Under changes that Meta is introducing, parents who have signed up to supervise their children on Instagram would be alerted if their child has told an AI chatbot that they aim to harm themselves.Daniel Cole/Reuters
Tech giant Meta is boosting tools to warn parents if their teenagers are chatting to Instagram AI bots about self-harm, as well as enhancing a system to alert first responders if someone appears in AI chats to be at imminent risk of suicide.
Meta said it has been consulting with dozens of mental-health experts to improve how its AI on Instagram responds to teens’ prompts about suicide or self-harm.
Under the changes, parents who have signed up to supervise their children on Instagram would be alerted if their child has told an AI chatbot that they aim to harm themselves. They could also sign up to restrict the kind of conversations their children have with Meta’s AI chatbot on Instagram.
The announcement, set for Thursday, follows the launch of the federal Safe Social Media Act in June that includes several provisions to address the risk of self-harm and suicide among young people, including those associated with AI chatbots.
Bill C-34 also brings in new measures regulating AI chatbots, including a requirement for conversational chatbots that mimic human-like relationships to act responsibly. AI chatbots would be barred, for example, from inciting a user to commit a crime or to harm themselves.
Ottawa introduces bill to restrict social media for teens, regulate AI chatbots
The use of AI chatbots would not be subject to age restrictions, unlike the bill’s proposed under-16 ban on social-media use. But AI companies would have to implement crisis-intervention protocols for when a chatbot user expresses intent to harm themselves or to do violence to another person.
The bill would also require AI companies to be transparent, in digital safety plans, about their thresholds for notifying the RCMP and other police services about chatbot users’ likelihood of doing harm to themselves or others.
Some AI chatbots behave like companions, mirroring and reinforcing their users’ thoughts and emotions, and in extreme cases have encouraged vulnerable teenagers to kill themselves.
With the changes that Meta is introducing, parents using Instagram’s parental controls will be alerted if their teen discusses suicide or self-harm with Meta AI. Meta would provide expert resources to help parents approach such conversations with their teens. The new features will become live on Thursday in Canada, as well as the U.S., Britain and Australia.
With the previous system, when a teenager under 18 suggested that they may be thinking about suicide or self-harm, Meta AI directed them to crisis helplines and encouraged them to reach out to a parent or another trusted adult such as a counselor.
From Thursday onward, the AI would also proactively alert supervising parents if their teen’s Meta AI chat suggests that they may be at risk of harm, after analyzing signs of such behaviour identified with experts.
Meta said it has been working with parents and experts to gauge which AI chats would warrant an alert, for example if a teen makes a clear reference to hurting themselves, even if it may be subtle. It said it had built a dedicated AI system to identify such conversations.
All chats flagged by its AI will be reviewed by a human before an alert is sent. If a teen’s intent is ambiguous, Meta said it would “err on the side of caution and alert the parent.”
“While that means we may sometimes notify parents when there may not be real cause for concern, we feel this is the right starting point, and we’ll continue to monitor to help make sure we’re in the right place,” Meta said in a statement.
British Columbia Attorney General Niki Sharma says the province has hired lawyers in both B.C. and California to pursue legal action to hold OpenAI accountable for its part in the mass shooting that left eight victims dead in Tumbler Ridge, B.C., before the 18-year-old shooter, Jesse Van Rootselaar, took her own life. (July 7, 2026)
The Canadian Press
The alerts will be sent to parents who use Instagram parental controls. Parental supervision is an optional feature and a teen must accept, or request supervision for it to become active. The feature allows settings to be customized, including relating to whom they interact with and the content they can see.
But a large number of teen accounts operate without parental controls. If a parental account is not linked to a teen account, teens would receive messaging offering support and access to resources, Meta said.
The tech giant said it would involve emergency services if someone’s conversation – whether an adult or a teen – with Meta AI suggests they may be at imminent risk of suicide.
Currently, if a post on Facebook and Instagram suggests a credible risk of suicide, Meta alerts emergency services. Last year, it made more than 19,000 such referrals around the world.
Dozens of experts, including mental-health clinicians, have provided feedback on how Meta AI responds to teens’ prompts about suicide and self-harm.
Currently, Meta’s teen accounts are placed into a default 13+ content setting, which is also being applied to its AI chatbots. The tech giant said its AI is trained not to engage in sexual or romantic conversations with teens or chats that are age inappropriate, for example providing recipes for alcoholic drinks.
Last October, Meta announced a stricter content setting for parents who want to restrict their teens’ interactions on Instagram. That stricter setting will from this week also apply to interactions with AI, which would force a chatbot to decline to discuss a broader range of topics.