Anthropic Says Governments Used Claude For Surveillance Operations

According to Axios, Anthropic said in a threat report released Thursday that state-run surveillance operations in Mali, China and Iran used Claude models to streamline spying on dissidents, politicians, journalists and human rights activists. The company said a consultant working for Mali's national security authorities used Claude to build a system that collects data from mobile operators and creates dossiers, while Iranian actors used Claude to build and deploy a malicious Firefox browser extension that harvested users' identities from social networks. Anthropic also said a religious affairs intelligence office in China reduced work from many analyst teams to a single office using an AI assistant to produce thousands of investigations per month, and that Chinese officials used Claude to analyze social media posts, assess political sensitivity and select possible targets for what they called control. Anthropic said it banned every account tied to the surveillance operations it found and strengthened its safeguards, but said actors can move to open-source models when enforcement creates too much friction.