get some color on the veg.

**Dylan Patel** [30:04]
Um, do you, do you, uh, tend to, um ... Do you tend to do the carrots and onions at the exact same time?

**Brandon Anderson** [30:11]
Yeah. Like, doing them all together is probably just easiest. So carrots, onions, and the white part of the scallions

**Dylan Patel** [30:19]
Okay. Um,

### Hyperscaler Dilemma

**Dylan Patel** [30:27]
so, so the hyperscalers have been the most profitable companies ever.

**Brandon Anderson** [30:30]
Yeah.

**Dylan Patel** [30:30]
Um, and now they're, they're about to face, like, sort of this, like, interesting conundrum, right?

**Brandon Anderson** [30:34]
Yeah.

**Dylan Patel** [30:34]
There's a huge innovator's dilemma here, whether it's, you know, Meta, Meta doesn't own the platform, um, and, uh, wherever people's eyeballs are is where people are gonna, like, uh, spend their cash. Um, or in, in, in the case of Google, right, AI can disrupt search.

Or in the case of Microsoft, right, productivity, uh, suite is where they make all the money, right?

**Brandon Anderson** [30:56]
Mm-hmm.

**Dylan Patel** [30:56]
Um, Office 365, Windows, et cetera. But things like Claude Code, Claude Bot, and, and future iterations of it, just take a little imagination-

**Brandon Anderson** [31:03]
Yeah

**Dylan Patel** [31:03]
... will displace those immediately.

**Brandon Anderson** [31:04]
Mm-hmm.

**Dylan Patel** [31:04]
Right? Um, you know, you've got like, uh, you know, same with Amazon, right? AWS is, is, is, is a general purpose, uh, you know, AI infrastructure and, or sort of infrastructure play. Uh, but there's a lot of risk with everything else, um, with everyone getting disrupted quite heavily.

**Brandon Anderson** [31:21]
Mm-hmm.

**Dylan Patel** [31:21]
And so they've got this, this, this dilemma where they could get, um, disrupted heavily. Um, at the same time, they've also got, uh, they've also got this challenge with regards to, um, potentially, you know, being beaten, right? So they, they have to invest hugely in AI.

They have to try and win AI.

**Brandon Anderson** [31:37]
Yeah.

**Dylan Patel** [31:38]
Um, and if they don't, then they're really, really screwed. Um, but right now, the demand for AI is insatiable-

**Brandon Anderson** [31:44]
Mm-hmm

**Dylan Patel** [31:45]
... and they can get pretty good returns just by building infrastructure and renting it out to the labs. Uh, but they'll obviously get way better returns if they, uh, if they have AI models in-house. So they need to spend like crazy to do this.

Um, and at the same time, um, if everyone else is like, sort of, it's like, um, it's Pascal's wager, right?

**Brandon Anderson** [32:03]
Mm-hmm.

**Dylan Patel** [32:03]
Um, if I don't spend like crazy and others do, I lose, right? If I don't believe in God, right, digital god coming, then, and others do, and it happens, then I'm, I'm a loser, right?

**Brandon Anderson** [32:12]
Yeah.

**Dylan Patel** [32:12]
Um, and so they've all got this dilemma, and the only solution is I have to spend more and more, you know, until, uh, uh, to, to tr- keep up in the race. And so, you know, this year's announcements of $180 billion of CapEx from Google and 200 from Amazon, right, which is, you know, four X what they were just doing a few years ago-

**Brandon Anderson** [32:31]
Mm-hmm

**Dylan Patel** [32:31]
... um, is, is quite incen- in- intense. Uh, but in addition to that, we're looking at, um, we're looking at this, like, skyrocketing in the next few years, right?

**Brandon Anderson** [32:41]
Yeah.

**Dylan Patel** [32:41]
There's no reason why Google will have any profit, uh, in '27 at all, right, in terms of cash flow.

**Brandon Anderson** [32:47]
Mm-hmm.

**Dylan Patel** [32:47]
They will just spend every dollar they make on, on AI infrastructure. Um, and I, I think that's at least my belief, um, and, and AI models and so on and so forth, because that, that, that's basically my belief.

And the market hasn't fully woken up to this realization.

**Brandon Anderson** [33:00]
Mm-hmm.

**Dylan Patel** [33:01]
Um, we've been saying it for a couple of years. Um, in fact, we even did a piece, uh, last year, which was like, how much debt can the hyperscalers borrow, right? 'Cause at some point, you know, they, they, they have to lever up on building capacity, right?

So an example of this is Meta, right? Meta is not as large as Google and Amazon, but they wanna be in the race.

**Brandon Anderson** [33:17]
Yeah.

**Dylan Patel** [33:18]
Um, and so they've already started taking some debt on to build their data centers. Now, obviously, they have a tremendously profitable business that could pay it off. Uh, they just have to stop spending the money, um, on CapEx that's not necessary.

