Florida woman faces felony charge after Claude chatbot flags threat against sheriff's office

A Florida woman faces a felony charge after Claude's safety systems flagged her written threat to "shoot up" a sheriff's office, prompting a human reviewer to alert law enforcement. The case, confirmed on October 5, 2026, highlights the direct legal consequences when AI monitoring tools trigger police involvement based on user inputs.

How the threat was detected

Carli Michelle Heller of Bonita Springs wrote a diary-style entry to Anthropic's Claude chatbot on September 26. She described plans to attack the sheriff's office. Claude's automated safety systems flagged the language. A human reviewer then assessed the entry and judged the threat credible, reporting it to authorities.

Deputies located Heller and detained her without incident. She now faces a felony charge. Anthropic has not disclosed the exact charge or the sheriff's office involved.

Anthropic's data-sharing policy

Anthropic states it may share user information in limited emergencies when disclosure is necessary to prevent death or serious physical injury. The policy places the company in a position familiar to social media platforms and other tech firms: balancing user privacy against potential harm prevention.

The incident follows recent lawsuits against OpenAI, including claims that the company failed to alert police about a potential mass-shooting threat. These cases are shaping the legal expectations around AI providers' duty to warn.

Why this matters for legal and compliance professionals

The Heller case tests the boundaries of user consent and provider liability. Companies deploying AI tools should review their terms of service and emergency disclosure policies. The threshold for when a human reviewer escalates an automated flag - and the documentation of that decision - will face scrutiny if challenged in court.

Professionals in privacy law, public policy, and regulatory affairs need to track how courts weigh the duty to report against privacy promises. For those building internal AI governance frameworks, courses on AI regulatory compliance and AI public policy offer structured approaches to these emerging risks.