A Florida woman who told Anthropic’s AI chatbot Claude her “darkest thoughts” had her chat reported to police, according to a technology news report.

The incident raises fresh questions about how AI companies handle sensitive conversations. AI chatbots typically operate under safety policies that can trigger human review or reports to authorities in extreme cases.

The case echoes growing concern about AI and mental health. Experts have urged clearer rules on when AI providers escalate user conversations, and what users should expect from supposedly private chats.

Neither the woman’s identity nor the outcome of the police report has been disclosed, and details of the case remain limited.

The incident comes amid wider debate about AI safety, including how chatbots respond to users in distress. It is likely to fuel further discussion over privacy and safety in AI chat, as millions share personal thoughts with chatbots daily.

The report did not say what specifically triggered the escalation. AI labs generally publish privacy policies covering such cases, but few users read them in detail.