I did live in Minnesota briefly after college. Um I'm from rural Georgia. Um yeah, we uh I did beekeep for like a year and a half basically. I feel like I've just kind of sort of went through a lot of life uh just next step, next step, next step. Doesn't seem like there's a clear immediate path. Looking back I can spin a narrative like oh obviously I wouldn't be doing this because my interest when I was eight was this and my interest when I was 12 was this but uh you know like moderating forums related to chips but you know I thought it was just like a serendipitous thing. Eventually sort of everything culminated in 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. 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 posting on WordPress, right? Like do it non-anonymously. Do it on Substack and start charging for it." A few years go by and then Doug joins the company. It was really a great moment because otherwise it might still just be like a niche anonymous blog where I'm still just doing random consulting rather than actually a company with 60 people.

Endgame scenarios are kind of insane. There's a variety of things that could happen. Status quo is the best, no war happens, China continues to industrialize itself with the industry that Taiwan has but there's no major event. Option two, which is what a lot of people think is the best, is Taiwan actually more stringently claims it's independent. That seems like a poor option just because that potentially causes China to move much more aggressively. There's two parties in Taiwan: the DPP and KMT. If the DPP wins, they're more pro-US, anti-China, more Taiwan independence. What I think is most likely is that there is some sort of political coup or action that destabilizes Taiwan but doesn't necessarily lead to a full-scale invasion. This is the best of both worlds for China; they don't have to enter an invasion but get to continually creep on Taiwan. The "galaxy brain" thing for an American to want is for the pro-US party to lose. You want KMT to win so China will be placated more, but it's not like TSMC starts disobeying American export restrictions because Taiwan utilizes American banking systems and equipment industries. China's placated by a friendly government, and yet China doesn't actually have any of the chips and the US continues to get access.

It'd be a travesty if a lot of Chinese researchers left American labs. Half or at least a third of researchers at labs are Chinese. At the same time, this is the greatest technology to ever fall into humanity's hands. Americans think they should control it, whether the moral justification is that we're the good guys or AI is a great weapon, so let's make sure we're the only ones with it. Where does that control start and stop? One could say let's just control the chips or the AI, but if you let them buy the chips, they're able to do everything they want anyway. China has great engineering; you only remove one piece of the puzzle, they're able to re-engineer that last bit. China's not that far behind. You look at agent swarms and it's not worse than Codex by a large amount. There is an argument that the current regime of export controls doesn't have the US leading in AI by enough. One could either antagonize China more, ban more things—does that risk a Taiwan invasion or alienate researchers? We only have a few years until the capabilities of AI accelerate GDP growth.

This year Google is spending $180 billion, Amazon spending $200 billion on AI infrastructure. This is 4x what they were doing just not too long ago. If they get returns of any degree, we're talking about trillions of dollars of economic value being added in the next handful of years. Right now the AI industry maybe does 50 billion of revenue, but we're seeing it explode. Anthropic is adding two or three billion of revenue a month now. We're in the takeoff period. Adoption has been so accelerated. Claude Code adoption went from 2% to 4% of GitHub commits in one month. We're at the stage where these are trillions of dollars of economic value. Worldwide software developer wages are $2 trillion; you end up with a pretty incredible amount of spend that could happen. These companies see insane amounts of demand. Anthropic added like $1.5 billion of compute in one month just to serve users. You extrapolate that out and they need hundreds of billions of dollars of compute. The bet here is that growth will continue to accelerate.

Nvidia embodies Silicon Valley spirit more than maybe any other company. Andy Grove's "Only the paranoid survive"—Jensen Huang is one of the most paranoid people in the industry. The moment he sniffed wind of the OpenAI Cerebrus deal, he updates his priors and changes course. As we look to Reuben and beyond, Jensen is starting to fully embrace heterogeneity. Much like this fried rice, there's no one individual ingredient that shines above all. He's got the CPX chip for context processing, not good at latency sensitive apps, but he's got his main line of GPUs and Groq chips. Nvidia knows they will lose because they have a business model deficit; Google and Amazon get to vertically integrate which saves tons of money. He has to be better by a ton to justify his margins. Nvidia will remain on top this year and next year, but moats are as shallow as they've ever been because things are moving so fast and the size of the numbers—hundreds of billions—allows anyone to justify hiring any talent.

What's the biggest bottleneck to speed? In 2023 it was definitely CoWoS (Chip on Wafer on Substrate). In 24, 25, it started to become data centers. Energy is a bigger deal in 25, 26. But the big bottleneck is now back again to semiconductors. Semiconductors are extremely cyclical. The buildings that chips are made in are the most complicated buildings people make; they have multi-year timelines. It requires all the complexity of electricians and plumbers but with chemicals and precursors through the fab. People have just not built enough fabs. These tools cost hundreds of millions of dollars. Google would buy a lot more TPUs but they can't ramp production fast enough, so they have to buy tons of GPUs. No one is getting enough capacity of semiconductors. Semis are the shortage, and the bottleneck to building more chips is more fabs. That's going to persist through the end of the decade.