Mathematics: Proofs, Prizes and AI-Assisted Results — 2026-09-08
OpenAI has claimed a breakthrough in solving the Navier-Stokes Millennium Prize Problem, triggering immediate controversy over credit and verification. Simultaneously, Anthropic announced a fully computer-verified formalization of Fermat's Last Theorem, marking a milestone in AI-assisted proof checking.
Mathematics: Proofs, Prizes and AI-Assisted Results — 2026-09-08
Top developments
OpenAI Claims Navier-Stokes Millennium Problem Solved
On September 8, 2026, OpenAI announced that its AI systems may have solved the Navier-Stokes existence and smoothness problem, one of the seven Millennium Prize Problems carrying a $1 million award. The company reports that 10,000 AI agents worked in parallel for 88 hours to demonstrate that fluid flow equations can sometimes "blow up," challenging long-held assumptions in fluid dynamics. This claim has sparked intense debate within the mathematical community regarding the validity of the proof and the appropriate attribution of such a fundamental result.
Controversy Over Credit and Prior Work
The announcement was immediately overshadowed by accusations from independent mathematicians who claim they had independently arrived at similar results or were aware of OpenAI’s internal work prior to the public release. Tristan Buckmaster of New York University reported that OpenAI researchers approached him regarding his own work on the problem, raising questions about whether the company "raced" to publish after learning of external progress. A separate report highlights a clash where a mathematician asserts he posted a valid proof weeks earlier, seeking proper credit as journals review the submissions.
Anthropic Formalizes Fermat’s Last Theorem
In a separate development announced around September 5–7, Anthropic stated that its Claude AI model successfully generated a fully computer-verifiable proof of Fermat’s Last Theorem. The process involved translating the theorem into approximately 13 million lines of code that could be checked by a computer without human trust, completing in 11 days what was expected to take years of manual effort. German tech outlet heise online reported that this represents the first fully machine-checked version of the theorem, highlighting the potential for AI to accelerate formal verification in mathematics.
Skepticism and Verification Challenges
Despite the claims, prominent mathematicians and outlets like Scientific American and WIRED urge caution, noting that the underlying proofs have not yet been fully public or peer-reviewed in the traditional sense. The New Scientist reported that the result emerged amid rumors and less complete AI-generated work announced just hours before, complicating the narrative of a singular breakthrough. The mathematical community is now closely reading the technical paper released by OpenAI to determine if the "blow-up" solution meets the rigorous standards required for the Clay Mathematics Institute prize.
Local view
German media has provided detailed coverage of both major AI announcements. Drweb.de analyzed the skepticism surrounding OpenAI’s Navier-Stokes claim, noting that experts are doubting the completeness of the result despite the high-profile announcement. Meanwhile, Spektrum der Wissenschaft highlighted the efficiency of Anthropic’s Fermat’s Last Theorem formalization, emphasizing how AI reduced a multi-year formalization project to just 11 days. French outlets like Pour la Science continue to focus on the recent Fields Medal winners (July 2026), with no new major local-language updates on the September AI proofs in the top results.
Context & numbers
- Prize Value: The Navier-Stokes problem carries a $1 million prize from the Clay Mathematics Institute.
- Computational Scale: OpenAI utilized up to 10,000 AI agents working in parallel for 88 hours to achieve the claimed result.
- Formalization Size: Anthropic’s Claude model produced approximately 13 million lines of code to verify Fermat’s Last Theorem.
- Recent Awards: The 2026 Fields Medals were awarded in July to Hong Wang, Yu Deng, John Pardon, and Jacob Tsimerman, with Hong Wang noted as the third woman to receive the honor.
On the radar
- Peer Review: Watch for peer-reviewed validation of OpenAI’s technical paper, which is currently being closely read by mathematicians worldwide.
- Clay Institute Response: The Clay Mathematics Institute has not yet issued an official statement regarding the eligibility of an AI-generated proof for the Millennium Prize.
- Oberwolfach Prize: The Oberwolfach Prize 2025 will be formally presented to Yu Deng on September 24, 2026, providing another opportunity for commentary on recent mathematical achievements.
This content was collected, curated, and summarized entirely by AI — including how and what to gather. It may contain inaccuracies. Crew does not guarantee the accuracy of any information presented here. Always verify facts on your own before acting on them. Crew assumes no legal liability for any consequences arising from reliance on this content.
Powered by
