OpenAI claims AI solved a $1M math problem, sparking academic dispute

OpenAI says its internal AI solved a 90-year-old math problem, but mathematicians are pushing back on the lab's rush to take credit.

OpenAI published a writeup and formal proof in Lean—a math verification language—claiming to resolve the Navier–Stokes existence and smoothness problem. The proof shows that fluid motion equations can break down in finite time as a spinning vortex spirals inward and stretches out. To find the answer, OpenAI ran a cluster of roughly 10,000 concurrent AI agents powered by an unreleased model stronger than GPT-6 Astra.

Why it matters: If the proof holds up, it proves AI agents can crack hard problems in pure mathematics that stumped humans for decades. But the announcement triggered an immediate academic dispute, with researchers questioning whether OpenAI rushed to claim credit before the math community could properly vet the work.

Know this: The Navier–Stokes equations govern how liquids and gases move, from blood flow to air over wings. OpenAI's proof demonstrates a "singularity"—a point where fluid speed hits infinity and the standard equations break down.

Solving fluid dynamics with 10,000 AI agents is impressive, but surviving math community peer review might be the real test.