GPT-6 Astra and Opus 5.5 produce a paper on MHD equilibrium counterexamples

Przemek Chojecki | PC @prz_chojeckiOriginally posted 2026.09.30

Open original on X

I've run these new results through GPT-6 Astra + Opus 5.5 and got even more counterexamples in families that strengthen Landreman’s results. The automated paper is here: https://www.ulam.ai/archive/explicit_C2_mhd_equilibria.pdf Figure by Opus 5.5 (+some calculations), final write-up by GPT-6 Astra. No Lean verification, just adversial LLM check-up.

This post quotes another post: x.com/0x_impai/status/2103161612677107942