Breakthrough

OpenAI Says It Solved One of the World’s Most Complex Problems — Did It?

OpenAI
OpenAI | Photo: Shutterstock

OpenAI issued a dramatic statement saying it had solved one of the world’s most complex and long-standing mathematical problems — the Navier-Stokes equations — in just 88 hours, according to a report by CNBC.

Mathematical equation | Photo: Shutterstock

The equations, which deal with the motion of liquids and gases, are one of the seven “Millennium Prize Problems” designated by the Clay Mathematics Institute, each carrying a $1 million prize for whoever solves it. According to the company’s reports, the task was accomplished using a complex system of about 10,000 autonomous artificial intelligence agents that operated simultaneously, communicated with one another, ran code and used information from the internet.

Suspicions of copying and a battle with rivals

Alongside the excitement over the technological achievement, OpenAI’s claims sparked a major uproar in the academic community, with Tristan Buckmaster, a mathematics professor at New York University, raising serious questions about the originality of the solution.

Buckmaster said he had worked on the problem concurrently with mathematician Levent Alpoge, who currently works at Anthropic — OpenAI’s direct competitor.

OpenAI
OpenAI | Photo: Shutterstock

He said that because OpenAI’s solution was suspiciously similar to the method used by him and Alpoge, he raised concerns that the company’s artificial intelligence models had been exposed to information or code from their working meetings conducted on the company’s systems.

The tech giant’s response and broader implications

OpenAI quickly denied the allegations that it had used private information or unauthorized research drafts. The company confirmed that the AI agents’ run began following rumors of progress in Buckmaster and Alpoge’s research, but insisted that the agents had not been exposed to the researchers’ work before it was made public.

OpenAI CEO Sam Altman, illustration
OpenAI CEO Sam Altman, illustration | Photo: Shutterstock

However, the company qualified its response, noting that it could not completely rule out the possibility that anonymized data from the use of its products had helped improve the models. The worlds of science and technology are now looking to the Clay Mathematics Institute, which has yet to issue an official response and will have to determine whether the machines’ proof is indeed valid and historic.

Key takeaways

  • OpenAI said it solved the Navier-Stokes equations in 88 hours, according to a CNBC report.
  • The problem is one of the seven Millennium Prize Problems designated by the Clay Mathematics Institute and carries a $1 million prize.
  • OpenAI said the result came from about 10,000 autonomous AI agents that worked in parallel, communicated, ran code, and used internet information.
  • The claim triggered questions in academia after mathematician Tristan Buckmaster suggested the solution looked suspiciously similar to work he had done with Anthropic’s Levent Alpoge.
  • OpenAI denied using private information or unauthorized research drafts, but said it could not completely rule out anonymized product data having helped improve its models.

Questions and answers

What did OpenAI claim it solved?
OpenAI said it solved the Navier-Stokes equations, a long-standing mathematical problem about the motion of liquids and gases. The company said it did so in 88 hours.
Why did the academic community question OpenAI’s claim?
Mathematics professor Tristan Buckmaster said the solution looked suspiciously similar to work he had done with Levent Alpoge. He raised concerns that OpenAI’s models may have been exposed to information from their work meetings.
Has the Clay Mathematics Institute confirmed the proof?
Not yet. The article says the institute has not issued an official response and will have to determine whether the proof is valid and historic.

Share this story

About the writer

Shay Nogai

Writes for TV10, Israel’s economic channel. This story was translated from the original Hebrew.

More stories by Shay Nogai

More for you