
What happened
OpenAI's internal system proved that smooth fluid motion can break down into a singularity in finite time, resolving the Navier–Stokes Millennium Prize Problem.
Why it matters
The question had been unresolved for roughly 90 years; OpenAI shared both a writeup and a formal proof in Lean, and does not claim the prize.
What to watch
The result hinges on whether the mathematical community accepts the proof; OpenAI notes its model is significantly more capable than GPT‑6 Astra.
WHO IT HITSMathematicians and fluid dynamicists will need to verify the proof; AI researchers and business leaders should watch for further demonstrations of AI's capability in scientific discovery.
Summaries like this, every morning.
OpenAI's announcement highlights a significant leap in AI reasoning, as the internal model outperforms GPT‑6 Astra in benchmarks. The effort began after rumors of other resolutions and involved a multi-agent system with 4.9 million messages and 300 billion output tokens. The proof's Lean formalization adds rigor, and OpenAI acknowledges concurrent work by Levent Alpöge and Tristan Buckmaster on a related Euler problem. This milestone suggests AI can tackle frontier mathematical challenges, but its acceptance will depend on peer review and the broader mathematical community's validation.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Chainalysis reported 138.5 million payments on Base using x402, up from 58,800 total transactions as of Q3 202…

SpaceX's "super intelligence" segment, which houses the xAI business and X ad revenue, booked $7.56 billion in…

A vulnerability dubbed PHP2Shell, dormant for over 10 years in the PHP language used by about 70% of websites…

Block unveiled Dessa, which it says is the first AI model trained entirely on human speech with no text data…

At Anthropic's "Machines of Loving Grace" exhibition and DevelopersIO 2026 TOKYO, Claude and Claude Code power…

A FitLog developer replaced repeated prompt warnings with a harness — a 48-line CLAUDE.md "map" linking six do…
