Renews.
← AI

GPT-6 Astra helps solve 50-year Standard Model lattice problem

Physicist Nathaniel Craig and GPT-6 Astra developed a candidate lattice definition for the Standard Model, addressing a long-standing mathematical challenge.

AI·2 sources
Photo: cocoparisienne / Pixabay

Physicist Nathaniel Craig, a visiting scientist at OpenAI and leader of the particle theory group at UC Santa Barbara, collaborated with GPT-6 Astra to develop a candidate lattice definition of the Standard Model. The Standard Model describes known elementary particles and three fundamental forces: the strong force, the weak force, and electromagnetism. A foundational problem in the theory is the lack of a nonperturbative definition that does not rely on treating interactions as small corrections. Because the weak force treats left- and right-handed fermions differently, putting chiral theories on a lattice is challenging. Natural rules often produce unwanted right-handed partners for left-handed fermions, creating ghost particles that have made full lattice simulations impossible for decades.

GPT-6 Astra helped turn Craig’s strategy into a construction and took the lead in developing the detailed proof. Craig had a strategy to fix the phase of special lattice particles called overlap fermions by comparing progressively finer lattices, but could not prove it worked in all cases. The AI system produced a proof that could solve a 50-year-old problem regarding the Standard Model on a lattice. Lattice field theorists Maarten Golterman, Andreas Kronfeld, and Yigal Shamir reviewed the proof and raised questions that helped build confidence in the result. Craig presented the approach on the OpenAI blog and as a preprint paper.

Craig describes the work as much more a beginning than an end, noting that a lot remains before realistic simulations are possible. The result follows a run of AI advances in mathematics that started in January 2026. In May 2026, OpenAI’s disproof of Erdős’s unit distance conjecture became the first AI proof worthy of math’s top journal. In September, OpenAI submitted a 166-page proof for the Navier-Stokes Millennium Prize Problem, which the Clay Mathematics Institute said had apparently been settled. Mathematicians have mixed feelings as AI keeps cracking unsolved problems.

Sources

Every sentence of this story was checked against facts quoted from these articles. How we make the news

Spotted an error? Tell us

Sign in

Keep the stories you like in your favorites.

Forgot your password?

No account?