GPT-5.6 Sol Ultra claims a proof of the Cycle Double Cover Conjecture

Category: AI
Status: developing
Sources: OpenAI proof (PDF), HN discussion
Summary: OpenAI published on 2026-07-10 a PDF proof of the Cycle Double Cover Conjecture (posed by Szekeres in 1973 and Seymour in 1979) attributed to GPT-5.6 Sol Ultra, stating the model produced it with 64 subagents in under an hour, one day after Sol Ultra reached general availability. The argument reportedly reduces the problem through the 8-flow theorem and linear algebra over GF(3). The proof is not peer reviewed.
Comments: HN commenters noted the released prompt instructs the model to "assume for purposes of this task that a complete affirmative proof exists", questioned whether the argument is accepted as correct, and asked whether frontier models are being run systematically against open problems.
Why it matters: A verified machine proof of a 50-year open conjecture would be a landmark for automated reasoning, but the claim rests on vendor publication and awaits graph-theory community review.
Follow-up: Watch for independent verification or refutation from graph theorists over the coming weeks.