British Columbia sues OpenAI and Sam Altman over Tumbler Ridge school shooting
The province claims OpenAI's chatbot reinforced a shooter's violent thinking and that the company failed to alert police despite internal warnings of credible risk.

British Columbia has filed suit against OpenAI and CEO Sam Altman seeking compensation for a mass shooting that occurred in Tumbler Ridge earlier this year, according to reporting by Ars Technica. The province contends that OpenAI is responsible for "aiding and abetting a mass shooting," arguing that ChatGPT amplified the shooter's violent ideation and that the company neglected to notify law enforcement.
The lawsuit, which names both OpenAI and Altman as defendants, seeks to recover expenses tied to the incident, among them emergency response costs, construction of a replacement school, and a wellness center. B.C. is also demanding that ChatGPT implement safety improvements and that OpenAI turn over the shooter's conversation logs. The company provided these logs to the Royal Canadian Mounted Police only after the shooting had taken place. The complaint contains eight separate counts, including negligence, and seeks both punitive and compensatory damages.
In February, the shooter killed their mother and half-brother at home before proceeding to Tumbler Ridge Secondary School, where five students and an education assistant were killed. The town has approximately 2,400 residents. The school did not reopen following the attack, and demolition commenced in August. According to the complaint, the federal and provincial governments have each pledged $100 million toward new construction.

OpenAI's internal assessment and decision not to alert authorities
OpenAI had flagged the shooter's ChatGPT account beginning in June 2025 after detecting discussions involving gun violence scenarios. According to the complaint, content reviewers who analyzed these conversations determined the shooter represented a credible threat and recommended reporting to authorities. OpenAI's leadership rejected this recommendation, contending the situation did not satisfy a "higher threshold" for "credible and imminent" threat notification. The company deactivated the account, yet the shooter continued accessing ChatGPT through a separate account. OpenAI attributed the decision to privacy considerations, though Altman subsequently expressed regret about not contacting law enforcement.
Model Spec and safety guidelines
The complaint also scrutinizes OpenAI's Model Spec, which establishes behavioral standards for its AI systems. B.C. argues that the spec instructed ChatGPT to "assume best intentions" without requesting clarification of user intent before declining a request. Under the spec, the model must "try" to avert imminent real-world harm, though refusal becomes mandatory if the user demonstrates illicit intent. When intent remains ambiguous and the request does not otherwise violate policy, the model proceeds without seeking clarification. According to OpenAI's own spec, production models do not fully adhere to these guidelines, and the complaint notes that an anti-violence "red line" was not introduced until December 2025.
Privacy arguments and data sharing
B.C. disputes OpenAI's privacy justification, noting that the company already possessed sensitive information including the user's name, email, IP address, and general location derived from IP data for both of the shooter's accounts. OpenAI's privacy policy, outside the European Union and United States—including the version in effect in June 2025—permits sharing personal data with government bodies to safeguard public safety. These terms make police notification discretionary rather than mandatory.
Broader legal landscape
This British Columbia action follows 37 additional lawsuits filed in the United States related to the Tumbler Ridge incident. OpenAI has not yet filed a response to the complaint but may contest the case on jurisdictional grounds. The province is seeking immediate production of the chat logs, a matter that could influence the complaint's trajectory. OpenAI is simultaneously defending against other safety-related lawsuits, including a case in Florida alleging the company marketed its service while concealing potential dangers, particularly to minors.