OpenAI's September 2026 solution to the Navier-Stokes Millennium Prize Problem using 10,000 coordinated AI agents over 88 hours, followed by its announcement of substantial progress on a second unsolved problem, has anchored trader sentiment at a 60% implied probability for OpenAI. The lab's demonstrated scaling of internal models for Lean-verified proofs and ongoing math output—hundreds of additional results released in early October—outpace Anthropic's contributions, which center on researcher Levent Alpöge's collaborative work on related fluid dynamics but lack comparable standalone breakthroughs. Competitive dynamics, including rumors that prompted OpenAI's initial push, reinforce this edge, though Anthropic's model capabilities and talent pipeline remain swing factors ahead of any formal Clay Institute review or next disclosure.