OpenAI’s recent demonstration of advanced AI-driven mathematical reasoning underpins the 78% market-implied odds favoring it over Anthropic. In early September 2026, OpenAI deployed thousands of agents on an unreleased internal model to produce and Lean-verify a solution to the Navier-Stokes existence and smoothness problem—one of the remaining Millennium Prize Problems—within roughly 88 hours plus formalization time. The company simultaneously disclosed substantial progress on at least one additional unsolved problem from the same list. While a concurrent effort by NYU mathematician Tristan Buckmaster and Anthropic researcher Levent Alpöge on related fluid-dynamics questions created brief controversy over data access and priority, the proofs differ and OpenAI’s scale, compute investment, and model capabilities appear decisive. Upcoming catalysts include further public releases on remaining problems and any Clay Institute verification timelines, which could reinforce or challenge the current trader consensus on OpenAI’s lead in frontier AI-assisted mathematics.