Uh, but, but Zuckerberg has woken up and fully realizes this. How much do I wanna cook the onions and car- carrots?

**Brandon Anderson** [33:34]
Just get some color on it, and then just put it in the plate with the eggs.

**Dylan Patel** [33:39]
Okay. Um, and so, so you've seen some hyperscalers such as, uh, Meta, they've, they're already taking debt on for their largest AI cluster in Louisiana.

**Brandon Anderson** [33:49]
Mm-hmm.

**Dylan Patel** [33:49]
Um, you know, they're taking like $40 billion of debt on for that. Uh, but they're in the market to take on much, much more. Um, Google and Amazon haven't taken on debt yet for AI infrastructure, but they will, right?

**Brandon Anderson** [33:58]
Mm-hmm. Yeah.

**Dylan Patel** [33:59]
And, and so I think, I think people really realize and panic and, and, and probably this year when they see all of these companies doing exactly this, right?

**Brandon Anderson** [34:08]
Yeah.

**Dylan Patel** [34:08]
Are they gonna... You know, what ends up happening, uh, when the most profitable companies that have ever existed, which have compounded at double digits for, uh, over, like a decade and a half-

**Brandon Anderson** [34:18]
Yeah

**Dylan Patel** [34:18]
... now all of a sudden say, "We're not gonna, we don't care about profit anymore. We're just building pixie dust," right?

**Brandon Anderson** [34:23]
Mm-hmm.

**Dylan Patel** [34:23]
"We're building digital God. Um, and if you believe in it, great. If you don't, then, you know, tough luck." Um, and, and all this CapEx predates infant revenue, right? 'Cause you need to have spent the CapEx, brought on the clusters and all that before you can have the, um, before you can ever have the, um, the revenue come online.

**Brandon Anderson** [34:41]
Mm-hmm.

**Dylan Patel** [34:41]
Right?

**Brandon Anderson** [34:41]
Like cloud revenue.

**Dylan Patel** [34:42]
And then the revenue starts off at lower margin, right?

**Brandon Anderson** [34:45]
Yeah. Also, it's time to c- cook the chicken, so just add some oil and then we'll cook chicken-

**Dylan Patel** [34:48]
Okay, sounds great

**Brandon Anderson** [34:49]
... when we're done. Yeah. Yeah, so you're saying investment before the revenue comes online?

**Dylan Patel** [34:55]
Yeah, yeah. So, so there's like a timeline lag when revenue comes online. Uh, there's a timeline lag in terms of like when you rent the infrastructure versus like, hey, you have to train the model before you can ever start to, um,

um, before you can ever start to

actually get the like, uh, AI service revenue, right?

**Brandon Anderson** [35:16]
Mm-hmm.

**Dylan Patel** [35:17]
Um, and so you've seen this like with like all the vendors, right? Like, you know, there, there's huge spend for OpenAI, Anthropic, and Google on training models. Um, and, and the others are doing it too, like, uh, Amazon and such, uh, before they ever end up with enough revenue, uh, oc- oc- g- generating revenue from the models from the services that they sell on top.

And so the market is just gonna really hate this.

**Brandon Anderson** [35:39]
Yeah.

**Dylan Patel** [35:39]
And I feel like that's gonna lead to, despite the fact everyone in San Francisco is gonna see revenue skyrocketing, they're gonna see all the amazing capabilities, uh, but they, we live in a bubble, right? If you told- Uh, we, we put out some research that was like, "Hey, 4% of commits on GitHub are Claude Code," and everyone at SF is like, "That's too low," right?

Like, it's like 100% of mine.

**Brandon Anderson** [35:58]
Yeah.

**Dylan Patel** [35:58]
Maybe it's 50% for all the boomers, right?

**Brandon Anderson** [36:00]
Mm-hmm.

**Dylan Patel** [36:00]
And it's like, no, no, no. Like, you know, we've got a lot of adoption to go. Um, and so people are gonna like, sort of like see all this amazing model progress and revenue growth and adoption in SF, but then like in New York and in London and like Hong Kong and other financial capitals of the world, Singapore, et cetera, people are gonna see the exact opposite, right?

**Brandon Anderson** [36:19]
Mm-hmm.

**Dylan Patel** [36:19]
They're gonna see, they're gonna see the most profitable companies ever are destroying their business model to build, um, capacity in something that maybe doesn't, necessarily doesn't have returns.

**Brandon Anderson** [36:30]
Yeah.

**Dylan Patel** [36:30]
Um-

**Brandon Anderson** [36:30]
Like a big bet towards

**Dylan Patel** [36:31]
... and, and so that I think is gonna... And then, and likewise, they're gonna see the general public fucking hates AI, and you're gonna, you likely see like- A real backlash to AI from both the financial class and the normal people of the world.

