OpenAI's Controversial Claim of Solving Navier-Stokes Problem

OpenAI announced on Tuesday that it has solved the Navier-Stokes problem, one of the Millennium Prize Problems, using an advanced AI model and significant computing resources. This claim, however, has sparked controversy as it came just a day after NYU mathematician Tristan Buckmaster and Anthropic's Levent Alpöge released findings on a related problem.

According to OpenAI, their solution was achieved in 88 hours with the help of 10,000 AI agents. They stated that they began training their model in late August and were inspired to tackle the Navier-Stokes problem after hearing rumors about Buckmaster and Alpöge's work. However, Buckmaster has accused OpenAI of using their research as a foundation without proper credit.

OpenAI has denied these allegations, asserting that no specific user data was accessed in solving the problem. They acknowledged the possibility that de-identified data from users could have contributed to their model's improvement but maintained that they did not utilize Buckmaster and Alpöge's prompts or proofs.

Why it matters

This incident raises significant questions about research ethics in AI and the collaboration between human mathematicians and AI systems. The controversy highlights the challenges of crediting contributions in a rapidly evolving field where AI plays an increasingly central role.

What we know

• OpenAI claims to have solved the Navier-Stokes problem, a Millennium Prize Problem. Confirmed by www.france24.com, www.technologyreview.com, www.theverge.com
• The announcement came a day after Buckmaster and Alpöge published related findings. Confirmed by www.france24.com, www.technologyreview.com, www.theverge.com
• OpenAI stated it will not claim the $1 million prize for solving the problem. Confirmed by www.france24.com, www.technologyreview.com, www.theverge.com
• OpenAI used an internal model with 10,000 AI agents to solve the problem. Confirmed by www.france24.com, www.theverge.com
• Buckmaster accused OpenAI of using his and Alpöge's research without credit. Confirmed by www.france24.com, www.technologyreview.com, www.theverge.com

Where the reports differ

• OpenAI claimed it did not access Buckmaster and Alpöge's work, while Buckmaster suggested they did. www.france24.com, www.technologyreview.com, www.theverge.com
• OpenAI stated that their proofs differ significantly from those of Buckmaster and Alpöge, while Buckmaster argued otherwise. www.theverge.com, www.technologyreview.com

Sources

• www.france24.com — OpenAI Claims Breakthrough in Math Problem Amid Controversy
• www.technologyreview.com — OpenAI's Math Breakthrough Sparks Controversy Over Research Ethics
• www.theverge.com — OpenAI Claims Breakthrough in 90-Year-Old Math Challenge