OpenAI Claims to Solve a 90-Year Math Problem, Sparks Row

OpenAI announced it had produced a solution to the Navier-Stokes existence and smoothness problem, one of the seven Millennium Prize Problems that has stumped mathematicians for roughly 90 years, using an internal, unreleased AI model, only for the announcement to immediately spark accusations of stolen work and academic malpractice from rival researchers, according to CBC News's reporting on the controversy.

The Navier-Stokes equations describe how fluids, including gases and liquids, move and change over time, giving scientists a mathematical way to predict fluid behavior under given conditions. The central unresolved question, first left open in 1934 after mathematician Jean Leray proved solutions exist only in a generalized sense, was whether smooth three-dimensional fluid motion could ever break down entirely, a question the Clay Mathematics Institute named one of its seven Millennium Prize Problems in 2000, each carrying a $1 million reward for a verified solution.

OpenAI's system produced an analytical proof, alongside a formal verification written in the programming language Lean, showing that an initially smooth fluid at rest can develop a singularity, a point where its physical properties spike to infinity, within a finite amount of time, according to OpenAI's own writeup of the result. The company disclosed genuinely striking operational detail behind the effort: agents were subdivided into communicating groups, with the group that produced the Navier-Stokes resolution involving on the order of 10,000 concurrent AI agents working together, arriving at their solution roughly 88 hours after the effort began, according to CNBC's reporting on the disclosure. OpenAI computer scientist Ven Chandrasekaran explained the significance directly at a press briefing: "Our proof does show that there exist fluids which start out perfectly normal, and under the Navier-Stokes equations, actually achieve infinite speed in a finite amount of time," a physically impossible outcome for a real fluid, suggesting the equations may not perfectly mirror physical reality under certain conditions.

The Navier-Stokes Controversy at a Glance

Almost immediately after OpenAI's announcement, two other mathematicians, Tristan Buckmaster of New York University and Levent Alpöge, who works at rival lab Anthropic, announced roughly 12 hours earlier that they had resolved several closely related problems themselves, with help from a variety of AI models, including OpenAI's own, according to Quanta Magazine's detailed reporting on the dispute. Terence Tao, a prominent mathematician, offered a genuinely pointed criticism of the broader dynamic this race has created: "The indiscriminate use of AI is turning the subject into a meaningless production quota 'game' that ultimately is of very little benefit, either to mathematics or to the world," according to CNN's reporting on his comments.

OpenAI addressed the timing and originality question directly in its own statement, acknowledging its effort began September 1 "after hearing a rumor" about progress on the puzzle that it later learned related to Alpöge and Buckmaster's work specifically. The company stated: "We (the researchers and the agents) did not see any of their work through any means until they released it publicly, in particular, no specific user data was accessed in order to solve this problem. While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models," according to CNBC's reporting on that disclosure, an unusually candid acknowledgment of genuine uncertainty about the provenance of its own result.

The Clay Mathematics Institute itself, the body that will ultimately verify any claimed solution and award the associated prize, struck a notably cautious tone. Institute president Martin Bridson said: "It is certainly an exciting day, as we contemplate the announcement of major advances in the human understanding of mathematics," according to Nature's reporting on the announcement, a measured statement that stopped well short of confirming OpenAI's result as verified or prize-worthy. Separately, mathematician Ravi Vakil offered a more sympathetic framing of the underlying achievement, regardless of the surrounding controversy, telling CBC News: "These are things that people have been chasing for a hundred years that we didn't think would be within reach," while acknowledging the drama "overshadows the real human accomplishment."

This story is worth understanding for any business evaluating AI's genuine capability at complex, open-ended reasoning tasks, since it represents one of the most significant claimed applications of large-scale, multi-agent AI coordination to date, connected directly to the broader agentic AI capability race we've tracked closely this year, including Jensen Huang's disputed "AGI has arrived" declaration tied to OpenAI's Astra model.

For businesses in research-intensive fields, this controversy is a genuinely useful cautionary case study in the attribution and provenance challenges that emerge when AI systems work on problems with active human research already underway, a dynamic worth understanding before deploying large-scale AI research agents on any competitive or academically contested problem.

What did OpenAI actually claim to solve?
OpenAI said its internal AI system solved the Navier-Stokes existence and smoothness problem, one of seven Millennium Prize Problems, by proving that smooth three-dimensional fluid motion can develop a singularity in finite time, a question unresolved since 1934.

Why is OpenAI's claim controversial?
Rival mathematicians Tristan Buckmaster and Levent Alpöge announced related results roughly 12 hours before OpenAI's announcement, raising questions about whether OpenAI's system may have drawn on their unpublished work, an allegation OpenAI has denied while acknowledging it cannot fully rule out indirect influence.

Has OpenAI's solution been officially verified?
Not yet. The Clay Mathematics Institute, which established the Millennium Prize Problems and would award the $1 million prize for a verified solution, has responded cautiously without confirming OpenAI's result as verified.

OpenAI announced its internal AI system solved the Navier-Stokes Millennium Prize Problem, a 90-year-old mathematics puzzle, using roughly 10,000 concurrent AI agents working for about 88 hours, only for the claim to immediately spark controversy after rival mathematicians Tristan Buckmaster and Levent Alpöge announced closely related results roughly 12 hours earlier. OpenAI has denied directly accessing the rival researchers' unpublished work while acknowledging it cannot fully rule out indirect influence through de-identified product usage data. The Clay Mathematics Institute, which would ultimately verify the solution and award its associated $1 million prize, has responded cautiously, and mathematician Terence Tao warned the episode reflects AI turning mathematics into "a meaningless production quota game" rather than genuine scientific benefit.