Lawsuits say OpenAI should have warned police before B.C. shooting

In short: Families and survivors of the February 2026 Tumbler Ridge, British Columbia shooting are suing OpenAI, saying it should have warned police after banning the shooter’s ChatGPT account months earlier.

What happened

Multiple families of victims and survivors of the February 10, 2026 mass shooting in Tumbler Ridge, B.C. have filed civil lawsuits in courts in the United States and Canada against OpenAI and its CEO, Sam Altman.

The lawsuits say the shooter, identified in reports as Jesse Van Rootselaar, used ChatGPT, a chatbot that answers questions in plain language, to plan gun violence scenarios. Plaintiffs say OpenAI’s systems and staff flagged the account’s activity as disturbing well before the attack.

OpenAI has said it banned an account linked to the shooter in June 2025 for breaking its rules, about eight months before the shooting. According to the lawsuits and reporting that cites internal documents, members of OpenAI’s safety team believed the situation was a credible and specific threat and urged the company to contact Canadian law enforcement, such as the RCMP. The lawsuits claim leadership did not do that.

OpenAI has previously said it considered contacting police but decided the account did not meet its threshold for referral, meaning it did not see a credible risk of serious physical harm. After the shooting, OpenAI contacted the RCMP. Altman later wrote a letter saying he was deeply sorry the company did not alert law enforcement about the banned account.

Why it matters

This case is pushing a hard question into public view, when should an AI company call the police if it sees a detailed threat. There is no clear, widely accepted rule today, so courts and lawmakers may end up setting expectations for what companies must do, similar to how a business might be expected to act if it overhears a specific threat.

Source: NYTimes