OpenAI announced that an advanced artificial intelligence model under development has resolved the Navier-Stokes problem, a longstanding mathematical challenge that has persisted for over a century. The company revealed that the breakthrough was achieved within days using extensive computational resources, amounting to millions of dollars in processing costs.

The Navier-Stokes equations, central to this problem, describe the motion of fluids such as air and water. These equations are foundational in fields ranging from aircraft design and weather prediction to medical research on blood flow. The challenge lies in proving whether solutions to the equations always exist and behave smoothly over time. The problem is one of seven that make up the Millennium Prize Problems, a collection of the most difficult unsolved questions in mathematics recognized by the Clay Mathematics Institute and accompanied by a $1 million award for a verified solution.

OpenAI’s team, led in part by researcher Sebastien Bubeck, reported deploying a system involving approximately 10,000 autonomous AI agents working concurrently, exchanging information as they raced toward a solution. The project began in late August with the training of a new model that surpassed the capabilities of publicly available AI. Shortly after, amid rumors that another group had resolved certain Millennium problems, OpenAI applied the model to all six unsolved questions remaining on the list. The Navier-Stokes problem showed particular promise and quickly became the company’s primary focus.

Mark Chen, OpenAI’s chief research officer, described the accomplishment as a major milestone in AI research, highlighting the technology’s potential to tackle some of the most difficult questions in science and mathematics. However, OpenAI emphasized that it does not intend to claim the Millennium Prize if its findings are validated. According to Clay Mathematics Institute policy, any purported solution must be published in a peer-reviewed journal and withstand at least two years of scrutiny within the mathematical community before prize consideration.

Professor Martin Bridson, president of the Clay Mathematics Institute, responded by underscoring the institute’s rigorous and deliberate evaluation process, cautioning that it remains too soon to confirm the AI’s claims as definitive.

The announcement coincided with the release of AI-assisted research by mathematician Tristan Buckmaster of New York University, along with Levent Alpoge of Anthropic, an AI firm competing with OpenAI. Their work addressed related mathematical equations but did not claim a full solution to the Navier-Stokes problem.

The developments underscore the rapidly evolving role of AI in advancing mathematics, a domain long considered resistant to automation. However, the emergence of machine-generated proofs has also prompted debate within the academic community over whether such results equate to genuine human insight or understanding. The coming months and years will be critical in assessing the long-term impact of AI on fundamental scientific inquiry.