OpenAI announced on August 18, 2026 local time, the launch of ChatGPT for Teens for users aged 13 to 17. The service builds on the existing ChatGPT and adds learning support features and age-appropriate safety measures by default. It is automatically applied when the system estimates that a user is under 18, or when a user reports being 13 to 17, and will roll out gradually to eligible accounts on individual ChatGPT plans, including Free.

ChatGPT for Teens reduces use cases where it simply provides homework answers and strengthens support that helps users think through problems and reach answers themselves. In Study Mode, it supports learning through guiding questions, step-by-step hints, and checks for understanding. A new Responsible Homework Reminder recognizes signs that a user may be trying to shortcut an assignment and encourages use of Study Mode, where problems are solved step by step. The service also supports quizzes that check knowledge and Learning Visualizations that present difficult concepts visually. Users and parents can add Study Hours, which turn on Study Mode by default during specific time periods. OpenAI says the product can be used not only for schoolwork, but also for creative work, exploring new ideas, and everyday questions.

Accounts believed to belong to users under 18 are identified not only through user self-reporting but also through ChatGPT's age prediction system. Age prediction may refer to multiple signals, including the general topics a user talks about, the times of day they use ChatGPT, how and when the account is used, and how long it has been since the account was created. ChatGPT for Teens is automatically applied to eligible accounts estimated to belong to users under 18, so users do not need to turn it on themselves. Users aged 18 or older who are mistakenly included can remove teen protections through age verification.

Safety measures apply protections for users under 18 in high-risk areas such as self-harm, violence, eating disorders, dangerous behavior, and explicit sexual content. When a user has been using the service for a long time, ChatGPT prompts them to take a break and reminds them again that ChatGPT is AI. It also shows a warning when a user tries to upload a sensitive image. OpenAI says it will restrict romantic or sexual role-play with teens and prevent ChatGPT from using romantic expressions, encouraging emotional dependence from users, or suggesting that it has feelings or consciousness.

Parents can optionally connect to a teen's account and manage some settings. They can set Quiet Hours, when ChatGPT cannot be used, and Study Hours, when Study Mode is enabled, and they can receive safety-related notifications in certain high-risk situations. Parents cannot view a teen's conversations, chat history, or real-time usage. OpenAI says that even in safety notifications, it shares only limited information needed for support and does not share full conversation text. The rollout began on August 18 for eligible accounts worldwide and will expand over the next few weeks. Available features may vary by region.

OpenAI announced on August 18, 2026 local time, the launch of ChatGPT for Teens for users aged 13 to 17. The service builds on the existing ChatGPT and adds learning support features and age-appropriate safety measures by default. It is automatically applied when the system estimates that a user is under 18, or when a user reports being 13 to 17, and will roll out gradually to eligible accounts on individual ChatGPT plans, including Free.

ChatGPT for Teens reduces use cases where it simply provides homework answers and strengthens support that helps users think through problems and reach answers themselves. In Study Mode, it supports learning through guiding questions, step-by-step hints, and checks for understanding. A new Responsible Homework Reminder recognizes signs that a user may be trying to shortcut an assignment and encourages use of Study Mode, where problems are solved step by step. The service also supports quizzes that check knowledge and Learning Visualizations that present difficult concepts visually. Users and parents can add Study Hours, which turn on Study Mode by default during specific time periods. OpenAI says the product can be used not only for schoolwork, but also for creative work, exploring new ideas, and everyday questions.

Accounts believed to belong to users under 18 are identified not only through user self-reporting but also through ChatGPT's age prediction system. Age prediction may refer to multiple signals, including the general topics a user talks about, the times of day they use ChatGPT, how and when the account is used, and how long it has been since the account was created. ChatGPT for Teens is automatically applied to eligible accounts estimated to belong to users under 18, so users do not need to turn it on themselves. Users aged 18 or older who are mistakenly included can remove teen protections through age verification.

Safety measures apply protections for users under 18 in high-risk areas such as self-harm, violence, eating disorders, dangerous behavior, and explicit sexual content. When a user has been using the service for a long time, ChatGPT prompts them to take a break and reminds them again that ChatGPT is AI. It also shows a warning when a user tries to upload a sensitive image. OpenAI says it will restrict romantic or sexual role-play with teens and prevent ChatGPT from using romantic expressions, encouraging emotional dependence from users, or suggesting that it has feelings or consciousness.

Parents can optionally connect to a teen's account and manage some settings. They can set Quiet Hours, when ChatGPT cannot be used, and Study Hours, when Study Mode is enabled, and they can receive safety-related notifications in certain high-risk situations. Parents cannot view a teen's conversations, chat history, or real-time usage. OpenAI says that even in safety notifications, it shares only limited information needed for support and does not share full conversation text. The rollout began on August 18 for eligible accounts worldwide and will expand over the next few weeks. Available features may vary by region.