Um-

**Brandon Anderson** [36:43]
Does it matter, though, if the public hates AI if it provides a lot of value towards enterprise and companies? Like, isn't the main value and profit coming from enterprises more so than the general public?

**Dylan Patel** [36:54]
Exactly. And I think, I think that's, that, like, is, like, the big fear, right?

**Brandon Anderson** [36:58]
Mm-hmm.

**Dylan Patel** [36:58]
You know, we've already had, like, decades of, like, people being like, "Hey, income inequality is bad." Um, and the value of labor has been, has been falling, right?

**Brandon Anderson** [37:07]
Yeah.

**Dylan Patel** [37:07]
Uh, the value of labor used to be way, way higher, right?

**Brandon Anderson** [37:10]
Mm.

**Dylan Patel** [37:10]
Um, you know, as a percentage of the economy. But as we've mechanized, as capital has become more and more important, as machinery has grown, um, we've sort of had this major change, which is that, um, you know, capital is, is taking more and more of the ear- uh, uh, earnings, uh, from, of, of, of, uh...

Capital's taking more and more of their earnings.

**Brandon Anderson** [37:32]
Mm-hmm.

**Dylan Patel** [37:32]
And, and, and so people are really mad about that, and now we're gonna start seeing huge, uh, job loss too, right? Like, hey, like, turns out there's shitloads of software developers just out of school, uh, who can't get jobs.

Okay, fine. But what about, like, the 2 million people who drive cars for a living? Okay, well, Waymo works well. Tesla Robotaxi's starting to be deployed. Zoox is starting to be deployed.

**Brandon Anderson** [37:54]
Yeah.

**Dylan Patel** [37:54]
Um, you know, we- we're starting to see really the beginnings of all that.

**Brandon Anderson** [37:59]
Yeah.

**Dylan Patel** [37:59]
Um, we're gonna see, you know, the stock market maybe does well or the econ- the GDP's gonna look good, but then normal people aren't gonna be accruing much value from it. And so more and more... You know, and eventually, like, the financial markets will not do too well either because software's imploding, because hyperscalers are gonna invest all their capital, and you're gonna end up with this, like, major, major, uh, weir- fear and worry for everyone in the industry.

**Brandon Anderson** [38:21]
Mm.

**Dylan Patel** [38:22]
Um, or, sorry, everyone in the world.

**Brandon Anderson** [38:24]
Mm-hmm.

**Dylan Patel** [38:24]
Um, and, and there's, like, an AI backlash, right? Um, and, and I think that's gonna be, like, the hottest button issue of, like, the next election.

**Brandon Anderson** [38:31]
Mm.

**Dylan Patel** [38:31]
Right? Um, if not the midterms.

**Brandon Anderson** [38:34]
Yeah.

**Dylan Patel** [38:34]
Right? Um, and it seems obvious to me that, like, any party that wants to win should just become the anti-AI party.

**Brandon Anderson** [38:40]
Mm.

**Dylan Patel** [38:41]
Um, because life as we know it is changing.

**Brandon Anderson** [38:43]
Mm-hmm. Taking a little detour, where do you think the alpha is or, like, the bet is? 'Cause you said that Nvidia's kind of covering their bases with the Ruben CPX, uh, the Groq chips, standard GPUs, um, and there's a lot more startups out there that are very specialized.

### Chip War

**Brandon Anderson** [39:00]
And even, like, a lot of YC companies, right, are, like, popping up and kind of tackling this industry. Do you think it will kind of turn out to be a play where, like, Nvidia at the end of the day still reigns and, um, like, crushes every other company that tries to take away the market share?

Or do you think there will actually be a lot of value accrued to these more specialized smaller companies?

**Dylan Patel** [39:22]
Um, in, in the, in the, in the chip space specifically?

**Brandon Anderson** [39:25]
Yeah.

**Dylan Patel** [39:26]
Um, yes, I think that's a really strong debate that, uh, people are having.

**Brandon Anderson** [39:31]
Mm-hmm.

**Dylan Patel** [39:31]
Um, is how much, how much value accrues to Nvidia? How much value accrues to the model companies? Um, how much do they start to, um, really, um... You know, do, do, do smaller chip companies take charge and, uh, win?

Um, and it is really a innovator's dilemma in the sense that, like, you know, hey, why did, why did Intel and AMD not win in AIG fuse? Well, it's because they were, they were making money off of CPUs, and Nvidia was focused on parallel computing.

**Brandon Anderson** [39:59]
Mm-hmm.

**Dylan Patel** [40:00]
Um, and now you've got, got sort of the same question, which is, um, you know, will, will Nvidia be able to innovate on all the things that needs to be innovated, um, or, or will...

There's a lot of scraped stuff on the bottom.

