OpenAI launches ChatGPT for Teens with learning and safety protections

OpenAI has launched ChatGPT for Teens, a dedicated experience for users aged 13 to 17. According to the company's August 18 announcement, people who report an age in that range—or whom its age-estimation system identifies as under 18—are placed into the teen experience automatically. The release brings learning tools, stronger default safeguards and additional family controls into one product configuration rather than leaving them as separate settings.

For learning, the service combines Study Mode with new responsible-homework reminders that can redirect attempts to obtain quick assignment answers toward step-by-step problem solving. Quizzes and Learning Visualizations remain available, while new Study Hours let teens or linked parents schedule periods when Study Mode becomes the default. OpenAI presents the product as support for learning outside the classroom; it is distinct from the institution-managed ChatGPT for Teachers offering.

The teen version also applies age-appropriate protections in areas including self-harm, violence, eating disorders, dangerous activities and sexual or graphic content. Linked parents can already set Quiet Hours, manage selected features and receive limited notifications in high-risk situations; OpenAI says eating-disorder-related notifications are being added. Other additions include break reminders, warnings before sensitive-image uploads and onboarding that explains available safety features. OpenAI also says its under-18 behavior rules prohibit romantic language, encouragement of emotional dependency, and claims that ChatGPT has feelings or consciousness.

For families, the automatic placement is significant because protection no longer depends only on a teenager or parent finding the right controls. It also raises practical questions: age estimation can make mistakes, automated safeguards are not perfect and parental alerts are intentionally limited rather than full conversation monitoring. Users should therefore treat the system as an extra layer, not a substitute for teachers, parents or professional support.

For makers and policymakers, this release turns child safety from a policy document into a product design. It offers a concrete pattern—age-aware defaults, learning-oriented interactions and graduated controls—but its quality will depend on measured outcomes and transparent evaluation. OpenAI says it is beginning to publish teen-specific system-card tests. That evidence will matter more than launch claims when schools and families judge whether the experience is genuinely safer and educationally useful.