Meta is beginning to roll out a system that can alert parents when a teenager’s conversation with Meta AI indicates a possible risk of suicide or self-harm. The feature is designed for families using Instagram’s parental supervision settings.
The change adds a new layer to parental controls that have largely focused on general activity within Instagram. It brings AI assessment and human review into decisions about whether a parent or guardian should be notified.
Human Review Comes Before an Alert
Meta says it will use dedicated AI models to identify conversations involving suicide or self-harm among teenagers. A conversation flagged by the system will not immediately generate a notification for a parent.
Each detected case will be reviewed manually to reduce the chance of mistaken alerts that could unnecessarily alarm families. However, Meta says it may still send an alert when uncertainty remains about a user’s intent after that review.
Parents or guardians who receive a notification will also be given guidance and resources from experts. The materials are intended to help adults respond to a sensitive situation and begin a more appropriate conversation with the teenager.
| Feature | Availability | Status |
|---|---|---|
| Meta AI conversation risk alerts | United States, United Kingdom, Australia, Canada | Beginning rollout |
| Meta AI conversation risk alerts | Global parental supervision users | Targeted before the end of 2026 |
| Emergency service contact option | High-risk teens and adults | In development |
Initial Rollout Covers Four Countries
The alert system is initially being introduced to supervised Instagram users in the United States, the United Kingdom, Australia, and Canada. Meta aims to make it available globally to all parental supervision users before the end of 2026.
The update arrives amid growing scrutiny of how AI chatbots respond when users are experiencing a mental health crisis. Teen protection has drawn particular attention because young users may turn to AI to discuss deeply personal issues.
Before introducing parental alerts, Meta AI had already been designed to direct teenagers expressing suicidal thoughts or self-harm concerns toward crisis support services. The chatbot is also intended to encourage users to speak with a parent or another trusted adult.
Limited Content Extends to AI Conversations
Meta is also expanding Limited Content protections to teenagers’ interactions with Meta AI. When enabled by a parent, the setting gives the chatbot more limited responses that are adjusted to the user’s age.
Limited Content was previously used to filter sensitive material on Instagram, including adult visuals and content related to self-harm. The expansion applies similar restrictions when teenagers use the platform’s AI chatbot.
In April 2026, Meta began giving parents access to categories of topics teenagers discussed with Meta AI during the previous seven days. The setting does not reveal detailed conversation content, and it includes guidance developed with experts to support open family discussions about AI use.
Emergency Contact Capability Remains in Development
Meta is developing an option to contact emergency services when a Meta AI conversation suggests a person faces a high and immediate risk of suicide. The planned capability would apply not only to teenagers but also to adult users considered to face a serious safety threat.
The company says the system is being developed with parents, more than 75 mental health professionals, and a teen safety advisory group. Those participants are helping determine which conversations are serious enough to trigger a parent or guardian alert.
Meta announced in February 2026 that it also planned notifications when teenagers repeatedly searched Instagram Search for terms related to suicide or self-harm. The newly rolling out AI conversation alerts put that earlier plan into practice.