OpenAI is giving teenage ChatGPT users a more restricted version of its AI assistant, combining additional safety measures with tools intended to make the chatbot less useful as a shortcut for homework.
ChatGPT for Teens is designed for users aged 13 to 17 and will also apply to people OpenAI’s age-prediction technology estimates are under 18. The launch builds on the company’s earlier work around parental controls and age detection, but packages those measures into a distinct experience rather than relying entirely on individual settings.
Education is a major part of the approach. Existing features including Study Mode, quizzes and Learning Visualizations are available to teenagers, while two newer systems attempt to change how students interact with ChatGPT when completing schoolwork.
The Responsible Homework Reminder is designed to recognise situations where a teenager appears to be asking ChatGPT to complete an assignment rather than help them understand it. In those cases, the assistant can steer the user toward Study Mode, which works through a problem more gradually instead of immediately supplying an answer.
Study Hours takes that idea further. Teenagers or parents can define periods when Study Mode becomes the default behaviour. During those hours, ChatGPT is designed not to provide direct answers, encouraging students to work through questions instead.
That distinction matters as generative AI becomes increasingly common in education. Chatbots can be useful tutoring tools, particularly when they explain difficult concepts interactively, but the same technology can also complete essays, solve problems and generate coursework with very little effort from the student. OpenAI’s approach effectively acknowledges that simply giving students access to a general-purpose chatbot creates a tension between assistance and substitution.
The teen experience also places tighter boundaries around potentially sensitive conversations. OpenAI says additional protections are enabled for subjects including self-harm, eating disorders, violence, dangerous activities, and sexually explicit or graphic material. ChatGPT is also instructed not to engage teenagers using romantic language, encourage emotional dependency, or present itself as conscious or capable of feelings.
There are smaller safeguards as well. Teen users can receive reminders to take breaks and warnings before uploading sensitive images, while onboarding has been adapted for younger users. Parents who link their accounts can continue using Quiet Hours and selected controls. OpenAI also says notifications for limited high-risk circumstances have been expanded to include situations involving eating disorders.
None of these measures removes the wider questions surrounding teenagers and generative AI. Age-prediction systems are unlikely to be perfect, and educational safeguards depend heavily on how reliably an AI system can understand a student’s intent. There is also a difficult balance between protecting younger users and giving teenagers enough independence to use AI productively.
Still, ChatGPT for Teens signals a shift away from treating younger users as simply another segment of the general ChatGPT audience. As AI assistants become more embedded in schoolwork and everyday life, the bigger test will be whether features such as Study Hours actually encourage learning rather than merely adding another obstacle for determined students to work around.
