Recent OpenAI breakthroughs, including an AI multi-agent system solving the Navier-Stokes Millennium Prize Problem in September 2026 and October releases of partial proofs on the Hodge conjecture for CM abelian varieties plus progress toward Birch and Swinnerton-Dyer, have elevated trader consensus that large language models can now tackle these longstanding challenges. The tight race between “no solution by end-2027” at 32.5% implied probability and Hodge at 30.5% reflects uncertainty over full verification, formalization in tools like Lean, and whether compute-intensive agent workflows will deliver complete resolutions before regulatory or timeline hurdles intervene. Birch’s 22.5% share stems from targeted algorithmic advances in elliptic curves, while lower odds on Riemann, Yang-Mills, and P versus NP highlight their greater structural complexity. Upcoming catalysts include further model releases and independent mathematician reviews that could shift sentiment rapidly.