Overview
- Parents and teens can link accounts to apply teen-specific filters that reduce graphic content, romantic or sexual role-play, viral challenges and extreme beauty ideals.
- Parents can set quiet hours, disable voice and image generation, turn off memory, and opt teen accounts out of having chats used to train OpenAI’s models.
- In rare cases of suspected acute distress, automated systems flag conversations for human review and parents may be notified, but chat transcripts remain private.
- Protections require teen consent and can be unlinked at any time with a parent notification, and ChatGPT still works without sign-in or age checks, creating bypass risks.
- The rollout starts on the web with mobile support coming, and OpenAI is developing age prediction and says sensitive chats may be routed to specialized reasoning models.