Research · Updated 17 Oct, 02:30 pm IST
Mathematicians question OpenAI’s claimed Lean proof of Navier–Stokes
Why it matters for readers: If you’re curious, it shows that AI can produce formal‑looking mathematics but experts may still question whether the work truly follows known approaches.
- OpenAI published a 166‑page potential proof of the Navier–Stokes problem written in Lean, a formal verification language.1
- Several mathematicians, including Gonzalo Cao‑Labora, say the detailed approach differs from the Spanish mathematicians OpenAI cited, making the similarity only surface‑level.1
- The Clay Mathematics Institute has been reviewing OpenAI’s Lean submission since it was released.1
- Before publication, Tristan Buckmaster alleged OpenAI built on work he and an Anthropic researcher had produced with public models; OpenAI denied using Buckmaster’s data.1
- In early October OpenAI released roughly 400 more results on GitHub, which many mathematicians said cheapened the field and made the work hard to evaluate.1
Get a brief like this every morning
Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.
Get started