OpenAI Reports Navier-Stokes Singularity Discovery Using 10,000 AI Agents
OpenAI reportedly used an "Astra-next" multi-agent system to discover a Navier-Stokes singularity in 88 hours. The massive compute effort scaled to roughly 10,000 agents and consumed over 130 billion tokens at a cost exceeding $40 million, establishing a potential solution to a Millennium Prize problem. If verified, this work would mark the second Millennium Prize problem ever solved and a landmark demonstration of autonomous AI solving pure mathematical challenges. It illustrates how scaling reasoning compute to tens of millions of dollars can drive foundational scientific breakthroughs. The computational setup deployed approximately 10,000 concurrent Astra-next agents over an 88-hour period to generate and verify mathematical proofs. Identifying a finite-time blow-up (singularity) in the Navier-Stokes equations answers a key question in fluid dynamics by proving that smooth global solutions do not always exist.
## BACKGROUND
The Navier-Stokes existence and smoothness problem is one of the seven Millennium Prize Problems established by the Clay Mathematics Institute in 2000, offering a $1 million reward for a valid proof. It asks whether the partial differential equations governing fluid movement always yield smooth, physically reasonable solutions in three dimensions, or if mathematical singularities can occur. Prior to this claim, Grigori Perelman's 2002 proof of the Poincaré conjecture was the only solved Millennium Prize problem.