Stanford professor: Claude/GPT are pretty bad at annotating DNA motifs

anshulkundaje · x · 2026-09-24

Stanford professor Anshul Kundaje reports that Claude/GPT perform "pretty bad" at helping annotate DNA motifs — a painstaking manual task. His team is building enough labeled data to train models to accurately annotate massive motif lexicons, suggesting an intrinsic capability worth adding.

Related event: Stanford team finds frontier LLMs like Opus 5 and GPT6 poor at DNA motif annotation out of the box(6 posts)→

Original post →

More from Research

Research channel →