**Brandon Anderson** [40:19]
Yeah.

**Dylan Patel** [40:20]
Yeah, yeah. Well, it's like there's stuff stuck in the pan, you know? So I'm trying to scrape it off. Um-

**Brandon Anderson** [40:24]
We could also get a new pan. I have two new pans.

**Dylan Patel** [40:27]
Actually, that would be amazing.

**Brandon Anderson** [40:28]
The final thing, we just gotta now add everything together. So first add some oil to the pan once it's dry, and then add the garlic ginger. But have it kind of lower. Um-

**Dylan Patel** [40:39]
Have it what?

**Brandon Anderson** [40:40]
Have, like, the temperature be lower, like, not too high.

**Dylan Patel** [40:42]
Oh, really? Okay.

**Brandon Anderson** [40:43]
Yeah.

**Dylan Patel** [40:45]
Nvidia is, um... They've kind of got this, like, innovator's dilemma. The nice thing is they embody Silicon Valley spirit more than maybe any other company-

**Brandon Anderson** [40:53]
Mm-hmm

**Dylan Patel** [40:53]
... uh, which is Andy Grove, right? Andy Grove, uh, from Intel.

**Brandon Anderson** [40:56]
Yeah.

**Dylan Patel** [40:57]
Um-

**Brandon Anderson** [40:57]
Only the paranoid survive.

**Dylan Patel** [40:58]
Only the paranoid survive, right? And, and I think Jensen Huang is, like, one of the most paranoid people- ... in the industry, right?

**Brandon Anderson** [41:04]
Yeah.

**Dylan Patel** [41:05]
Um, he's, he's constantly, like, freaking out, changing internal things. Like, you know, in a good way though, right?

**Brandon Anderson** [41:11]
Yeah.

**Dylan Patel** [41:11]
Like, truly founder mode. Um, all right, uh, the aromatics are very aromatic.

**Brandon Anderson** [41:16]
Okay. Then just add, um, all the veg. Don't add everything, um, 'cause, like, proportions. But add, like, the onions, carrots, eggs, peas, and then chicken.

And then once you have that, then add the rice, and, like, should mix everything together. And then at the very end you're gonna add soy sauce and then some sugar and salt to adjust.

Yeah. But Jensen's very-

**Dylan Patel** [41:52]
Oh, yeah, Jensen, I think-

**Brandon Anderson** [41:53]
Yeah

**Dylan Patel** [41:53]
... Jensen's very paranoid, um, and that makes him, like, an amazing founder-

**Brandon Anderson** [41:56]
Mm-hmm

**Dylan Patel** [41:57]
... um, and CEO. Um, and so you, you have all these people freaking out, but it's like the moment he sniffed wind of the OpenAI Cerebras deal-

**Brandon Anderson** [42:06]
Yeah

**Dylan Patel** [42:06]
... he immediately went out and was like, "Okay, I don't actually... I didn't actually... wasn't building this technology because I didn't believe in it, but now I do because OpenAI's trying to acq- uh, is trying to use Gro- uh, Cerebras.

So I'm just gonna go acquire Groq." Right? Like, you know-

**Brandon Anderson** [42:18]
Yeah

**Dylan Patel** [42:18]
... it's like, that's, like, why he did it, right?

**Brandon Anderson** [42:19]
Mm-hmm.

**Dylan Patel** [42:19]
So it's like, you know, there's, there's a bit of, like, um... You know, the moment he, he sniffs anything, he changes course and tune, updates his priors. Um- And I think that's, like, really impressive. And so as you step forward to, like, hey, what about, um...

You, you step forward to, like, okay, well, what does that mean for his hardware roadmap? Well-

**Brandon Anderson** [42:38]
Yeah

**Dylan Patel** [42:38]
... before he was, like, making one, one... You know, just a few kinds of architectures and chips. Uh, but primarily it was all, like, very similar, right?

**Brandon Anderson** [42:45]
Mm-hmm.

**Dylan Patel** [42:46]
It was a large GPGPU. Um, and it was, like, having... It was, like, the best memory, the best networking, everything sort of the best as possible. Um, and sort of, like, one size fits all, uh, with the main line of, like, A100, H100, B200, right?

Um, but as we look to Ruben and beyond-

**Brandon Anderson** [43:03]
Mm-hmm

**Dylan Patel** [43:03]
... right, Jensen is really, like, starting to fully embrace heterogeneity, right?

**Brandon Anderson** [43:09]
Mm-hmm.

**Dylan Patel** [43:09]
Um, you know, much like this fried rice, right, there's no one individual ingredient that shines above all, right? You've kinda gotta have a little bit of everything. Um, and so, so this is like b- this is, like, sort of, like, what Jensen's believing here.

So he's got, you know, he's got this CPX chip-

**Brandon Anderson** [43:24]
Yeah

**Dylan Patel** [43:24]
... right? Which is made for, um, context processing, prefill. It's pretty good at video and image gen as well.

