OpenAI Execs Push Back on Navier-Stokes Proof Authorship Dispute

OpenAI's release of an AI-generated proof of the Navier-Stokes problem sparked an attribution and authorship dispute with another team (Tristan Buckmaster and Anthropic employee Levent Alpöge). On September 9, OpenAI CEO Sam Altman and VP Sebastien Bubeck each published lengthy responses, saying OpenAI had proactively proposed collaboration and a joint release, and never asked for the other side's names to be removed.

Confirmed

• Sam Altman stated that OpenAI initially believed the other team had also fully solved the problem and hoped for a joint release; upon learning they had only solved the Euler equations, not Navier-Stokes, OpenAI offered to let them publish first, but they refused to coordinate throughout.
• Sebastien Bubeck posted a detailed point-by-point response to the authorship dispute with Levent: the controversial screenshot showed him proactively reaching out to Levent to coordinate paper release timing, which he says was in good faith; he stressed he never asked for Levent to be removed as an author of his own work.
• Noem Brown clarified in a repost that the model's capabilities far exceed today's LLMs and the result should not be credited to any individual prompt.

Unconfirmed

• The two sides tell conflicting stories about the coordination process. Whether the other team actually refused to coordinate and how contributions should be divided currently rest solely on OpenAI's account, with no full public response from the other team yet.

Why it matters

• The dispute touches on norms for authorship and priority in AI-assisted mathematical research: when AI does the core work of a proof, there is no established convention for delineating credit among human collaborators, and this incident may become a reference case for future community discussions.

2026-09-09 ~ 2026-09-09 · 19 related posts

Full story(4 episodes)→

• Episode 1: Navier-Stokes Proof Sparks Data-Leak Controversy, OpenAI Responds(2026-09-08, 67 posts)
• Episode 2: OpenAI Accused of Copying Anthropic Employee's Remarks in Apparent Jab(2026-09-08, 2 posts)
• Episode 3: OpenAI Execs Push Back on Navier-Stokes Proof Authorship Dispute(2026-09-09, 19 posts)
• Episode 4: Rumors say OpenAI solved Navier-Stokes amid priority dispute(2026-09-09, 2 posts)

• [source] OpenAI's Sebastien Bubeck denies authorship demand in Navier-Stokes credit clash with Anthropic — SebastienBubeck · 2026-09-09
• Bubeck: OpenAI began Millennium-problem push after viral rumors Anthropic solved two — SebastienBubeck · 2026-09-09
• [source] Altman responds to Navier-Stokes proof feud: OpenAI offered rival team to go first — sama · 2026-09-09
• Sebastian Bubeck clarifies release-coordination drama: 'I never asked to remove Levent's authorship' — MaziyarPanahi · 2026-09-09
• Altman reveals OpenAI offered rivals first publication after finding they solved Euler but not Navier-Stokes — daniel_mac8 · 2026-09-09
• Sebastien Bubeck responds to Navier-Stokes proof controversy: never asked to strip authorship — jachiam0 · 2026-09-09
• Sebastien Bubeck publicly refutes recent claims on X — Wonderful_Buffalo_32 · 2026-09-09
• Sebastien Bubeck clarifies Navier-Stokes drama: 'I never asked to remove Levent's authorship' — daniel_mac8 · 2026-09-09
• Sebastien Bubeck clarifies math-research controversy: outreach was release coordination, not authorship grab — kimmonismus · 2026-09-09
• [source] Microsoft's Sebastien Bubeck fires back in collaboration spat: 'nothing was locked, you didn't talk to us' — kfountou · 2026-09-09
• Bubeck clarifies credit dispute: OpenAI reached out to HuggingFace only to coordinate releases — basedjensen · 2026-09-09
• OpenAI's Bubeck says the Millennium Problem project began over rumors Anthropic had solved two — basedjensen · 2026-09-09
• Bubeck says team chased Millennium problems after viral rumors Anthropic solved two, denies scooping — anshulkundaje · 2026-09-09
• Bubeck reveals Twitter rumors spurred OpenAI's Navier-Stokes push, proofs differ — burny_tech · 2026-09-09

5 near-duplicate retellings: rohanpaul_ai · rohanpaul_ai · jasondeanlee · basedjensen · littmath