🤖 Princeton professor disputes OpenAI's claim of solving the Navier–Stokes problem
Mathematics professor Sergiu Klainerman published an essay titled “AI will not make mathematicians obsolete” in the journal Inference on September 21, 2026 — a response to OpenAI's September 8 announcement: an internal system of ~10,000 agents produced a solution to the Navier–Stokes problem, one of the seven “Millennium Prize Problems” of the Clay Mathematics Institute, in 88 hours, with an analytical note and formalization in Lean.
🌍 This is the first case where a Millennium Prize problem has been claimed as solved by a machine, and immediately — a public professional review of the result by a leading mathematician. Klainerman asserts that the machine solved only a weakened forced version: the singularity — infinite flow velocity at a finite time — arose under an artificially chosen external force from a state of rest, while the true unforced problem remains open. A pragmatic signal for the industry: the open repository openai/NavierStokesAndEuler and the Lean formalization allow the proof to be checked independently, while the choice of meaningful problem formulations remains with humans.
👤 The loud “AI solved a Millennium Prize problem” is more nuanced than it sounds: the real unforced Navier–Stokes equations are untouched. At the same time, there is no basis from open data to call this “eval-gaming”: the system formally solved exactly the problem that was formalized, and Klainerman calls the allowance of the forced formulation, permitted by Charlie Fefferman, an error — this is a dispute within mathematics itself. The essay and OpenAI's post can be read independently to form one's own opinion. To the fear of young mathematicians “will I be replaced?”, the author answers categorically: the disappearance of mathematics as a human activity is not included in the list of AI risks.
Source 1: https://inference-review.com/article/ai-will-not-make-mathematicians-obsolete
Source 2: https://openai.com/index/navier-stokes-solution/