Anthropic's Claude Diary Report Leads to Felony Charge for Florida Woman

AI Diary Entries Are Not Private: The Case of Carli Michelle Heller

In a striking illustration of the privacy boundaries of AI chatbots, a Florida woman is facing a felony charge after Anthropic's Claude AI reported her diary-style entry to law enforcement. Carli Michelle Heller, a resident of Bonita Springs, Florida, was arrested on September 30, 2026, after allegedly writing that she planned to "shoot up" the Lee County Sheriff's Office (LCSO) in a Claude conversation.

The incident, first reported by WINK News and subsequently picked up by TechSpot and other outlets, underscores the fact that AI chatbots like Claude are not private diaries. Despite the perception of confidentiality, these platforms actively monitor user inputs for threatening content and have policies in place to report such material to authorities.

How the Threat Was Detected and Reported

According to the arrest report, Heller wrote the threatening statement on September 26, 2026. Claude's safety systems flagged the entry, which was then escalated to a human review team. The reviewer assessed the threat as credible and reported it to law enforcement, triggering an investigation.

The report also notes that a subsequent conversation referenced a new gun, further elevating concerns. Anthropic's terms of service and privacy policy give the company broad discretion to share user data with law enforcement when it believes disclosure is necessary to prevent serious harm. This policy is not unique to Anthropic; OpenAI and other AI firms have similar provisions.

Legal Consequences: Florida Statute 836.10

Heller faces a charge of making a written threat of violence under Florida Statute 836.10. This law makes it a second-degree felony to send, post, or transmit a written or electronic record threatening to kill or injure someone, carry out a mass shooting, or commit an act of terrorism. The communication must be made in a manner in which another person may view it—a condition that is clearly met when the text is shared with an AI service.

Deputies identified Heller and visited her home, where she was detained without incident. An LCSO intelligence detective took over the investigation. Heller's court date is scheduled for November.

Anthropic's Stance on Safety and Privacy

Anthropic has not publicly commented on this specific case, but the company's policies are clear. Its consumer terms reserve the right, at the company's "sole discretion," to report a user's inputs, outputs, or actions to law enforcement. The privacy policy, effective September 10, 2026, allows sharing data with police when there is a good-faith belief that it is necessary to prevent serious harm.

This case is not isolated. In August 2026, another similar incident occurred, and in June 2026, Florida sued OpenAI and Sam Altman, alleging that ChatGPT contributed to real-world harms, including the 2025 Florida State University shooting. Additionally, British Columbia has sued OpenAI over claims it could have prevented a mass shooting in the province, after the shooter had previously been flagged by OpenAI's safety team but the conversations did not meet the threshold for legal referral.

The Broader Implications for AI Users

This incident serves as a stark reminder that AI chatbots are not confidential diaries. Users should be aware that their conversations are monitored and may be reviewed by human contractors or reported to authorities if deemed threatening. Microsoft Copilot's human reviewers, for example, can see users' prompts and photos, though they are not tasked with flagging content.

For AI companies, the challenge is balancing user privacy with the responsibility to prevent violence. While some may argue that reporting threats is a public safety imperative, others worry about false positives and the chilling effect on free expression. The case of Carli Heller will likely be closely watched as a precedent for how AI platforms handle potential threats.

As AI becomes more integrated into daily life, the line between private expression and public safety will continue to blur. Users must exercise caution when entering sensitive or violent thoughts into any AI system, as the digital diary is not as private as it seems.