OpenAI Releases 700+ AI-Generated Math Manuscripts, Sparking Academic Dispute

The significance lies less in the volume of manuscripts than in the absence of peer review and problem-level disclosure, which leaves the central claim unverifiable by the standards the academic community applies. That gap explains the dispute: mathematicians are being asked to accept a benchmark-style result without the transparency that normally accompanies a proof. The open question is whether OpenAI submits the work to independent verification or releases the underlying problems and solutions, which would move the debate from assertion to evidence.