OpenAI claims 10,000 agents solved Navier-Stokes Millennium Prize problem in 88 hours
OpenAI announced that an unreleased AI model, driven by 10,000 autonomous agents, solved the Navier-Stokes Millennium Prize Problem in 88 hours. The breakthrough, which demonstrates that fluid dynamics equations can break down over time, required millions of dollars in compute resources—roughly 1,000 times the cost of previous mathematical AI efforts. Mark Chen, OpenAI’s chief research officer,…
Key points
- OpenAI used 10,000 AI agents and millions in compute to solve the Navier-Stokes problem in 88 hours.
- Mathematician Tristan Buckmaster accused OpenAI of copying his approach, though OpenAI denied using his specific prompts.
- OpenAI will not claim the $1 million prize, as the Clay Institute requires peer review and a two-year validation period.
The announcement triggered immediate controversy with mathematician Tristan Buckmaster and Anthropic researcher Levent Alpoge, who had recently published AI-assisted work on related equations. Buckmaster alleged that OpenAI adopted their specific approach after rumors of their progress spread, though OpenAI denied using their prompts or proofs. While OpenAI stated it would not claim the $1 million prize, the Clay Mathematics Institute emphasized that rigorous peer review and a two-year validation period are required before any solution is officially recognized.
This incident highlights the accelerating role of large-scale agent swarms in high-level mathematics, raising debates about the nature of machine-generated understanding versus human insight. The dispute over credit underscores the competitive tension between major AI labs and the academic community as AI systems begin to address problems that have remained unsolved for decades.
OpenAI says 10,000 AI agents cracked one of math's hardest problems in 88 hours
phys.org · 9 September 2026
Loading the full article…
This text was published by phys.org and written by Alex PIGMAN. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
18sources- Hacker News discussion · 228 pointsnews.ycombinator.com
- Hacker News discussion · 166 pointsnews.ycombinator.com
- Reddit discussionreddit.com
- Last Week in AI #344 - Navier–Stokes, Pacing the Frontier, AI MisuseNewsletter · Last Week in AI ·
- OpenAI maths bombshell sparks debate about who gets credit in age of AIPress · nature.com ·
- AI companies must work with the research community to protect attributionPrimary source · Nature Machine Learning ·
- Scott Aaronson says that labs, "having been burned by the hostile response to the Navier-Stokes proof, are now sitting on solutions to some very major problems until they figure out a better way to handle things"Press · scottaaronson.blog ·
- OpenAI just wants to winPress · The Verge AI ·
- ‘Immature playground boasting’: Mathematicians uneasy at OpenAI’s latest scalpPress · The Guardian AI ·
- OpenAI Claims Another Huge Mathematical Result Amid Fights Over Credit, Ethics, and PrivacyPress · singularityhub.com ·
- OpenAI’s Navier-Stokes release included a Lean 4 formal proofPress · johndcook.com ·
- OpenAI’s sly mathematical breakthrough sends a chill through academiaPress · The Verge AI ·
- OpenAI's secret model settles a $1M math problemNewsletter · The Rundown AI ·
- OpenAI claims to have solved the 90-year-old Navier-Stokes math problem in 88 hoursPress · CNBC Technology ·
- [AINews] OpenAI reports Navier-Stokes singularity find in 88 hours using Astra-next, roughly 10,000 agents and 130B tokens (>$40M), a contender for second ever Millennium Prize awardedNewsletter · Latent Space ·
- What OpenAI’s latest controversy tells us about the future of mathPress · MIT Technology Review AI ·
- Some thoughts on the Navier–Stokes Millennium Prize ProblemNewsletter · Simon Willison ·
- OpenAI claims to have solved maths problem that stumped humans for decadesPress · The Guardian AI ·
- OpenAI says it cracked 90-year-old maths problem in 88 hoursPress · BBC Technology ·
- OpenAI claims huge maths breakthrough on a famed ‘Millennium Problem’Primary source · Nature Machine Learning ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- MIT researchers show neural networks can build own learning rules · 1 src
- Anthropic research suggests small AI models may self-replicate on IoT devices · 6 src
- Anthropic co-founder predicts full RSI possible by 2027 · 1 src
- Sakana AI’s MASS method improves model self-judgment but risks baking errors into training · 1 src
- Microsoft Research finds clearing chat history boosts AI agent performance by 66% · 1 src
Comments
via GitHub Discussions