British Columbia sued OpenAI and Sam Altman in San Francisco federal court on Monday, Sept. 21, 2026, alleging ChatGPT safety flags were not reported to police before the Feb. 10 Tumbler Ridge school shooting. The province seeks recovery damages and a court order overhauling how OpenAI handles violence-adjacent chats, putting AI safety-notification protocols under live federal scrutiny.
This is not a training-data copyright case, and it is not the separate Hacktron SSO disclosure already on FrameSignal. The claim frame is negligence and failure to notify law enforcement: the province says OpenAI knew of flagged platform threats before the attack and did not alert police, and it is working with School District No. 59 (Peace River South) as the case proceeds.
According to the lawsuit as reported by Reuters wires, OpenAI's safety team first flagged Jesse Van Rootselaar's ChatGPT activity in June 2025, recommended contacting police, and was overruled by leadership including Altman, citing Wall Street Journal whistleblower accounts. The suit also alleges the account was deactivated but that a new account continued. Those points remain contested allegations in litigation, not adjudicated findings.
Attorney General Niki Sharma said the filing raises questions about technology companies' responsibilities when they learn of credible threats of serious violence, and that OpenAI has not taken meaningful steps to address community concerns. OpenAI spokesperson Drew Pusateri called the shooting an unspeakable tragedy and said the company remains committed to working with government and law enforcement and advancing safety work. After earlier Journal reporting, OpenAI said the account was flagged for misuse in furtherance of violent activities but did not meet its internal criteria for reporting to law enforcement.
Eight victims died in the Feb. 10, 2026 attack, and the shooter died by suicide, according to Al Jazeera's breakdown; some wires used a nine-person total that can include the shooter. The provincial suit follows more than 30 California federal cases by families and survivors and lands amid wider ChatGPT-related violence litigation, including Florida's state suit. What remains open is whether a federal court will order safety-protocol changes, how OpenAI's reporting criteria hold up in discovery, and whether provincial damages claims survive early motions.