By Benson Daniel
OpenAI has announced what it describes as a major breakthrough in mathematics, claiming that an advanced artificial intelligence system has solved the Navier Stokes problem, one of the most difficult and famous unsolved problems in modern mathematics.
The problem, which has challenged mathematicians for nearly 90 years, concerns equations used to describe the movement of fluids such as water and air. It asks whether smooth solutions to the Navier Stokes equations can always remain smooth or whether they can develop singularities, where the equations effectively break down.
The Navier Stokes problem is one of seven Millennium Prize Problems identified by the Clay Mathematics Institute as among the most important unresolved questions in mathematics. A correct solution carries a $1 million prize.
OpenAI said its breakthrough was achieved using an internal AI system more powerful than its latest publicly known models. The company reportedly deployed around 10,000 AI agents to work on different aspects of the problem, with the system producing its proposed solution after about 88 hours of intensive computation.
The resulting mathematical argument runs to about 165 pages and was subsequently subjected to further checking by another AI system. OpenAI said its result shows that the Navier Stokes equations can develop a finite time singularity under certain conditions, providing an answer to a question that has occupied mathematicians for generations.
However, the announcement has already generated controversy within the mathematical community. Tristan Buckmaster, a mathematics professor at New York University, raised concerns after saying that he and another researcher had been working on a related approach and were close to publishing their own findings.
Buckmaster said their research had been conducted partly with AI tools and questioned whether material entered into OpenAI’s services could have contributed to the company’s result. OpenAI has denied using the researchers’ work improperly and said its own solution was developed independently. The company has, however, acknowledged that it cannot completely rule out the possibility that de identified user data may have contributed indirectly to improvements in its models.
OpenAI has also indicated that it will not seek the $1 million Millennium Prize associated with the problem. Instead, the company says the development demonstrates how increasingly capable AI systems could accelerate scientific research and help mathematicians tackle problems that have resisted conventional approaches for decades.
The claim will now face intense scrutiny from mathematicians, who will need to examine the proposed proof carefully before the result can be regarded as an accepted solution. Nature reported that the announcement represents the first claim by a computer system to have solved a truly major open problem in mathematics, although independent expert verification remains crucial.
If confirmed, the breakthrough would mark a significant milestone in the development of AI for scientific research and could strengthen arguments that advanced AI systems are moving beyond assisting researchers to making potentially original contributions to fundamental mathematics.
Leave a comment