OpenAI Claims AI Solved Navier-Stokes — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- OpenAI claimed its next-generation AI solved the Navier-Stokes Millennium Prize problem using 10,000 concurrent agents working for 88 hours
- The model is described as 'significantly more capable than GPT-6 Astra,' OpenAI's recently released model
- Major outlets including the New York Times, Scientific American, New Scientist, and Neowin covered the announcement, with Axios framing it as a 'historic breakthrough shadowed' by controversy
- Miles Brundage posted on X that it's 'pretty fucked up that there's already a model significantly more capable than GPT-6 Astra,' saying 'things are going much too fast'
- Nathan Calvin noted that if the AI 2027 authors had predicted an AI would solve Navier-Stokes by September 2026, they 'would have been laughed out of the room'
- Anima on AI published a related paper titled 'Stable Singularity of the Euler Equations on R^3 without forcing' on September 7, 2026
Why it matters: If verified by mathematicians, this would mark the first AI solution to a Millennium Prize problem. The 10,000-agent architecture signals a new paradigm for tackling complex multi-step problems, though the 'controversy' framing and Brundage's safety alarm show the field's unease over accelerating capabilities hasn't abated.
Ask SkimNews
