AIThe Decoder3h ago
Mathematicians say OpenAI doesn't understand its own Navier-Stokes proof
TL;DROpenAI's claimed math proof doesn't match what experts say it actually does.
Why it matters: Raises questions about AI reliability for complex reasoning and whether companies fully validate breakthrough claims.
OpenAI submitted a 166-page solution to the Navier-Stokes problem but apparently doesn't fully understand its own proof. Several groups of experts dispute the claimed methodology, saying the actual approach differs from what OpenAI described. And because language models are…
Read full articleSource: The Decoder · Opens in new tab