**Brandon Anderson** [43:31]
Mm-hmm.

**Dylan Patel** [43:31]
Um, but it's not really good at latency sensitive applications. You know, they've of course got their main line of GPUs, and now they've got this, these Groq chips, right?

**Brandon Anderson** [43:39]
Mm-hmm.

**Dylan Patel** [43:39]
So, so, uh, you know, they're, they've... NVIDIA's sort of got every single, uh, aspect, uh, or type of chip possible now within his company. Um, and he's, he's continuing to try and, like, ex- uh, innovate and, and move as fast as possible in all these things.

**Brandon Anderson** [43:53]
Yeah.

**Dylan Patel** [43:53]
Um, and so when we think about, like, hey, what ends up happening, um, with NVIDIA in, in this case, it's, um, it's, you know, Je- Je- NVIDIA knows they will lose because they have a business, uh, model deficit, right?

**Brandon Anderson** [44:11]
Mm-hmm.

**Dylan Patel** [44:11]
Google, Amazon, they get to vertically integra- integrate, and vertical integration always saves tons of money. Um, so he has to be better than everyone-

**Brandon Anderson** [44:19]
Yeah

**Dylan Patel** [44:19]
... um, by not just, like, a little bit, by, by a ton to justify his margins.

**Brandon Anderson** [44:24]
Mm-hmm.

**Dylan Patel** [44:24]
Otherwise, the vertical integration of, of his competitors will win out.

**Brandon Anderson** [44:27]
Mm-hmm.

**Dylan Patel** [44:28]
Um, and, and, and so this is sort of, like, I think the big challenge. Um, and I think, I think the story is not finished, right? NVIDIA will remain on top this year and next year, um, based on what we see, but others will gain some ground.

Uh, and the question is what happens in the long term? Um, and, and honestly, you know, the, the, the, the cards are up in the air, right? Um, no one has the, the right to win. No one has a destiny to win.

Uh, things are moving so fast, whoever, whoever does the... you know, innovates the hardest will win. Not, not necessarily, like, oh... You know, and, and, and I think moats are as shallow as they've ever been, right?

**Brandon Anderson** [45:02]
Mm.

**Dylan Patel** [45:02]
Uh, because how fast things are moving, the v- the, the size of the numbers that are being thrown around now, right, it's hundreds of billions of dollars for each indi- uh, major hyperscaler.

**Brandon Anderson** [45:12]
Yeah.

**Dylan Patel** [45:12]
The size of the numbers are so large that you can just go and justify hiring anyone, any talent. The moats become much smaller. Um, and this, this is sort of, like, pretty, pretty big deal with regards to, you know, does NVIDIA win or not, right?

**Brandon Anderson** [45:27]
Yeah. I guess, what do you think's the biggest bottleneck for speed to keep us from going as fast as possible? Is it memory? What do you think is the main bottleneck?

### Bottleneck

**Dylan Patel** [45:37]
You know, I think I have zero wok hei, by the way.

**Brandon Anderson** [45:39]
Yeah.

**Dylan Patel** [45:40]
Um-

**Brandon Anderson** [45:41]
I don't, I don't think I'll have any wok hei with an induction stove

**Dylan Patel** [45:42]
Dude, I swear to God, if Uncle Roger finds this video, I'm gonna cry. I'm like... He's like... He's gonna be like, "No MSG, haiya. No wok hei, haiya. Induction furnace, what are you doing?" Um, no, yeah, but I, I, I, I...

But anyway, sorry. Um, you know, what's the biggest bottleneck to speed? You know, I think you cook fried rice much faster if you have a wok. Um, so okay. All right, I'm done. I'm done. I'm not good enough to justify that.

There's, like, 100 other mistakes I've made, you know? Um, but, like, I think the biggest bottleneck to, like, hey, why, why only $200 billion this year for Amazon? Why not 500, right?

**Brandon Anderson** [46:17]
Yeah.

**Dylan Patel** [46:17]
Um, I think, I think there's, like, a number of limiting factors, and it's sort of, like, year by year it's been different, right?

**Brandon Anderson** [46:22]
Yeah.

**Dylan Patel** [46:22]
In 2023, it was definitely all, um, related to chips, right? Semiconductors, CoWoS, uh, which is chip-level wafer-on-substrate. Um, l- driving up production of this was very, very difficult. And then as we step forward to 2024,

as we step forward to 2020... Oh, I didn't throw any sugar or soy sauce, no wonder.

**Brandon Anderson** [46:44]
Oh, did you serve it already?

**Dylan Patel** [46:45]
No, I didn't.

**Brandon Anderson** [46:45]
No.

**Dylan Patel** [46:45]
I almost did.

**Brandon Anderson** [46:46]
Okay. Yeah. Final step, just sugar, soy sauce to taste. So taste it as you go.

**Dylan Patel** [46:51]
I don't know how much sugar to do, but...

Dude, this shit's about to be sweeter than Panda Express.

