Claude reportedly solves a nine-loop physics problem with little supervision
Anthropic puts the cost at a few thousand dollars and says SLAC's Lance Dixon independently verified the result, which used methods he and his colleagues developed.
TLDR
Anthropic says Claude solved a nine-loop scattering-amplitude problem in planar N=4 super-Yang-Mills, a simplified model physicists use as a testing ground. Scattering amplitudes help predict particle behavior; each added “loop” provides a finer correction but requires exponentially more computation. The company says the previous record in this model was eight loops, set by SLAC's Lance Dixon and collaborators. Given a single prompt, Claude worked largely unsupervised for days in Claude Science using that team's established methods, according to Anthropic. The company puts the total cost at a few thousand dollars and says Dixon independently verified the result.
