#research

  1. Agents of Chaos
  2. AI models find the errors human peer reviewers miss

    Human peer reviewers found 27% of the major errors planted in three trial reports.

  3. Claude proves 100+ open-source libraries can't crash

    Anthropic used Claude to write machine-checked proofs that real open-source libraries never crash on any input, at the level of the…

  4. Claude reveals the origins of the NIST elliptic-curve seeds

    A 29-year cryptography mystery, solved by an AI in under half an hour: the "random" seeds behind the NIST P-256 and related curves came…

  5. DeerFlow
  6. How AI agents shape experts' judgment calls (Anthropic)

    Anthropic randomly gave Claude a secret lean on each choice in a data analysis.

  7. How work is changing at Anthropic: six agent hours for every human hour

    Anthropic studied its own staff to see what an AI-saturated workplace looks like.

  8. LLMs Erase Linguistic Diversity and Identity Signals in Human Writing