OpenAI has presented ChatGPT for teenagers, a new experience of its chatbot designed specifically for users aged 13 to 17. The system incorporates additional protections against sensitive content, tools to promote learning, and controls that allow parents to limit certain functions or set usage schedules.
This is not a different application or a new independent chatbot. ChatGPT will automatically activate this experience when the user states that they are between 13 and 17 years old or when OpenAI's age prediction systems estimate that the account belongs to a person under 18 years old.
The company claims that its goal is to help teenagers "learn, think critically, deepen their understanding, and use artificial intelligence confidently", while preventing the chatbot from becoming a substitute for teachers, family members, or personal relationships.
How ChatGPT for teenagers works
The teenage version maintains the general capabilities of ChatGPT, but modifies the way the system responds to certain content and requests.
One of its main innovations is the reinforcement of the study mode, which uses guided questions, step-by-step explanations, and knowledge checks to help the student understand a subject.
If a teenager tries to directly obtain the solution to an assignment, ChatGPT may show a reminder and guide them towards study mode. Instead of simply providing the answer, the system will try to engage the user in solving the problem.
The experience also incorporates: questions and step-by-step explanations, quizzes to check knowledge, educational visualizations of complex concepts, reminders related to homework, schedules in which study mode is activated automatically, alerts to take breaks after prolonged periods of use, reminders that ChatGPT is an artificial intelligence, and warnings before uploading certain private or sensitive images.
OpenAI argues that artificial intelligence should complement learning and not serve to evade it. The company believes that these tools can especially help students who do not have support at home when they encounter difficulties with a subject.
More protection against self-harm, violence, and sexual content
ChatGPT for teenagers incorporates by default reinforced measures to reduce exposure to potentially harmful or inappropriate content for their stage of development.
The restrictions especially affect conversations related to self-harm and suicide, violence and graphic content, eating disorders, viral challenges or dangerous activities, age-restricted products and services, explicit sexual content, romantic, sexual or violent role-playing games, as well as extreme beauty standards and harmful diets.
The new experience also introduces limits on the emotional behavior of the chatbot. ChatGPT should not use romantic language with teenagers, promote emotional dependency, or imply that it has feelings or consciousness.
OpenAI points out that these measures are designed to reinforce relationships in the real world and prevent minors from perceiving artificial intelligence as a person or as a substitute for their family and social environment.
What parents will be able to control
Parents or guardians will be able to link their own ChatGPT account with that of the teenager. The connection will not be automatic: it must be initiated by an invitation and accepted by the other account.
Once linked, parents will be able to manage certain options, including: setting "quiet hours" during which the teenager will not be able to use ChatGPT, scheduling study hours, activating or deactivating voice mode, controlling image generation, managing access to certain ChatGPT functions, deciding whether conversations can be used to improve models, managing the use of saved memory, and receiving alerts in certain serious risk situations.
Study hours and quiet hours work differently. The former cause new conversations to start in study mode, but do not prevent access to ChatGPT. Quiet hours, on the other hand, block access during the scheduled periods.
OpenAI has also announced that it will expand security notifications to include certain cases related to eating disorders.
Parents will not be able to read conversations
That said, parental controls will not allow parents to read conversations, check the adolescent's history, or know their activity in real time. OpenAI may send a notification in limited situations where it detects a serious concern for the minor's safety. These alerts may include an overview of the risk and help resources, but will not include the transcript of the conversation.
The company also warns that these alerts do not constitute a real-time surveillance system, may not detect all cases, and do not replace professional care or emergency services.
Both the adolescent and the parent or guardian can unlink the accounts at any time. If the minor does so, the linked adult will receive a notification.
How ChatGPT knows that a user is a minor
ChatGPT will be able to activate the experience for adolescents using three types of information: the age declared when creating or setting up the account, a previously verified age, and OpenAI's age prediction systems.
If the system considers that an account belongs to a person under 18 years old, it will automatically apply the corresponding protections.
Adults mistakenly classified as adolescents can request an age verification to remove the restrictions. OpenAI uses the services of Persona for this procedure, an external company responsible for verifying the documentation or image provided by the user.
When ChatGPT identifies that the account holder has turned 18, it can remove the protections for adolescents and end the linkage with the parental account.
When ChatGPT will be available for adolescents in Spain
OpenAI has begun the rollout of ChatGPT for adolescents on August 18, 2026. The experience is being implemented globally among eligible accounts of free and paid personal plans.
The company has not officially stated that the European Union or Spain are excluded from the launch but does warn that the rollout will be gradual over the coming weeks and that the availability of some tools may vary by region.
The only temporary exception explicitly mentioned by OpenAI is Australia, where full availability is expected on September 8.
Thus, Spanish adolescents may not yet find the new features in their account.
ChatGPT is still not intended for children under 13 years old. OpenAI also states that users between 13 and 18 years old must have the consent of their parents or guardians to use the service.