Overview
- OpenAI said an unreleased internal system used about 10,000 AI agents over roughly 88 hours and millions of dollars of compute to produce a proposed solution and a Lean formalization of the Navier‑Stokes problem.
- New York University mathematician Tristan Buckmaster publicly accused OpenAI of following his unpublished approach and asked whether his Codex sessions helped train the model.
- OpenAI denied that researchers or agents saw specific user drafts but said it cannot fully rule out that de‑identified product usage informed model improvements.
- The company withdrew sponsorship of Caltech’s Mathathon after organized pushback from academics who say the announcement shifted verification work onto unpaid mathematicians.
- Independent community vetting, journal review and Clay Mathematics Institute procedures remain the central paths to prize eligibility and lasting acceptance, and mathematicians warn the episode could change norms for sharing, authorship and data governance.