“Reposting because this is similar to a weaker improvement of OAI’s result discussed…”

Reposting because this is similar to a weaker improvement of OAI’s result discussed on Frank Calegari’s blog last night, thus increasing my credence this is correct. See BLH’s comment thread: https://galoisrepresentations.org/2026/10/06/the-openai-problem-dump/

Quoting @captain_sude: https://github.com/CaptainSude/elliptic-curves-cm-modularity/releases/tag/v1.0.0

If OpenAI’s proof of the modularity conjecture over imaginary quadratic fields hold up, we have a full proof that every elliptic curve over a CM number field is modular.