Smithsonian Magazine · People & society

AI May Have Solved Navier-Stokes Problem, Sparking Credit Controversy

OpenAI announced that an internal AI system has potentially solved the Navier-Stokes problem, a complex fluid dynamics equation that has puzzled mathematicians for nearly 200 years and carries a $1 million prize.

The AI system, comprising around 10,000 agents, worked autonomously for 88 hours on the problem. This achievement highlights AI's growing capability in tackling advanced mathematical challenges.

The Navier-Stokes problem, one of seven Millennium Prize Problems, questions whether the equations describing fluid motion always yield smooth solutions or if 'blowup' scenarios exist where solutions become infinitely fast.

OpenAI claims its AI identified a 'blowup' scenario involving a spiraling vortex, verified using the Lean programming language. While theoretical, the underlying equations are crucial for applications like aircraft design and climate modeling.

The potential solution has ignited controversy due to an alleged association with the work of two human researchers, one employed by OpenAI's rival Anthropic. These researchers had also made significant progress on the problem using large language models.

One of the researchers, Tristan Buckmaster, questioned whether their work, including drafts shared in OpenAI's Codex, might have been accessed or used by the AI system. OpenAI denies accessing specific user data but acknowledges the possibility of de-identified data improving models.

OpenAI has stated categorically that user prompts could not have influenced the system. The company will not claim the $1 million prize, and the ultimate credit for the solution remains uncertain amidst these claims and counterclaims.

AI-samenvatting op basis van de bron.