GoKawiilOpenAI's Navier-Stokes AI proof draws credit dispute from NYU mathematician
OpenAI announced that an AI model produced a Lean-verified solution to a key case of the 200-year-old Navier-Stokes equation, one of the Clay Millennium Prize problems, after running over a thousand agents for more than 50 hours at a cost of millions of dollars in compute. The announcement came shortly after mathematician Tristan Buckmaster and Anthropic researcher Levent Alpöge posted their own related advances, and Buckmaster has accused OpenAI of rushing to claim credit after learning of their progress.
GoKawiil's interpretation of the reporting above, not reported fact.
The episode highlights both AI's growing capability to tackle unsolved problems in pure mathematics and the messy human dynamics of credit and priority that can emerge when competing labs race toward the same breakthrough. It also raises questions about how AI-assisted mathematical discoveries should be verified, attributed, and compared when multiple groups use different AI tools to reach similar results simultaneously.
- OpenAI says an AI model produced a Lean-formalized solution to a case of the Navier-Stokes equation.
- The effort required massive compute, reportedly costing millions of dollars, using over a thousand agents.
- Mathematician Tristan Buckmaster disputes OpenAI's credit claims, saying the company acted after learning of his and Anthropic researcher Levent Alpöge's own progress.
The Millennium Problems by Keith Devlin — If the Navier-Stokes headlines left you curious what the fuss is about, Devlin's book walks through all seven Clay Millennium Prize problems in plain language, including the fluid equation OpenAI claims to have made progress on. It's a great way to understand why mathematicians care so much about who gets credit for cracking them.
See The Millennium Problems by Keith Devlin on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: wired.com — Will Knight, 2026-09-08
Published there as: “OpenAI Just Claimed a Huge Math Discovery. Some Academics Are Crying Foul”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.