OpenAI’s September 2026 announcement that an internal model directed 10,000 autonomous AI agents to produce and Lean-formalize a finite-time singularity proof for the Navier-Stokes equations has accelerated trader focus on whether another lab will publicly claim a second Millennium Prize solution before year-end. The effort, estimated at millions in compute and building on concurrent human-AI work by Anthropic’s Levent Alpöge and NYU’s Tristan Buckmaster, demonstrated scalable agent swarms tackling long-standing conjectures while highlighting competitive dynamics among labs racing to demonstrate advanced reasoning benchmarks. Subsequent October reports of OpenAI progress on Yang-Mills existence and related problems have reinforced expectations of rapid follow-on results, tempered by ongoing expert review, Clay Institute timelines, and the distinction between partial claims and full prize-eligible resolutions.