THE FRONTIER SIGNALSTHE EVIDENCE LEDGER
How much AI research is already led by AI?
Self-reported, model-assisted ratings on a fixed task basket. No measured category was fully autonomous; this is not an independent cross-lab capability score.
What was claimed?
Anthropic reports that Claude led 26% of its measured AI R&D work in August, with human supervision.
What does the evidence support?
Company claim, independent confirmation unresolved
Its published method offers a window into internal development workflows beyond public chat tools.
What are the limits?
Self-reported, model-assisted ratings on a fixed task basket. No measured category was fully autonomous; this is not an independent cross-lab capability score.
Claim or evidence date: 2026-08. Last verified: .
Colors classify evidence; they do not rank truthfulness. Explore hype versus reality →