OpenAI’s September 8 release of an analytical proof plus publicly verifiable Lean formalization for a finite-time singularity in the Navier-Stokes equations has anchored the 93% market-implied odds against retraction by year-end. Traders see the heavy compute commitment—roughly 10,000 concurrent agents over 88 hours on an internal model exceeding GPT-6 Astra capabilities, plus 130 billion output tokens—as evidence of deliberate internal validation rather than hasty claims. Priority disputes with NYU and Anthropic researchers center on credit and data provenance, not proof invalidity, while the company’s explicit decision not to pursue the Millennium Prize further signals caution. Still, independent mathematician scrutiny or an undetected Lean error could force a public correction if verification fails.