A social media rumor claimed Claude had solved Navier-Stokes, but Terence Tao said he was not aware of any significant new developments. The claim spread fast, yet no paper, proof file, or expert review material was public. That leaves readers with a familiar AI story: a dramatic assertion, but no evidence that survives inspection.
Andrew Curran's 2.5 million views
Andrew Curran posted, “Claude has solved the Navier-Stokes equations.” The post drew over 2.5 million views. It also pushed a Millennium Prize Problem into the center of a rumor cycle before any proof could be checked.
The Navier-Stokes existence and smoothness problem is still listed as unsolved by the Clay Mathematics Institute. In practical terms, the claim would require public mathematical evidence that other researchers could inspect, and none was available in the material circulating online.
Terence Tao on September 3
On September 3, Tao discussed the potential impact of AI on mathematical research and used Navier-Stokes as an example. He described a hypothetical AI research process in which autonomous systems generate candidate mathematics at large scale and then verify results with Lean.
That scenario is a workflow, not a proof. It sketches how AI might help mathematics move from guesswork to formal verification, but it does not show that Claude produced a valid solution here.
What the rumor leaves open
The tension is simple. Curran’s prediction said Anthropic had solved a Millennium Prize Problem related to Navier-Stokes. Tao’s clarification said he was unaware of any significant new developments, and he said his earlier posts were only about a hypothetical research scenario.
For readers following the claim, the only concrete next step is to wait for proof materials that can be checked line by line. Until then, the rumor remains a rumor, and the unresolved question is whether Claude generated any proof at all or only became the subject of one.
