The archive.
85 notes from the work. Browse at your own pace.
Google's ScientistTwo was graded by two AI reviewers. Only one of them was held out, and that is the whole result.
8 min readAnthropic's new project coordinator is real. The team-sharing pitch in the video covering it is not, yet.
4 min readThe AI drug 'reversing aging' was never tested for that. A different AI drug already has conditional approval.
6 min readA 40-year-old search algorithm beats modern embeddings at agentic research. Only one version of it does.
9 min readChunking is not dead. The talk that says so smooths one of its own numbers.
6 min read"Discovery intelligence" is a real research term. The video's argument with MIT about it is built on a claim MIT never made.
5 min readA talk says gamify your AI agent like a LitRPG character. A contemporaneous essay proposes something else entirely.
6 min readA Stanford professor says TCP is done for AI clusters. The industry's real money is betting on something else.
7 min readKnowledge agents need a different shape than coding agents. The harder problem is not search.
7 min readTwo open-weight models beat GPT-6 Astra on agent benchmarks without changing a single weight. Here is what actually improved.
7 min readUnion Alpha was never really a mystery model. It's a procurement question wearing a mystery's clothes.
5 min read