Vibeleaderboard
← All Intel
Intel / post

New on the Science Blog: Yes, Claude can do Nine Loops.

Source
Anthropic
Date
Anthropic@AnthropicAI

New on the Science Blog: Yes, Claude can do Nine Loops. Theoretical physicists predict how particles behave using formulas called scattering amplitudes. These are notoriously hard to compute, so researchers work with layers of increasingly fine corrections called “loops”—each added loop makes the answer more precise but takes exponentially more computation. Most calculations stop at two or three loops. Eight loops was the previous record in a simplified model physicists use as a testing ground (planar N=4 super-Yang-Mills), set by SLAC's Lance Dixon and collaborators. Last month, physicist and science writer @4gravitons issued a challenge: could an AI push past eight loops in this model, using only the compute budget an academic could reasonably access? Given a single prompt describing the nine-loop problem, Claude ran largely unsupervised for days in Claude Science and solved it using methods developed by Dixon and his colleagues, at a total cost of a few thousand dollars. Dixon independently verified the result, and von Hippel wrote about the experience for our blog. Read more: https://t.co/CS2f2qoIhJ

Key takeaways · AI-distilled
  • Scattering amplitudes, the formulas physicists use to predict particle behavior, are computed in successive 'loop' corrections that each cost exponentially more; most calculations stop at two or three loops.
  • The previous record, eight loops, was set by SLAC's Lance Dixon and collaborators in planar N=4 super-Yang-Mills, a simplified model used as a testing ground; physicist @4gravitons challenged an AI to beat it on an academic-scale compute budget.
  • From a single prompt, Claude ran largely unsupervised for days in Claude Science, used methods developed by Dixon and colleagues, and reached nine loops for a total of a few thousand dollars; Anthropic says Dixon independently verified the result.
Why it matters

Claude ran largely unsupervised for days to compute a nine-loop scattering amplitude, beating the previous eight-loop record, independently verified by the physicist who set that record, for a few thousand dollars in compute.

More from Anthropic
Recommended reads
Comments

Checking sign-in…

Loading comments…