AI
OpenAI's Navier-Stokes Breakthrough Draws Accusations of Academic Misconduct
The AI company claims an AI model solved a 200-year-old fluid dynamics equation, but another researcher alleges it rushed into the work after learning of competing progress.
WIRED — AI
OpenAI announced it has used artificial intelligence to produce a solution to the Navier-Stokes equation, a foundational 200-year-old problem in physics describing how fluids behave. The breakthrough suggests AI can tackle some of mathematics' hardest remaining challenges—yet it immediately sparked allegations from the academic community that the company acted unethically in its pursuit.
The Breakthrough#
According to Sebastien Bubeck, a mathematician and AI researcher at OpenAI, the company began work on the problem on August 28. The effort culminated in what it claims is a formally verified solution, written in Lean, a formal language for mathematical proofs. OpenAI deployed over a thousand AI agents working for more than 50 hours to complete the task, consuming computational resources valued in the millions of dollars.
The Navier-Stokes equation is one of the world's most famous unsolved problems. Cracking it would earn a solver one of the $1 million Clay Millennium Prize awards.
The Allegations#
The triumph, however, was immediately clouded by accusations from Tristan Buckmaster, a mathematician at New York University, and Levent Alpöge at Anthropic. According to Buckmaster's account, OpenAI learned of his and Alpöge's independent progress on a related problem last week and then rapidly escalated its own efforts. He claims OpenAI leadership offered him proposals that would have allowed the company to announce the solution without crediting Alpöge, and alleges he received evasive answers when questioning whether OpenAI had accessed their computational logs.
OpenAI's Response#
OpenAI executives firmly denied the accusations in a press briefing. Bubeck stated: "We, whether it's the researchers or the agents, did not see any of their work until it was released publicly last night." The company acknowledged Buckmaster and Alpöge's contributions and emphasized that OpenAI's solution was methodologically distinct from their work, though OpenAI did not provide detailed comparisons.
Why It Matters#
As AI systems take on increasingly sophisticated mathematical research tasks, questions about credit attribution, data access, and the conduct of AI labs in competitive environments are likely to multiply. The dispute underscores tensions within the AI research ecosystem as these tools challenge traditional academic norms.
