25 Fields Medalists: AI labs’ math race is “severely misaligned”

On 11 Sep 2026 twenty-five Fields Medalists published “A Severe Misalignment of AI in Mathematics” at mathandai.org (signatories include Terence Tao, Peter Scholze, Maryna Viazovska, Artur Avila, and others). They argue that LLM progress on hard open problems is real, but labs’ race to claim solutions as marketing benchmarks is misaligned with mathematics’ goals of conceptual insight, careful writeups, attribution, and mentoring. Rushed true/false dumps without isolation of methods or citation raise plagiarism concerns and can scoop human researchers mid-work. The Economist and Hacker News covered the backlash in the wake of OpenAI’s Navier–Stokes claim and related math PR. Primary text: mathandai.org.