Altman responds to proof dispute: OpenAI offered rivals first release and the prize
zacharynado · x · 2026-09-09
OpenAI CEO Sam Altman posted a lengthy response to the dispute over a recent AI mathematical proof. He said he spent much of the weekend talking with the team involved and that Seb and everyone else acted with integrity and generosity.
Key points: - OpenAI initially believed the other team had also solved the problem and wanted a joint release; - Upon learning the rivals had Euler but not Navier-Stokes, OpenAI offered to let them go first and suggested they take the prize, optionally making Tristan lead author on a rewrite of the OpenAI proof; - They felt it was hard to extend the same offer to Anthropic employee Levent, who was unwilling to talk or coordinate; - Altman stressed OpenAI remained open to other solutions and would have greatly preferred cooperation.
More from Models
- Community urges backing up quantized models amid takedown risk, shares torrent mirrors — Kahvana · 2026-10-25
- Leaked 33B dense model from a dark-horse lab could be the best dense model yet — TheZachMueller · 2026-10-25
- OpenAI and Anthropic week 43: Claude Duet, drug-discovery lead via GPT-Rosalind, life sciences open call — btibor91 · 2026-10-25
- Anthropic tokens on OpenRouter surge ~4x to nearly 3T daily, defying 'bubble peak' talk — scaling01 · 2026-10-25
- Nanbeige4.3-33B open-sourced on Hugging Face, claims to beat Qwen3-32B — Mysterious_Finish543 · 2026-10-25
- ChatGPT Voice leaks internal citation tokens as ship emojis in live output — rubanbhatia · 2026-10-25