Should AI chatbots alert police to violent threats? A test case in Canada

Published by Al Jazeera on October 17, 2026, the feature uses the Tumbler Ridge shooting earlier this year as a case study for one of AI's hardest open questions: when, if ever, must a chatbot company break user privacy to warn authorities about a violent plan it has seen? The 18-year-old attacker had asked ChatGPT for guidance before carrying out Canada's deadliest school shooting in almost four decades, in which nine people were killed before he died by suicide. In September 2026 the province of British Columbia filed a civil lawsuit against OpenAI and chief executive Sam Altman; victims' families have filed separate suits, and a Washington State man has been charged with conspiracy. The article weighs calls for mandatory threat reporting against risks of mass surveillance, over-reporting and chilling legitimate conversations about mental health. Legal scholars quoted in the piece note there is no clear statutory duty for AI firms to call police, unlike the limited 'duty to warn' rules on therapists in some jurisdictions. The story is now one of the most-read tech-ethics features of the week, framing the policy debate heading into regulators' autumn hearings in the US, Canada and the EU.