Safety tools for supervised accounts
Meta previewed new parental controls that let adults curb or fully block teens’ chats with AI characters on Instagram, Facebook and Messenger. Supervisors can be notified when the teen starts a conversation with an AI, and set quiet hours or shut the feature off entirely. Teens will see clearer disclosures that they are messaging an AI system and prompts with links to safety resources. Meta says the controls are designed to address risks around maturity, misinformation and unwanted advice in AI exchanges.

How the controls will roll out
The company plans phased availability and says schools and caregivers helped test the designs. Meta also updated its internal guardrails so sensitive queries trigger safer responses and fewer hallucinations. The controls tie into existing Family Center dashboards, allowing granular app-by-app settings while keeping messaging privacy intact for human chats. The big test is enforcement—whether teens can easily switch accounts, and whether safety labels deter risky prompts. Regulators will watch how Meta records, stores and audits teen-AI interactions and how quickly harmful content gets suppressed.