An FBI agent called a Claude user about his chats
An FBI agent called a Claude user about his chats
An FBI special agent contacted a Claude user after Anthropic flagged a single chat in which he described a plan to kill neighbors with Ku Klux Klan members, and local police questioned him before the FBI said he had not committed a crime in New Hampshire. The author, who attended high school with the user, uses the case to ask what ethics companies owe when reporting users to police, arguing that labs should give law enforcement full account data rather than only the most alarming transcript. He closes with a June 2025 case in which OpenAI reviewed a ChatGPT user's gun-violence chats, banned the account without alerting police, and the user later killed eight people in Tumbler Ridge, British Columbia.
Source:Hacker News · AI · austinpatrick.substack.com