MENLO PARK, Calif. (Diya TV) — Meta has introduced a system that alerts parents when a supervised teenager’s conversation with Meta AI indicates a possible risk of suicide or self-harm.
The alerts are now available to parents using Instagram’s parental supervision tools in the United States, United Kingdom, Australia and Canada. Meta plans to make the feature available globally by the end of 2026.
Meta AI already directs teenagers who express thoughts of suicide or self-harm to crisis helplines and encourages them to contact a parent, counselor or another trusted adult. The new system adds a separate notification for the supervising parent when Meta determines that the teenager may face a safety risk.
Meta said it developed a specialized artificial intelligence system to identify conversations that include direct or subtle references to self-harm. A human reviewer will examine every conversation flagged by the system before the company sends an alert.
The company acknowledged that the system may produce false alarms. Meta said reviewers will notify parents when a teenager’s intent remains unclear rather than risk overlooking a potentially serious situation.
Parents who receive an alert will also receive resources developed with experts to help them discuss suicide and self-harm with their teenager. Meta has not publicly provided detailed performance data showing the system’s accuracy or the expected rate of incorrect alerts.
The feature only applies when a parent has established supervision through Instagram. It does not automatically notify the parents of every teenager who uses Meta AI.
The alerts expand controls Meta introduced earlier in 2026. Instagram began notifying supervising parents when a teenager repeatedly searched for suicide- or self-harm-related terms within a short period. Meta also began allowing parents to view general topics their teenagers had discussed with Meta AI during the previous seven days.
Meta said it consulted more than 75 clinicians who specialize in adolescent mental health. The clinicians reviewed hundreds of sample responses involving suicide and self-harm and evaluated whether Meta AI responded appropriately to teenagers.
Based on that feedback, Meta said it is modifying the chatbot to acknowledge a teenager’s feelings before directing the user to offline support. The company said the goal is to avoid responses that end sensitive conversations too abruptly while continuing to encourage contact with qualified people and emergency resources.
Meta is also developing a system that could contact emergency services when a conversation involving an adult or teenager indicates an imminent risk of suicide. That capability remains under development and is separate from the parental alert system that launched Thursday.
The company already contacts emergency services when it identifies Facebook or Instagram posts that indicate a credible and immediate suicide risk. Meta said it made more than 19,000 such referrals worldwide in 2025. That figure comes from the company and has not been independently verified.
The announcement also extends Meta’s Limited Content setting to conversations with Meta AI. Parents can activate the setting through Instagram’s supervision tools to make the chatbot decline a broader range of prompts.
Meta automatically places users under 18 into its standard Teen Account setting, which the company describes as appropriate for users 13 and older. Under that setting, Meta AI is designed to reject sexual or romantic interactions with teenagers and refuse requests involving subjects such as alcoholic drink recipes. The Limited Content option applies stricter restrictions.
The alerts create a balance between safety monitoring and teen privacy. Meta said the company consulted parents and experts when deciding which conversations should trigger a notification. However, the company has not released detailed information about how long flagged conversations remain stored, how many employees can access them during review or how teenagers are informed when their conversation prompts a parental alert.
The system also depends on Meta correctly identifying a user as a teenager and the parent activating supervision. Teenagers who enter an inaccurate birth date or use an account without parental supervision may not receive the same protections.
Meta’s changes come as technology companies face increased scrutiny over how conversational AI systems interact with minors, particularly during mental health crises. Researchers, parents and child-safety organizations have raised questions about whether chatbots can recognize distress consistently, respond without reinforcing harmful behavior and direct vulnerable users toward qualified human assistance.
Meta has not presented the parental alerts as a substitute for professional mental health care. The feature instead serves as an intervention tool intended to inform a supervising parent when a teenager’s conversation may require immediate attention.
