On September 21, 2026, British Columbia Attorney General Niki Sharma announced that the provincial government was filing a lawsuit against OpenAI in California “for its failure to notify law enforcement of threats made on its platform prior to the mass shooting at Tumbler Ridge Secondary school.” The shooting took place on February 10, 2026. The Province said it was working with the board of education for School District 59 Peace River South as the case proceeds.
The statement frames the case around a duty to act on credible threats: the lawsuit, it says, “raises serious questions about the responsibilities of technology companies when they become aware of credible threats of serious violence.” Sharma also said she had written to federal ministers with proposals to amend the Criminal Code of Canada “to ensure a pathway to human accountability for AI’s actions,” and that OpenAI had “not taken meaningful steps to address the concerns raised by survivors, families and the Tumbler Ridge community.”
The government placed the suit in a line of public cost-recovery litigation, noting that British Columbia has previously used the courts against tobacco and opioid manufacturers and vaping companies. That comparison is the novel part: well-known chatbot-harm suits such as Garcia v. Character Technologies were brought by bereaved families, while here a government is the plaintiff, treating an AI provider’s handling of threat signals as a matter of public harm.
Why it matters: it tests whether a chatbot operator that detects violent intent in conversations has a legal duty to warn authorities, a question every consumer AI company with an abuse-monitoring team now faces. What it does not show: a complaint is a set of allegations, not findings. The government’s statement does not itself set out the evidence about what OpenAI saw or decided, and no court has ruled on the claims or on whether a California court is the right forum for them.