We’ve updated the Model Spec, our living document outlining intended model behavior, to strengthen and more clearly codify the principles that reflect how we build experiences for teen users.
New section: Under-18 (U18) Principles
ChatGPT’s new Under-18 (U18) Principles builds on the existing safety rules that apply to all users, adding age-appropriate guidance where appropriate for the developmental needs of teens, aged 13-17. This update clarifies how those rules are intended to apply in teen conversations, recognizing that teens benefit from clearer boundaries, reduced exposure to potentially harmful content and stronger real-world support when risks arise. The assistant should meet teens where they are, engaging with them in a respectful and transparent manner, while refusing to participate in self-harm, sexualized or violent immersive roleplay, dangerous activities, substance misuse or any efforts to conceal harm. When credible risks arise, the model should prioritize prevention and early interventions, offer safer alternatives and encourage involvement of parents, guardians and other trusted adults or professionals – making clear that AI can provide guidance and information, but cannot replace real-world care.
Other updates
This release also includes minor edits and clarifications for consistency and readability throughout the document.
More information can be found inthisblog post, and the latest version of the Model Spec is available atmodel-spec.openai.com.