OpenAI Claims AI Agent Solved the Navier–Stokes Blow-up Problem

OpenAI announced that its AI agent answered the long-standing Navier–Stokes blow-up problem—the equations describing fluid motion do blow up in certain cases—accompanied by a machine-checkable Lean formalized proof. NYU mathematician Tristan Buckmaster, using a self-paid ChatGPT Codex subscription and collaborating with Anthropic employee Levent Alpöge in their spare time, reached the result about 88 hours before OpenAI's announcement (August 15). The new dynamic of machine proofs outpacing human understanding, combined with a divided mathematical community, marks this as a landmark moment for AI in frontier mathematics.

Confirmed

• OpenAI announced a Navier–Stokes blow-up construction with Lean formalized code claimed to be machine-checkable (relayed by @thione and others).
• Per @rubenhassid: Buckmaster, on a self-paid Codex subscription with Levent Alpöge, solved the problem on August 15; OpenAI announced its solution about 88 hours later. The episode also raised privacy and data-boundary questions.
• PDE scholar Scott Armstrong said the proof is indeed messy and hard to read, but a comprehensible, narratively appealing proof should be extractable; the Lean repo contains more lines than he personally wrote for his last paper (relayed via @RexDouglass).
• The same week, 25 Fields Medalists (including Terence Tao) signed a statement on the serious mismatch between AI and mathematics; on September 3, GPT-6 Astra's FrontierMath performance drew attention (WSJ column by Ben Cohen, relayed by @新智元).

Unconfirmed

• One responder (relayed by Adam Marblestone, responding to Claude Fable's essay After Math) read the paper and Lean code personally and questioned the gap between a machine-checkable logical proof and human-comprehensible mathematics; assessments remain under discussion.
• Armstrong estimated human mathematicians may need weeks to a month to fully digest the proof, and whether the Leray bottleneck is truly broken remains to be verified (relayed by @soumitrashukla9).

Why it matters

• If the blow-up construction holds and is digested, the nearly century-old Leray bottleneck could fall—a major contribution to PDE theory.
• The event showcases proof-before-understanding: machine-generated, formally verified proofs that humans interpret afterward, potentially reshaping how mathematics is organized.
• The divided reactions—from optimism about extracting readable proofs to 25 Fields Medalists warning of AI-math mismatch—signal ongoing norm and trust battles as AI enters frontier mathematics.

2026-09-13 ~ 2026-09-15 · 8 related posts

Full story(20 episodes)→

• Episode 1: Rumors claim GPT-6 cracks Navier-Stokes, credibility doubtful(2026-09-05, 2 posts)
• Episode 2: OpenAI Claims 10,000 Agents Solved Navier-Stokes in 88 Hours(2026-09-09, 13 posts)
• Episode 3: Existence Proof Effect: Rumor of Solvability Drove OpenAI's Navier-Stokes Attempt(2026-09-09, 3 posts)
• Episode 4: Mathematicians Accuse OpenAI of Racing to Claim Their Unfinished Proof(2026-09-09, 6 posts)
• Episode 5: Anthropic Faces Ethics Backlash Over Math Proof and Scooped Collaboration(2026-09-09, 2 posts)
• Episode 6: Rumors swirl that OpenAI nears Hodge and BSD conjectures as Millennium Prize problems fall(2026-09-10, 27 posts)
• Episode 7: OpenAI Claims Navier-Stokes Solved, Progress on a Second Millennium Problem(2026-09-10, 21 posts)
• Episode 8: Mathematician accuses OpenAI of training Astra on his unpublished work; OpenAI denies(2026-09-10, 9 posts)
• Episode 9: Navier–Stokes Authorship Dispute Fuels Debate Over AI Labs' Scooping Culture(2026-09-10, 6 posts)
• Episode 10: OpenAI denies Codex prompts influenced training as "stolen chat data" claims unravel(2026-09-11, 8 posts)
• Episode 11: OpenAI's Millennium Prize Claim Mired in Authorship and Scooping Dispute(2026-09-11, 10 posts)
• Episode 12: AI Circles Speculate Millennium Prize Problems May Fall to AI(2026-09-11, 2 posts)
• Episode 13: Math Community Splits Over AI's Role After OpenAI's Millennium Prize Claim(2026-09-11, 8 posts)
• Episode 14: Rumor: OpenAI Aims Internal Model at Riemann Hypothesis and P vs NP(2026-09-11, 7 posts)
• Episode 15: OpenAI's Navier-Stokes Proof Passes Rechecks but Faces Scrutiny(2026-09-11, 2 posts)
• Episode 16: OpenAI's 10,000-Agent Navier-Stokes Claim Sparks Math Community Credit War(2026-09-12, 18 posts)
• Episode 17: Misread as Anti-AI, Terence Tao Sparks Debate Over the Value of Mathematics(2026-09-12, 18 posts)
• Episode 18: Fields Medalists' Open Letter on AI Sparks Cross-Disciplinary Debate(2026-09-12, 6 posts)
• Episode 19: OpenAI Claims AI Agent Solved the Navier–Stokes Blow-up Problem(2026-09-13, 8 posts)
• Episode 20: OpenAI Claims 10,000 Agents Solved Navier-Stokes Problem in 88 Hours(2026-09-14, 11 posts)

• Mathematician pushes back on OpenAI PDE proof criticism: messy but salvageable — RexDouglass · 2026-09-13
• [source] GPT-6 Hits FrontierMath, Claims Navier-Stokes Proof in 88 Hours; 25 Fields Medalists Push Back — 新智元 · 2026-09-13
• [source] Mathematician cracks Navier–Stokes with Codex, then OpenAI agents did it in 88 hours — rubenhassid · 2026-09-13
• PDE researchers say OpenAI's Navier-Stokes blow-up proof inverts how humans learn math — soumitrashukla9 · 2026-09-14
• OpenAI says its AI agents solved Navier-Stokes blow-up, a Millennium Prize Problem — Chris_Armstrong · 2026-09-14
• Analyst Reads OpenAI's Navier–Stokes Proof and Lean Code to Test the 'Unintelligible Proof' Critique — scottleibrand · 2026-09-14
• [source] OpenAI says internal system solved 90-year-old Navier–Stokes Millennium Problem with a Lean proof — thione · 2026-09-15

1 near-duplicate retellings: Chris_Armstrong