Fields Medalist Hugo Duminil-Copin: Frontier Models Already Outperform Me at Many Math Tasks
miniapeur · x · 2026-09-02
Fields Medalist Hugo Duminil-Copin (Université de Genève / IHES) published an essay on how AI is directly affecting the mathematical community.
- He states plainly that frontier models are far better than he is at many mathematical tasks, with encyclopedic command of his field while he sometimes can't recall his own proofs
- Personal experience: in April, ChatGPT Pro proved the difficult percolation problem his colleagues privately called pc<1 — a problem they were actively working on; he says it's now only a matter of time before the field's most famous conjecture, θ(pc)=0, falls too
- Central tension: problems with urgent applications clearly benefit, but whether mathematics should welcome the flood of AI-generated proofs is a legitimate, pressing question
- He deliberately skips predicting AI's future and notes mathematicians have a role in AI safety research
Related event: Fields Medalist Reflects on What Mathematicians Lose by Using AI(3 posts)→
More from AGI Musings
- Anthropic borrows lawyers' ethics for Claude agents that closed 186 deals with no human approval — rohanpaul_ai · 2026-10-18
- Researcher argues we may solve AI-human alignment before humans align with each other — 3scorciav · 2026-10-18
- EigenGender: private-industry coup isn't the default, but worth guarding against — EigenGender · 2026-10-18
- Writing instructor's meltdown: "LLMs are literally just spicy autocomplete" — Decent-Decision-9028 · 2026-10-18
- When will a machine win a Nobel Prize? — rand_longevity · 2026-10-18
- If humans had a chain of thought, most would fail alignment training — robleclerc · 2026-10-18