Fields Medalist: Frontier models exceed humans in math, AI proving conjectures may become the norm
S_Conradi · x · 2026-09-01
Hugo Duminil-Copin, a Fields Medalist, shares a practitioner's perspective on AI in mathematics. He notes that while he avoids long-term predictions, current frontier models are undeniably superior to him in many tasks, possessing encyclopedic knowledge of his field where he sometimes struggles. While AI will benefit problems with direct applications, he questions whether the rapid proliferation of AI-generated proofs is universally welcomed. He recounts colleagues' experience where ChatGPT Pro proved a difficult percolation problem ($pc < 1$) and fears the famous conjecture $\theta(pc) = 0$ may soon fall to AI.
Related event: Fields Medalist Says Frontier Models Already Beat Him at Math Tasks(4 posts)→
More from AGI Musings
- Hinton hits back at Jensen Huang: 'He has good reason to say AI isn't dangerous — he sells Nvidia chips' — BlackHC · 2026-10-16
- Yudkowsky blasts new interpretability finding as primitive; critic pushes back on doomer logic — AlexTensor · 2026-10-16
- Beff Jezos declares 'civilizational inflection point' in the acceleration of technological progress — beffjezos · 2026-10-16
- Data scientist: pivoting to AI engineering is the safest career bet for the next 5 years — mdancho84 · 2026-10-16
- AI will be superhuman at science regardless, and Europe can't opt out — teortaxesTex · 2026-10-16
- beffjezos to Aaron Levie: techno-optimism is hyperstitious — basedjensen · 2026-10-16