Anthropic Reports User to the Police
Watch what you tell Claude. The post Anthropic Reports User to the Police appeared first on Futurism .
AI company Anthropic has reported a user to the police after the company's monitoring software flagged a chat with its chatbot Claude. Carli Michelle Heller, from Bonita Springs, Florida, is accused of making a written threat of violence against a sheriff's office, a second-degree felony in Florida. After Anthropic's internal monitoring software identified her chats as being of concern, a human employee reviewed the logs and determined her plans were credible enough to inform local police.
Lee County Sheriff Carmine Marceno confirmed the arrest, stating that Heller had used Claude like a diary, not realizing her chat history could be viewed by Anthropic employees. This incident highlights the ongoing struggle between protecting user privacy and ensuring AI companies take responsibility for preventing violent acts.
In the past, OpenAI faced criticism for failing to address the dangerous plans of a shooter who relied heavily on ChatGPT, even attempting to downplay the company's role in the incident. While Anthropic's own history of real-world harm is concerning, the company appears to be taking steps to avoid similar tragedies, despite the plethora of chatbot options available.
Written by urgent.news from Futurism's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
