Attack
Anthropic details the strengthening of Claude's protections against abuse linked to the US midterm elections.
Requests breaking election integrity rules appear in about 0.002% of conversations.
According to tests by the ISD and the McCain Institute, Claude could miss a malicious campaign split into harmless queries.
New classifiers aim to better spot novel and multi-step abuse, including fake news sites.
A fake news site run from abroad used Claude to promote Russian, Iranian and North Korean narratives.