OpenAI Model Reportedly Solves Field Medalist's Open Problems, Rattling Mathematicians

Around October 13, Hugo Duminil-Copin, Fields Medalist and professor at the University of Geneva, publicly recounted that an OpenAI math model that only came into existence in late August (spending roughly 3 hours of thinking per problem across 4,000 problems total) touched, in a single release, every major open problem he had publicly mentioned in his career; dozens of papers collided with his ongoing research, and the 16 problems he was working on were all claimed solved by OpenAI. He admitted, "Tuesday night I cried, and my crying woke my daughter," said the model "steamrolled my field," and warned that mathematics could slide toward closure.

Confirmed

• Duminil-Copin publicly described being overwhelmed by the OpenAI model, with all 16 in-progress problems claimed proven; his statements carried strongly emotional wording.
• The model reportedly only existed as of late August, running with about 3 hours of thinking per problem across 4,000 problems in total.
• OpenAI released solutions to multiple major open problems in mathematics, which the community has broadly described as "rough and verbose."

Why it matters

• Manoel Horta Ribeiro wrote an analysis of the impact on the mathematical community: PhD student Sophie Maclean said in a video, "I've cried a few times, afraid of losing the mathematics I love." His core argument is that empiricism is sweeping through mathematics, and AI's problem-solving approach strikes at the identity of the mathematical community and the value system of the discipline.
• Theoretical computer science figure mahditcs offered another warning: AI progress itself is not the biggest threat to the math/TCS communities; the real danger is the communities' arrogant and dismissive response to AI, which could make the whole field increasingly irrelevant; mathematics' core business is training the next generation of top mathematicians, and failing at that through arrogance would be the greater loss.

The episode reflects the tension between AI's leap in capability and the career structure of fundamental science, and has already triggered multi-layered responses in the math community, from personal emotion to industry-wide reflection.

2026-10-13 ~ 2026-10-13 · 5 related posts

Full story(19 episodes)→

• Episode 1: Terence Tao says AI math research is moving the wrong way(2026-10-07, 2 posts)
• Episode 2: Tao's "Mathematicians did not ask for this" sparks backlash as OpenAI math manuscripts face scrutiny(2026-10-08, 58 posts)
• Episode 3: Tao's Blog Debates Career Advice for Math Students in the LLM Era(2026-10-09, 2 posts)
• Episode 4: Mathematician Questions Terence Tao's AI Strategy(2026-10-09, 2 posts)
• Episode 5: Terence Tao Has Mixed Feelings About AI-Generated Math Proofs(2026-10-10, 2 posts)
• Episode 6: Terence Tao's "Math 2.0" Talk Sparks Debate on Proof Abundance and Explainability(2026-10-11, 60 posts)
• Episode 7: OpenAI Drops 722 AI-Generated Math Manuscripts, Leaving a Fields Medalist in Tears(2026-10-11, 13 posts)
• Episode 8: Ben Goertzel Slams Math Establishment Over LLM Theorem-Proving(2026-10-11, 3 posts)
• Episode 9: Tao's Thought Experiment Sparks Debate on AI Discoveries Beyond Human Understanding(2026-10-11, 3 posts)
• Episode 10: Terence Tao Calls for More Compute for AI Math, Misread on Reddit(2026-10-11, 2 posts)
• Episode 11: Mathematicians Clash Over Human Math's Relevance in AI Era(2026-10-12, 4 posts)
• Episode 12: Mathematician Says Most Colleagues Are Excited About AI, Not Resistant(2026-10-12, 2 posts)
• Episode 13: Chinese Mathematicians Led by Yau Launch AI-Math Alliance of Eight Institutions and 1,000 Scientists(2026-10-12, 6 posts)
• Episode 14: OpenAI Matrix Multiplication Paper Sparks Debate Over Theoretical vs Practical Value(2026-10-12, 3 posts)
• Episode 15: Reddit User Rebuts Mathematicians' Complaints About AI Solving Math Problems(2026-10-12, 2 posts)
• Episode 16: OpenAI's Math Breakthroughs Raise Questions of AI 'Plagiarism'(2026-10-12, 2 posts)
• Episode 17: OpenAI's 700+ AI math papers spark mathematicians' boycott call(2026-10-12, 4 posts)
• Episode 18: OpenAI Model Reportedly Solves Field Medalist's Open Problems, Rattling Mathematicians(2026-10-13, 5 posts)
• Episode 19: Mathematicians vs AI Boosters: Do Unreadable AI Proofs Halt Math Progress?(2026-10-13, 6 posts)

• TCS Community Warned: AI Isn't the Threat, Our Reaction to It Is — Dr_Atoosa · 2026-10-13
• [source] Empiricism Comes for Mathematics: AI Proofs Break a Centuries-Old Alignment — manoelribeiro · 2026-10-13
• [source] Fields Medalist Duminil-Copin 'paralysed' as OpenAI model claims proofs for all his career's open problems — tegmark · 2026-10-13
• Fields Medalist says OpenAI 'annihilated' his field by solving all 16 of his problems — inductionheads · 2026-10-13

1 near-duplicate retellings: manoelribeiro