Before OpenAI’s announcement, analytic programs on singularities and forced blow-up in fluid equations were already pushing the frontier.
Mathematicians such as Diego Córdoba, Luis Martínez-Zoroa, and collaborators have developed lines on singularity formation and blow-up scenarios in fluid models, with rigorous partial results over years.
Those programs—together with advances on forced Euler and harmonic-analysis techniques—form the scientific landscape any AI announcement must sit in.
The story does not start on press-release day: the community was already asking whether smooth solutions can break in finite time.
Since 2000, the Clay Mathematics Institute lists seven problems; Navier–Stokes existence and smoothness is one, with a one-million-dollar prize.
In September 2026 OpenAI’s agent announcement on Navier-Stokes reopened credit debates with Buckmaster and Alpoge’s Euler work; Clay/Millennium verification remains in progress and OpenAI said it is not claiming the prize.
On Epoch AI’s FrontierMath Erdős set (68 open Erdős problems), pre-release GPT-6 Astra was the only one of five models with a nonzero score (~3% = 2 problems) in the standard ~$300-budget run: it disproved #74 and proved #126. Looser extra runs do not count toward the official score.