Oh, shit, I completely forgot about the soy sauce. Um,

anyways, um-

**Brandon Anderson** [47:07]
2023, chips

**Dylan Patel** [47:07]
... you know, you know, in 2023 it was CoWoS, it was-

**Brandon Anderson** [47:09]
CoWoS

**Dylan Patel** [47:09]
... semiconductor supply chains. As we step forward to '24, '25, it started to become data centers.

**Brandon Anderson** [47:14]
Mm-hmm.

**Dylan Patel** [47:14]
Um, energy is a bigger deal in '25, '26. Um, but as we step forward, right, you know, supply chains are fast, and they react quickly, right? So this, this current, like, whole thing of, like, oh, data centers are the shortage, yes, data centers are a shortage.

**Brandon Anderson** [47:27]
Mm-hmm.

**Dylan Patel** [47:27]
Yes, power is a shortage. Um, but at the end of the day, actually, there's a lot of other shortages around, too, right? Um, you know, and, and, and, and when you think about power, okay, well, like, if you were not creative, right, and you just relied on grid power, well, there's only three companies that make dual combined cycle reactors.

**Brandon Anderson** [47:46]
Mm-hmm.

**Dylan Patel** [47:46]
But as you step forward to, like, oh, okay, well, what if I make... Uh, what if I want, um... What if I w- what if I take something else, right? What if I take, um, aeroderivative engines? Okay, there's a few more vendors.

What if I take industrial gas turbines? There's a few more vendors. What if I take, um...

What if I take, uh, medium speed reciprocating engines, right? Um, which are these, like, sort of, like, any, any company that makes diesel engines, there's dozens of them.

**Brandon Anderson** [48:14]
Mm-hmm.

**Dylan Patel** [48:15]
Uh, they can make medium speed reciprocating engines, and I can use... I could connect those up to make, um, power for the data center, right? So when I look at, when I look at, like, hey, what did... Who, who, who sort of broke these bounds, right?

Elon was the first one to sort of say, "Well, well, no. I don't care about the actual rules. Let me just, like, Let me just put power generation on site with low quality mobile turbines, right?

**Brandon Anderson** [48:34]
Mm-hmm.

**Dylan Patel** [48:34]
Not turbines even, right? Uh, industrial gas tu- uh, engines, um, you know, reciprocating engines, et cetera, et cetera. Um, so, so Elon broke all these rules, and now the, the whole industry's reacted fast enough because there's so many suppliers, right?

**Brandon Anderson** [48:47]
Mm-hmm.

**Dylan Patel** [48:47]
And the lead time to, like, ramp up production of these things is ultimately not nearly as long as it is in the semiconductor supply chain. So going back to your question of like, "Hey, what's the big bottleneck?" Um, well, there's, there, the, the big bottleneck is now back again to semiconductors, right?

**Brandon Anderson** [49:01]
Yeah.

**Dylan Patel** [49:01]
Semiconductors are extremely cyclical. Uh, the, the, the buildings that chips are made in are the most complicated buildings people make. Um, you know, they are, they are, um, you know, they have multi-year t- timelines. Uh, they require not, not just all the complexity of, like, el- electricians and plumbers that da- uh, that data centers do, but they actually require a lot more complex, um, because there's, there's all sorts of, uh, chemicals and, and precursors and so on and so forth that are going through, uh, the data center, right?

Or, or through the fab. And so, you know, people have just not built enough fabs. And then, you know, that's, that's ignoring all the complicated tools, right? These tools cost hundreds of millions of dollars in some cases.

**Brandon Anderson** [49:39]
Yeah.

**Dylan Patel** [49:39]
Um, they're the most complicated thing people make. Um, and so you end up with, wow, so the, the most, the, the, the challenge here is not just, um... It's, it's, it's ramping up production of semiconductors.

**Brandon Anderson** [49:51]
Mm-hmm.

**Dylan Patel** [49:51]
And, and so now we've entered an age in, especially in '26, but as we go into '27, '28, um, you know, and, and when we look in 2026, Google would buy a lot more GPUs, but they can't ramp production fast enough, right?

And so they have to buy tons of GPUs. When we go to '27, it applies again, right? Google simply cannot buy enough GPUs, and they have to buy-

**Brandon Anderson** [50:08]
GPUs

**Dylan Patel** [50:08]
... tons of GPUs. Um, and, and when you look across, um, the entire supply chain, no one is getting enough capacity of semiconductors.

