OpenAI is expanding protections for teenagers using ChatGPT, introducing a system designed to give users ages 13 to 17 a more age-appropriate experience while giving parents greater control over how the chatbot is used.The changes come as AI chatbots become increasingly common among young people for schoolwork, information, creativity and personal advice. OpenAI says its approach combines age prediction, stronger safeguards around sensitive subjects and parental controls rather than creating a completely separate chatbot for teenagers.
The company has been developing these protections for months. OpenAI has also been working with the American Psychological Association on research into how young people use AI and where additional safeguards may be needed.
What is ChatGPT’s teen experience?
ChatGPT’s teen experience is designed for users between 13 and 17.
Rather than requiring every teenager to manually activate a special mode, OpenAI is rolling out an age-prediction system that estimates whether an account may belong to someone under 18. The system can consider signals associated with account activity and usage patterns.
If ChatGPT is not confident about a user’s age, OpenAI says it will default to the safer teen experience.
The account remains active, meaning younger users can continue to use ChatGPT for learning, creativity and everyday questions. The main difference is that additional safeguards can apply to certain sensitive subjects and interactions.
How does OpenAI determine whether someone is under 18?
OpenAI says age prediction does not rely solely on the date of birth entered when an account was created.
The system can examine a range of signals associated with how an account is used, including:
- General topics discussed with ChatGPT
- Patterns in when the account is used
- How the account is used over time
- How long the account has existed
- Other account-related signals
OpenAI acknowledges that age prediction is not perfect.
If an adult is incorrectly placed into the teen experience, the company allows the person to verify their age. OpenAI’s current age-verification system uses Persona, and OpenAI says it receives the information necessary to establish age rather than the uploaded identification document or selfie itself.
What changes when ChatGPT identifies a teen account?
The teen experience applies additional safeguards to several categories of content.
These include:
- Graphic violence and gore
- Risky viral challenges
- Sexual or romantic roleplay
- Violent roleplay
- Content involving self-harm
- Content promoting extreme beauty standards
- Unhealthy dieting
- Body shaming
The purpose is not to prevent teenagers from discussing difficult subjects altogether. Instead, OpenAI says ChatGPT should respond to sensitive subjects more cautiously and in a way that is more appropriate for younger users.
OpenAI has also said it wants ChatGPT to remain a tool for learning and creativity rather than become a substitute for real-world relationships.
What can parents control?
Parents or guardians can link their account to a teenager’s ChatGPT account through parental controls.
Once accounts are linked, parents can manage a range of settings, including:
- Quiet hours
- Voice Mode
- Image generation
- Memory
- Model-improvement settings
- Study Mode
- Sensitive-content protections
Quiet hours allow parents to establish a period during which ChatGPT cannot be used from the linked teen account.
Parents cannot use parental controls to read their teenager’s conversations.
OpenAI specifically says that linking accounts does not give a parent or guardian access to the teen’s chat history.
Can parents receive safety alerts?
Yes, but the notifications are designed to be limited rather than functioning as continuous monitoring.
OpenAI says parents can receive notifications in certain situations where its systems detect indications that a teenager may need urgent support.
The company has expanded these notifications over time and says they are intended for narrow safety circumstances rather than ordinary conversations. They are not designed to notify parents simply because a teenager discusses politics, news, fictional writing or everyday frustrations.
This distinction is important because parental controls are intended to provide safety oversight without turning a teenager’s account into an account that parents can routinely monitor.
What is Study Mode for teenagers?
Study Mode is designed to make ChatGPT more useful as a learning tool rather than simply an answer generator.
OpenAI has made Study Mode available as a parental-control setting. When enabled by a parent, new main ChatGPT conversations can start with Study Mode switched on.
The approach is consistent with OpenAI’s broader push to position ChatGPT as a learning aid.
Instead of simply providing an answer to a homework question, Study Mode is intended to guide students through problems and encourage them to understand the material.
Parents can therefore choose whether Study Mode should be enabled by default for their teenager.
Will teenagers still be able to use ChatGPT normally?
Teenagers can continue using ChatGPT for a broad range of activities.
The safeguards primarily affect how the system handles certain sensitive topics and interactions.
OpenAI says teen users can still:
- Ask questions
- Learn new subjects
- Work on school assignments
- Write and brainstorm
- Explore ideas
- Create content
- Use ChatGPT for everyday problem-solving
The experience is therefore not intended to be a restricted educational chatbot.
Instead, it is a version of ChatGPT with additional protections around areas where younger users may face greater risks.
What happens when a teen turns 18?
The teen experience is not permanent.
When OpenAI identifies an account holder as 18 or older, teen protections can be removed and the account can transition to the adult experience.
OpenAI says eligible users are notified before the transition, and if the account is connected to a parent or guardian, that parental-control connection ends when the account becomes an adult account.
Adults who have been incorrectly placed in the teen experience can also verify their age rather than waiting for the system to correct itself.
Why is OpenAI introducing these protections now?
The changes come as AI becomes increasingly integrated into teenagers’ daily lives.
Young people are already using AI systems to search for information, study, write, create and seek advice.
That creates potential benefits but also new concerns.
A teenager may treat an AI system as an authoritative source even when the system is wrong. There are also concerns about young users turning to conversational AI during moments of emotional distress.
OpenAI has acknowledged those challenges and has been working with mental-health experts and organizations including the American Psychological Association to better understand how young people interact with AI.
Why is age detection controversial?
Age detection creates an unavoidable trade-off.
To provide different experiences to teenagers and adults, an AI company needs some way to determine a user’s age.
But estimating someone’s age from behavioral signals can produce mistakes and raises legitimate privacy questions.
OpenAI acknowledges that its prediction system can get someone’s age wrong. Adults who are incorrectly classified can use age verification to have the teen protections removed.
The company says its approach is intended to minimize the amount of sensitive information it receives during verification. For example, Persona handles the identity document or selfie used for verification, while OpenAI says it receives only the information needed to establish age.
Are these safeguards foolproof?
No.
OpenAI itself says safeguards are not perfect and can potentially be bypassed by people deliberately attempting to evade them.
That means age prediction and content safeguards should not be viewed as substitutes for parental involvement, education or responsible use.
The company has repeatedly encouraged parents to talk with teenagers about what AI can and cannot do, particularly when AI is used for sensitive or emotionally difficult subjects.
What does this mean for parents?
For parents, the new system provides more control without giving them unrestricted access to their teenager’s conversations.
The most useful controls are likely to be the practical ones: setting quiet hours, restricting image generation or voice use, enabling Study Mode and deciding whether memory should be available.
The safety notifications are designed for more serious circumstances.
Parents should also understand that AI safeguards cannot replace human judgment. A chatbot may be useful for explaining algebra, brainstorming an essay or helping a student study, but it should not become the only source of support for a teenager dealing with serious emotional, medical or safety concerns.
What does this mean for the future of AI and children?
OpenAI’s teen protections reflect a broader shift in how technology companies are approaching AI products for younger users.
The central question is no longer simply whether teenagers should be allowed to use AI.
Millions already do.
The harder question is how these systems should behave when the person using them is still developing emotionally and cognitively.
OpenAI’s answer is to combine age prediction, content safeguards, parental controls, study-oriented features and limited safety notifications.
Whether that combination is effective will depend on how accurately the company can identify ages, how well its safeguards perform in difficult situations and how families use the controls available to them.
For now, the direction is clear: ChatGPT is moving toward an experience in which age is increasingly important to how the system responds.