OpenAI reports an internal multi-agent system solved the Navier-Stokes Millennium problem
OpenAI says roughly 10,000 coordinated agents powered by an internal model more capable than GPT-6 Astra produced an analytical solution to the Navier-Stokes existence and smoothness problem after about 88 hours and 130 billion output tokens. The company released the writeup and a Lean formalization, with GPT-6 Astra used for formalization and verification.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why it moved the index
A frontier system completing a famous open mathematical problem with a released machine-checkable proof is strong evidence that coordinated agents can automate high-level scientific reasoning. Confidence is below the top range because independent mathematical review is not yet established and the central internal model is unnamed.
Assessment history
-
R1
Toward 88 · confidence 84
Initial inclusion from a newly timestamped primary capability report with a released formal proof.
08 Sept 2026
Share this page
-
DoomBench assesses “OpenAI reports an internal multi-agent system solved the Navier-Stokes Millennium problem” as evidence moving toward doom, with magnitude 88 and confidence 84 out of 100 in the capability gains category.
-
The DoomBench assessment of “OpenAI reports an internal multi-agent system solved the Navier-Stokes Millennium problem” is based on reporting from OpenAI and records the editorial rationale, source quality, attribution, and revision...
-
DoomBench summarizes “OpenAI reports an internal multi-agent system solved the Navier-Stokes Millennium problem” as follows: OpenAI says roughly 10,000 coordinated agents powered by an internal model more capable than GPT-6 Astra...
https://www.doombench.com/news/openai-reports-an-internal-multi-agent-system-solved-the-navier-stokes-millennium-problem-2026-09-08