**Brandon Anderson** [50:15]
Mm-hmm.

**Dylan Patel** [50:16]
Um, yes, they can put them in data centers. Yes, they can get the energy, uh, through maybe ghetto methods like putting reciprocating engines, right? Uh, diesel reciprocating engines or gas engines, but, like, you know, may- maybe not the most clean or, uh, efficient thing, but they can do it.

And so you end up with, oh, okay, semiconductors are the shortage, but what's the bottleneck to building more fabs, uh, or to, to building more chips is more fabs, and people just have not built these fabs yet, right?

**Brandon Anderson** [50:38]
Yeah.

**Dylan Patel** [50:38]
And that's, I think, the big bottleneck now.

**Brandon Anderson** [50:40]
Yeah, that makes sense.

**Dylan Patel** [50:40]
Um, and that's gonna persist through the end of the decade or until AI, you know, sort of slows down.

**Brandon Anderson** [50:45]
How are you feeling, Dylan? Are you good?

**Dylan Patel** [50:46]
Uh, I, I turned around and I saw yours briefly, so now I'm trying to, like- ... wipe the edges of my bowl so it looks beautiful, you know?

**Brandon Anderson** [50:52]
Yeah. No need to worry too much.

**Dylan Patel** [50:53]
No, no, no, no. People are gonna judge me hardcore.

### Tasting

**Brandon Anderson** [50:58]
How was that, Dylan? Was that fun?

**Dylan Patel** [51:00]
Uh, it, it was quite fun. I think there was the right amount of stress and, uh-

**Brandon Anderson** [51:03]
Stressful

**Dylan Patel** [51:03]
... involved. You know, I think I maybe didn't share my thoughts as well as normal, but-

**Brandon Anderson** [51:07]
Okay, that's good

**Dylan Patel** [51:07]
... maybe that's more natural and fun.

**Brandon Anderson** [51:09]
Yeah, like a mix of things. Great.

**Dylan Patel** [51:10]
Yeah.

**Brandon Anderson** [51:10]
And this is yours. Okay, let's try yours first, and then we'll tr- well, let's try the restaurant actually first and then-

**Dylan Patel** [51:14]
Control, control.

**Brandon Anderson** [51:15]
Yeah, control. Okay, cheers. Real cool.

**Dylan Patel** [51:18]
I always feel like you gotta cheers the k- this food.

**Brandon Anderson** [51:22]
Mm. That was, that was good.

**Dylan Patel** [51:23]
Mm-hmm.

**Brandon Anderson** [51:24]
Okay. And we'll try yours now.

**Dylan Patel** [51:27]
Is that okay?

**Brandon Anderson** [51:27]
A lot of meat. I love it.

I'll try this as well.

**Dylan Patel** [51:32]
Cheers.

**Brandon Anderson** [51:36]
Mm. Yeah, I like that.

**Dylan Patel** [51:36]
I like mine more.

**Brandon Anderson** [51:37]
Yeah. No, it's very... It's just, like, a lot stronger. Like, it's very deep. Mine's probably gonna be bland compared to that.

**Dylan Patel** [51:47]
So I, I, I think, um, you know, there's this, like, debate in the world, right? French people, they don't season their food so much.

**Brandon Anderson** [51:55]
Mm-hmm.

**Dylan Patel** [51:56]
They're all about the ingredients shining. Then you got like, you know, equator, uh, equatorial people, right, slop, like Indian food and, like-

**Brandon Anderson** [52:03]
Slop

**Dylan Patel** [52:03]
... Caribbean food and, like, Southeast Asian food, just like throw on the spice, throw on the sugar, throw on the, like, everything. And it's like, okay, well, then is this... A- and sort of like the, the elitist French would say, "That's 'cause your ingredients suck.

You have to throw all-" "... the slop on there," right?

**Brandon Anderson** [52:17]
Overdo it.

**Dylan Patel** [52:18]
Yeah, but I don't know. I'm, I'm, I'm a, I'm a slop son, you know?

**Brandon Anderson** [52:20]
Yeah. I mean, it is very tasty, so let's try this. The palates are...

Yeah, I think compared to yours, it's, like, a lot blander.

It's very light.

**Dylan Patel** [52:35]
Not gonna lie.

**Brandon Anderson** [52:37]
You like your, yours better, right?

**Dylan Patel** [52:39]
Yeah.

**Brandon Anderson** [52:39]
Yeah, I could tell. Honestly, yours is like-

**Dylan Patel** [52:42]
So there's a, there's a mess-up of technique, right? The bottom of my, uh, pan was getting burnt.

**Brandon Anderson** [52:46]
Mm-hmm.

**Dylan Patel** [52:47]
And you know, we can't get a wok hei because this is induction cooktop. But because the bottom was getting burnt, the smoke flavor I think was imparting into the rice. And I got a wok hei. It was intentional.

