Social media giant Meta has launched a system that alerts parents if their teenager discusses suicide or self-harm with its Artificial Intelligence (AI) chatbot.
The feature expands upon current protocols where the company directs young users to crisis hotlines or advises them to speak with an adult when self-harm indicators appear in chat logs.
The notification system functions through Instagram parental supervision tools. It is currently active for users in the United States (US), United Kingdom (UK), Australia and Canada, with global deployment scheduled by the end of the year.
The system operates via the Instagram parental supervision feature, which requires guardians to opt in. Accounts for new users under the age of 16 are automatically set to Teen Accounts by default, requiring parental permission to change restrictions.
Users aged 16 and 17 can alter these settings independently unless an adult actively supervises the account.
To identify conversations involving self-harm, Meta deploys an automated system. Due to the performance history of the technology, the company states that human moderators review flagged conversations before notifications are distributed.
Meta stated that the company worked with parents and experts to determine which interactions require notification, including instances where individuals make subtle references to self-harm. The platform stated that it will err on the side of caution and alert parents if the intent of a teenager remains ambiguous.
Guardians receive the alerts via app notifications alongside a separate email, text or WhatsApp message, depending on the contact details provided to the platform. The alert also includes resources and suggestions developed by experts on how to approach conversations regarding suicide and self-harm with children.
The policy arrives as competing platforms face legal scrutiny regarding automated systems.
OpenAI faces lawsuits alleging that its ChatGPT chatbot encouraged users to commit suicide or murder.
Reports indicate that approximately 1.2 million individuals discussed suicide planning or intent with ChatGPT within a specific timeframe, representing 0.15 percent of weekly active users.
Meta is also developing an additional safety protocol for both teenagers and adults. The upcoming feature allows the company to contact emergency services for wellness checks if a chatbot conversation indicates imminent self-harm.
The company has not announced a launch date for the tool, though it already monitors public posts and search queries on Facebook and Instagram for similar risks. Meta stated that it made over 19,000 such emergency referrals globally last year.
Meanwhile, Advocacy groups and medical professionals expressed varying views on the update. Brendan Bouffard, a staff attorney for the children’s advocacy non-profit Fairplay, stated that the announcement should be greeted with skepticism.
Bouffard cited 2025 research from the organisation, alleging that safety tools within Meta Teen Accounts were ineffective or non-existent.
While Meta stated that the research misrepresented the functionality of its safety mechanisms, Dr John Ackerman, clinical manager for the Center for Suicide Prevention and Research at Nationwide Children’s Hospital, stated that monitoring tools can assist parents in identifying concerns early and initiating supportive dialogue.
