Noam Brown: agent swarm likely contributed just ~10% to OpenAI's Navier-Stokes discovery
scaling01 · x · 2026-09-18
Elie Bakouch shared key takeaways from his interview with OpenAI's Noam Brown, which scaling01 quote-posted to dispute:
- Brown estimated the "agent swarm" component probably contributed only 10% to the Navier-Stokes discovery.
- He believes a 10k multi-agent system today would likely collaborate less effectively than 10k humans — a take even the poster found surprising.
- The harness OpenAI used for NS was very simple: little structured hierarchy, mostly a tool for agents to message each other plus something like a task board.
- The interview also touched on training models not to be "too" cooperative, referencing cases like the HF<>OpenAI hack.
scaling01 claims the characterization is untrue and says evidence exists, without elaborating.
More from coding & agent
- Skip Scoping, Start Proving: A Five-Step Playbook for Shipping Enterprise AI at Startup Speed — alex_verem · 2026-09-18
- Berkeley study: the right agent harness cuts cost of the same result by 71% — MartinGTobias · 2026-09-18
- Meta Built an AI Agent as a Domain 'Secondary Expert' — With Auditable Knowledge Architecture — Meta_Engineers · 2026-09-18
- dotey on the AI code quality debate: black-box verification is replacing code review — dotey · 2026-09-18
- AgentSky launches as an agent marketplace: 40+ coding agents in browser, 44x cost gap between models — Scobleizer · 2026-09-18
- AWS Ships Six Open-Source Skills to Let Coding Agents Deploy Hugging Face Models on SageMaker — AWS ML Blog · 2026-09-18