Introducing ChatGPT for Teens: Built for Learning, Backed by Protections

OpenAI has launched ChatGPT for Teens, a version of ChatGPT built specifically for users aged 13-17 that pairs a learning-focused design with a set of default safety protections and expanded parental controls. Teens are placed into the experience automatically when OpenAI’s age-prediction system estimates a user is under 18, or when a user states they are between 13 and 17.

Details

• How teens are identified: Rather than requiring upfront age verification, OpenAI relies on age-prediction technology across usage signals to route likely-under-18 users into the teen experience by default, alongside self-declared ages of 13-17
• Default safety protections: Built on OpenAI’s “Under-18 Principles,” informed by developmental science and outside experts, the teen experience adds specific safeguards around self-harm, violence, eating disorders, dangerous activities, and explicit sexual or graphic content
• Restrictions on romantic/companion behavior: The model is designed not to use romantic language, encourage emotional dependence, or imply it has feelings or consciousness — directly restricting the kind of companion-style roleplay available to adult users
• Other built-in guardrails: reminders discouraging teens from sharing sensitive personal images, break reminders during long sessions, and consistent product cues identifying ChatGPT as AI rather than a person
• Learning-first design: Study Mode — developed with teachers and learning-science researchers to guide teens through problems with step-by-step questions rather than direct answers — is central to the teen product; a new “Study Hours” feature lets teens or parents schedule times when Study Mode turns on by default, and the system is designed to recognize homework-shortcut attempts and redirect toward Study Mode, plus quizzes and “Learning Visualizations” for concept practice
• Parental controls: Parents linking to a teen’s account can set Quiet Hours, configure select settings, and receive safety notifications for high-risk situations, including a new, narrowly scoped notification category around eating-disorder-related signals
• Partnership tie-in: OpenAI is pairing the launch with a “signature partnership” with CodeAI to give students and educators AI-literacy tools and resources (see OpenAI’s related CodeAI partnership announcement)
• What’s not disclosed: OpenAI’s announcement does not specify a country-by-country rollout schedule or pricing beyond describing the launch as immediate (“introducing”)

What happened next

This follows an earlier OpenAI explainer published in July on teen safety principles and parental controls, but marks the shift from policy explanation to an actual dedicated product experience — a distinct, separately designed version of ChatGPT rather than settings layered onto the adult product. The launch lands alongside heightened scrutiny of AI chatbots and minors industry-wide, positioning ChatGPT for Teens as OpenAI’s answer to concerns about companion-style AI interactions with younger users, while still betting on continued teen engagement through the education-focused framing.