SemiAnalysis
Technology research and analysis firm specializing in semiconductor and AI infrastructure.
- VC’s paper gains look wild
- 2 Companies Will Control Most of the World's Compute by 2028. Dylan Patel Did the Math.
- Meta Agrees to Pay Billions to Settle Claims It Harmed Children
- Gavin Baker - AI Market Jitters - [Invest Like the Best, EP.485]
- Surgical robots break out
- Dylan Patel, SemiAnalysis, Nebius, Glean, Legora.. 12 Hot Takes From The Biggest Names in AI
- Anthropic’s Safety Superpower (Stratechery Article 6-15-2026)
- The U.S. Orders Anthropic to Shut Off Access to Mythos
- Anthropic Urges Its Competitors to Hit the Pause Button
- Reiner Pope – Chip design from the bottom up
- Amazon’s Durability (Stratechery Article 5-5-2026)
“Stripe is the foundation of Passport, the access management software that undergirds Stratechery, SemiAnalysis' institutional offerings, and The Ankler.”
Source→“About 96% of all the tokens that go through these entire sessions are cached”
Source→“Positron AI raised an $875 million Series C led by NEA, Atreides Management, Valor Equity Partners, Andra Capital, SemiAnalysis Capital and Jim Clark. The funding values the company at $5 billion.”
Source→“GB300... deliver 3x to 5x more performance per watt than the generation before”
Source→“OpenAI could potentially start turning a profit even with the big rise of Codex and 5.6.”
Source→“Anthropic started turning a profit in Q2... if we go back a year ago, everything that they, all the money they had was venture funded losses.”
Source→“By the time you're towards the end of 2028, if this trend continues, which I see nothing that's stopping it, you've got them just controlling most of the usable flops in the world on their own.”
Source→“Jane Street reportedly extracts more value per megawatt than Anthropic does.”
Source→“Cerebras bet against the GPU. It just IPO'd at $56B.”
Source→“All 4 now spend everything they generate on capex and still raise debt on top of it.”
Source→“OpenAI and Anthropic could control most of the world's usable AI compute by the end of 2028, as their superior ability to monetize models lets them outbid rivals for scarce capacity.”
Source→“Our friend Dylan Patel at his company, he's an ASI maxi, but he's at 30%. That's probably the highest one I've heard. I've actually heard of 50. And there's $25 trillion in knowledge work. Let's take your 20% number. That's $5 trillion.”
Source→“SemiAnalysis founder Dylan Patel is targeting $400 million for a new venture vehicle titled SemiAnalysis Capital Fund I.”
Source→“Semi-analysis, for instance, leader in many respects. They've done excellent work on everything up to the power source... They do everything up to the point where they don't assess where the gas will come from.”
Source→“There's multiple Chinese model labs who are telling all the inference guys, hey, our next model is not going to be open source. We're going to license it to you. So open is dying quickly, unfortunately.”
Source→“We go through every single permit and regulatory filing around every single data center with AI. And it's like, and we take satellite photos of every data center and we like, we're able to label our data set and then recognize what generators people are using, what cooling towers and the construction progress and substation. All this stuff is like automated and it's only possible because of Gen AI.”
Source→“There are rumors that semi-analysis recently passed a hundred million of revenue... We have 90 people and a big chunk of them are technologists, engineers across the whole supply chain. And then a big chunk is people who are formerly at hedge funds.”
Source→“We got CoreWeave and Crusoe and Nebius and Oracle and Microsoft and Amazon and Google and OpenAI to contribute to us compute... We've got over $50 million of hardware donated to us. Once we launched TPUs and Tranium, it would actually be over $100 million of hardware... Maybe about 15 different chip types, all running these benchmarks every single day on all the latest models.”
Source→“SemiAnalysis recently estimated that a $200 plan gets you $8,000 worth of Claude tokens and $14,000 worth of Codex tokens. Of course both are fighting for user and developer mindshare, but they're also fighting to have access to actual usage data to make their models better.”
Source→“SemiAnalysis says the $200-a-month Claude Max and ChatGPT Pro plans are far more generous than commonly assumed. Based on API pricing, it turns out they offer roughly $8,000 and $14,000 a month worth of tokens, respectively.”
Source→“SemiAnalysis estimates 'orbital compute costs more than 4x terrestrial compute in 2026 and is unlikely to reach cost parity until around 2040 unless Earth-based power and data-center bottlenecks become more severe.'”
Source→“Cited as one of only five gold-tier providers in Semi Analysis's ranking of ~100 GPU clouds. Gold-tier providers had 5-15% lower total cost of ownership even at identical GPU pricing.”
Source→“SemiAnalysis: 'AWS will lose the future of computing' — Thompson argues this analysis was correct but time-bound, describing a training-dominated world that has since been superseded.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.