Came across an update from OpenAI: its newly launched science blog has published its first guest post. Nathaniel Craig of UCSB’s particle theory group (who is also a visiting scientist at OpenAI) explains how he and GPT-6 Astra explored a candidate lattice-based nonperturbative definition of the Standard Model of particle physics.
According to the post, although the Standard Model agrees remarkably well with experiments, it has long lacked a complete definition that doesn’t treat interactions as “small corrections”—a problem that has remained open for more than fifty years. Craig had an idea, but hadn’t been able to prove it in all the cases that needed to be covered. Astra helped turn the idea into a construction and led the detailed proof. There was also a preprint completed with Astra’s assistance in October.
The post is careful to note that this is still only a candidate proposal: it offers more of an answer in principle, is still a long way from being usable for realistic lattice simulations, and will need discussion by other researchers.
#OpenAI #GPT6 #Standard Model
According to the post, although the Standard Model agrees remarkably well with experiments, it has long lacked a complete definition that doesn’t treat interactions as “small corrections”—a problem that has remained open for more than fifty years. Craig had an idea, but hadn’t been able to prove it in all the cases that needed to be covered. Astra helped turn the idea into a construction and led the detailed proof. There was also a preprint completed with Astra’s assistance in October.
The post is careful to note that this is still only a candidate proposal: it offers more of an answer in principle, is still a long way from being usable for realistic lattice simulations, and will need discussion by other researchers.
#OpenAI #GPT6 #Standard Model
