Meta begins alerting parents when teens discuss suicide or self-harm with Meta AI

The alerts are live for supervised Instagram accounts in the US, UK, Australia and Canada, with a global rollout planned by the end of 2026.

Meta is notifying supervising parents when a teenager’s Meta AI conversation indicates a possible risk of suicide or self-harm

Parents in the US, UK, Australia and Canada can now receive an alert when Meta identifies a supervised teenager’s conversation with Meta AI as suggesting a risk of suicide or self-harm.

The protection operates through Instagram’s parental supervision tools. Meta says a dedicated AI system identifies conversations that may warrant intervention, including subtle references to self-harm, but every flagged chat is manually reviewed before the parent is contacted.

The human review does not remove the possibility of false alarms. Where a teenager’s intent remains ambiguous, the company says it will favor notifying 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 states.

Parents receiving an alert will also be given resources intended to help them approach the conversation. Meta AI already directs teenagers to crisis helplines and encourages them to contact a parent, counselor or another trusted adult when they indicate that they may be considering suicide or self-harm.

The intervention remains tied to opt-in parental supervision. Teenagers whose Instagram accounts are not supervised are not covered by the new parent alert system.

Larry Magid, Chief Executive Officer and Co-Founder of ConnectSafely, frames the approach as a balance between privacy and parental intervention.

“While I believe that teens have a right to privacy, I also believe parents need to be informed if their teen may be at risk of hurting themselves,” he says. “I appreciate how Meta struck the right balance; protecting teen privacy while ensuring parents have the information they need to support their teen.”

Emergency service referrals are still in development

Meta is separately developing a system that could contact emergency services when an adult or teenager appears to be at imminent risk of taking their own life during a conversation with Meta AI. Unlike the parent alerts, that capability has not yet been introduced.

The company already makes referrals when Facebook or Instagram posts indicate a credible suicide risk. Meta says it made more than 19,000 such referrals worldwide last year, enabling first responders to carry out wellness checks. That figure relates to its existing work across the two social platforms, rather than the Meta AI system now under development.

Clinicians review Meta AI responses to teenagers

More than 75 clinicians specializing in teenage mental health reviewed Meta AI responses to hundreds of prompts about suicide and self-harm. Their feedback covered the appropriateness of responses, areas for improvement and how the chatbot handles different levels of risk.

Meta says it is using the findings to prevent conversations from ending too abruptly and to ensure the AI acknowledges a teenager’s feelings while directing them toward offline help.

Dr. Ji-yeon Lee, a Licensed Psychologist in the US and Professor of Counseling Psychology at Hankuk University of Foreign Studies in South Korea, describes the review as extending beyond immediate crisis responses: “It examined not only immediate responses to suicide and self-harm concerns, but also the broader conversational context, appropriate follow-up, and the varying levels of risk that can exist even within high-risk situations.”

Meta is also extending Instagram’s Limited Content setting to conversations with Meta AI. Teen Accounts are automatically assigned a default 13+ content setting, while parents can opt for the stricter alternative, under which the AI will decline a broader range of prompts.

The parent alerts are scheduled to become available to supervising parents worldwide by the end of 2026.