OpenAI has begun rolling out an age prediction system for ChatGPT to identify accounts likely belonging to users under 18 and apply additional safety features.

As per an announcement by OpenAI, the new system will estimate a user's age by analysing behavioural and account-level signals, including account age, typical activity times, usage patterns, and a user's stated age.

If an account is flagged as likely underage, ChatGPT will automatically restrict access to sensitive content categories such as graphic violence, sexual role-play, depictions of self-harm, and content promoting extreme beauty standards or harmful viral challenges.

Users incorrectly placed in the under-18 experience can verify their age through a selfie using the Persona identity service to restore full access. Parents also have the option to use additional parental controls, such as setting quiet hours or disabling certain features.

The rollout is underway for consumer plans, with implementation in the European Union scheduled for the coming weeks to comply with regional requirements. OpenAI stated the approach is informed by child development experts and that it will continue to refine the model's accuracy.