**Brandon Anderson** [52:57]
Yeah. Yeah, you can kinda taste it here, but it's, like, hard. Very good. Okay, are there any call-outs, things that you want people to know? Are you hiring, um-

**Dylan Patel** [53:07]
Mm

**Brandon Anderson** [53:07]
... clients?

**Dylan Patel** [53:09]
Yeah, I am, I am hiring. Um, we're at 60 people now. Um, we work with all the top companies in the world, major AI labs, major hyperscalers-

**Brandon Anderson** [53:17]
Mm-hmm

**Dylan Patel** [53:18]
... um, semiconductor companies, data center companies, et cetera. We cover the entire swath from, um, AI infrastructure, AI models, uh, tokenomics, right, usage of AI models, um, where, who, who's using them, what are they using them for, what's the cost of running it, all these sorts of things.

**Brandon Anderson** [53:31]
Yeah.

**Dylan Patel** [53:32]
So that's the area we're really expanding into this year, um, and last year, and so I think that's the audience that also matches sort of this, this audience that you have here.

**Brandon Anderson** [53:39]
Mm-hmm.

**Dylan Patel** [53:39]
So if anyone wants to track those things, right, usage of AI, the person who was working on this before, unfortunately for them but also a call-out, they got hired by Anthropic, right? You know, so sort of like, you know, the person who was working on this got c- got hired by Anthropic-

**Brandon Anderson** [53:51]
Yeah

**Dylan Patel** [53:51]
... as a, as a sort of like-

**Brandon Anderson** [53:52]
They're thriving

**Dylan Patel** [53:52]
... you know, please don't poach any more of my people. Um, but- ... I think it's a, it's a good pathway, right?

**Brandon Anderson** [53:57]
Yeah.

**Dylan Patel** [53:57]
It's like it's, it's showing-

**Brandon Anderson** [53:58]
Very much so

**Dylan Patel** [53:58]
... your work is public. You know, if you're gonna kill it, you're gonna kill it. You have all the resources behind you of knowing and understanding infrastructure.

**Brandon Anderson** [54:03]
Yeah.

**Dylan Patel** [54:03]
I think that's the big call-out is, like, we're hiring for that role. We pay well. We have he- healthcare. We hire globally. We're in eight or 10 different countries like US, Japan, Taiwan, Singapore, France, Germany-

**Brandon Anderson** [54:12]
Wow, global, yeah

**Dylan Patel** [54:12]
... Israel, yeah, Canada, um, UK, right? So we're everywhere. Um, yeah, I think that's, that's, that's the allure. We get to work with, uh, all the coolest people.

**Brandon Anderson** [54:20]
Great. That's amazing. Well, thank you so much, Dylan.

**Dylan Patel** [54:22]
Yeah.

**Brandon Anderson** [54:22]
It was a great time.

**Dylan Patel** [54:23]
Well, thank you for having me. Yes.

**Brandon Anderson** [54:24]
Yes.

**Host** [54:27]
I really care about the quality of the chicken. Okay, this is like chicken egg fried rice, right? Yeah.

**Dylan Patel** [54:33]
Yeah.

**Host** [54:33]
Oh. Lydia, what's your... How do you assess? Hm.

**Dylan Patel** [54:37]
I think mine's is a little intense.

**Host** [54:39]
Yours is intense. There's probably too much soy sauce. It-

**Dylan Patel** [54:42]
That's fair

**Host** [54:42]
... it's really good.

**Dylan Patel** [54:42]
I forgot to throw it in, and then I just threw it all in.

**Host** [54:45]
Ah, YOLO'd. How much did you have? See, he ha- he left, like-

**Dylan Patel** [54:49]
He little left a little bit, yeah.

**Host** [54:50]
Yeah. I would say, you know, like I, I would take th- I would take Alan's. I would take-

**Dylan Patel** [54:55]
I would, I would also take his

**Host** [54:56]
... except there's soy sauce for me. I don't know, but mm-hmm. I like Dylan's. Okay.

**Dylan Patel** [54:59]
Mm.

**Host** [54:59]
Yeah.

**Brandon Anderson** [55:00]
It has a lot of flavor.

**Host** [55:00]
I like salty. There's one each. Brandon, you're the tiebreaker.

**Brandon Anderson** [55:03]
I'm on the side of the slop.

**Dylan Patel** [55:04]
Yes.

**Host** [55:05]
Yes. There we go.

**Dylan Patel** [55:06]
Wait, what do you vote? What do you vote?

**Brandon Anderson** [55:07]
He voted for you.

**Host** [55:08]
Oh, I voted for you too.

**Brandon Anderson** [55:08]
Yeah, you tied with me 1 to 1.

**Dylan Patel** [55:09]
Wait, I thought you...

**Brandon Anderson** [55:10]
Yeah.

**Dylan Patel** [55:10]
You worked at a restaurant, bro.

---

This library is powered by PodHood (https://podhood.com), the podcast website platform.