California (Web Desk): OpenAI has claimed a major breakthrough in mathematics, saying its advanced artificial intelligence system has produced a solution to the 90-year-old Navier–Stokes problem in just 88 hours.
According to the company, around 10,000 AI agents were deployed to work on the highly complex mathematical challenge.
The agents were divided into different groups and equipped with the ability to access online information and run computer code as they explored possible approaches to the problem.
OpenAI said the AI system began working on the challenge on September 1 and produced what it describes as a solution after approximately 88 hours of continuous computational work, finishing on September 5.
The Navier–Stokes problem deals with mathematical equations used to describe the movement of fluids, including the flow of air and water. It is considered one of the most difficult unsolved problems in modern mathematics.
The Clay Mathematics Institute included the problem among its seven Millennium Prize Problems in 2000. A prize of $1 million has been offered for a valid solution to each of these longstanding mathematical challenges.
Despite OpenAI’s claim, questions have already been raised about the validity and originality of the proposed solution.
New York University professor Tristan Buckmaster has expressed concerns, saying he and mathematician Levant Alpoge, associated with rival AI company Anthropic, had already been researching the same problem.
Buckmaster reportedly noted similarities between OpenAI’s approach and their private research and called for further scrutiny over how the company arrived at its result.
OpenAI has rejected the suggestion that it used anyone’s private research or personal data. The company said it began working on the problem independently after hearing reports about mathematicians conducting research in the area.
However, it acknowledged that it could not completely rule out the possibility that general user data may have indirectly contributed to improvements in its models.
So far, the Clay Mathematics Institute has not officially verified OpenAI’s proposed solution or publicly confirmed that the long-standing problem has been solved.
If the claimed proof is independently examined and ultimately accepted by the mathematical community, it could represent a significant milestone for artificial intelligence and scientific research.
For now, however, the claim remains subject to expert review and formal mathematical verification.