Anthropic says it disrupted every operation in its Claude misuse report

Anthropic says its threat intelligence report details attempts to misuse Claude and how it detected and disrupted every operation described. The company characterizes the cases as atypical and among the most sophisticated misuse it has seen. It says it used lessons from the cases to strengthen safeguards and, where appropriate, shared findings with authorities and other AI companies. Anthropic says publishing the report will help others spot similar activity on their platforms. A user quoting the announcement offers a surveillance critique, arguing that Claude itself is “a form of mass surveillance.”