Meta adds a seven-day topic log for teen AI chats and expands parent alerts

Meta is moving AI supervision from vague reassurance to auditable signals. Parents who supervise Teen Accounts can now see the topics their teens asked Meta AI about during the last seven days, starting in the US, UK, Australia, Canada and Brazil. That is a meaningful shift because AI use inside social apps has been largely invisible to parents even when the same apps already expose messaging and screen-time controls.

The new Insights tab appears across Facebook, Messenger and Instagram, on web and in-app. Meta says parents will see topic areas such as School, Entertainment, Lifestyle, Travel, Writing and Health and Wellbeing, and can drill into subcategories inside each bucket. The company also says the topic can still appear even when Meta AI refuses to answer a question. That detail matters. It makes the supervision layer about attempted use, not only successful responses, which is where many safety systems lose context.

Meta is pairing the topic log with a stricter policy story. The company says teen AI experiences were shaped by 13+ movie-rating criteria and parent feedback, and it is building separate alerts for conversations related to suicide and self-harm. It also said the number of US teens enrolled in supervision has more than doubled since last year. Taken together, those details suggest Meta wants parental controls to become an AI governance surface rather than a buried settings page that families rarely open.

Two additions stand out beyond the product UI. Meta worked with the Cyberbullying Research Center on conversation starters for parents, and it formed an AI Wellbeing Expert Council with members affiliated with the National Council for Suicide Prevention, the University of Michigan, the University of Texas and the University of Southern California. The next question is whether families use the seven-day topic history as a real intervention tool or treat it as another dashboard nobody checks.