OpenAI's GPT-5.6 Sol Ultra reportedly solves a 50-year-old math problem in under an hour
OpenAI's GPT-5.6 Sol Ultra produced a proof of the Cycle Double Cover Conjecture in under an hour, using 64 subagents working in parallel. The conjecture had remained unsolved for 50 years. Mathematician Thomas Bloom calls the proof surprisingly elementary but criticizes the lack of citations for known prior work. The bigger question remains: Does AI just recombine existing knowledge, or does it create something new? The article OpenAI's GPT-5.6 Sol Ultra reportedly solves a 50-year-old math problem in under an hour appeared first on The Decoder.
OpenAI's GPT-5.6 Sol Ultra used 64 subagents working in parallel to prove the Cycle Double Cover Conjecture, which had remained unsolved for 50 years, in under an hour. Mathematician Thomas Bloom called the proof surprisingly elementary but criticized the lack of citations for known prior work.
Multi-agent parallel collaboration has become a core paradigm for frontier reasoning. The model of 64 subagents collaboratively tackling a single problem validates the feasibility of agent swarm architecture in complex scientific reasoning, marking a shift of AI agents from 'assistive tools' to 'autonomous research entities'.
AI agents autonomously completing mathematical theorem proofs will accelerate the automation of scientific research. However, the lack of citations in the proof exposes the current agent system's shortcomings in academic normativity—verifiability and reproducibility are necessary conditions for agent research to move from the lab to academia.
For industries requiring deep reasoning such as drug discovery, materials science, and financial modeling, multi-agent reasoning architectures provide a replicable research automation solution.
Observe whether OpenAI will productize multi-agent mathematical reasoning capabilities, and how academia will establish citation and verification standards for AI-generated proofs.