Many teenagers already use ChatGPT for homework, coding practice, college essays and, honestly, for advice they would not ask an adult for. So the real question for parents is not "should my child use AI?" It is "how do I make sure they use it safely?"
Here is what ChatGPT's rules, protections and parental controls actually do today, checked against OpenAI's own pages, and what I would set up first.
The age rules, in plain words
OpenAI's Terms of Use say you must be at least 13 years old (or the minimum age to consent in your country), and anyone under 18 needs a parent or legal guardian's permission. OpenAI's help centre is even more direct: ChatGPT is not meant for children under 13, and if it is used in a classroom with children under 13, an adult must be the one actually using it.
Other tools have different rules. Anthropic's consumer terms say Claude users must be at least 18. So if your 15 year old says "everyone uses Claude", that is not an account they are meant to have.
What ChatGPT already does for under-18s
When ChatGPT believes an account belongs to someone under 18, it automatically switches on ChatGPT for Teens. It decides this from the age on the account, a verified age, or its own age prediction, which looks at signals such as the topics discussed, the times of day the account is used and how long the account has existed.
In the teen experience, OpenAI says ChatGPT handles some topics with extra care and reduces content such as:
- Graphic violence or gore
- Viral challenges that could push risky behaviour
- Sexual, romantic or violent roleplay
- Content that promotes extreme beauty standards, unhealthy dieting or body shaming
OpenAI also says it will not show ads in teen accounts.
Age prediction is not perfect, and OpenAI says so itself. That cuts both ways. An adult can be wrongly placed in the teen experience, and a teen who lies about their age may not always be caught. This is why linking your account matters.
What changed on October 30, 2026
OpenAI announced new limits for teens and all users on October 30, 2026:
- For under-18s, ChatGPT will restrict practical information about weapons and security systems that could be misused, even when the question looks harmless and there is no sign of violent intent.
- For all users, ChatGPT can end chats where someone seems to be fixating on violence, and OpenAI may temporarily pause accounts after repeated chats like that. It is also limiting detail about past mass attacks, including the weapons used and how they were obtained.
- Parents will be notified if their linked teen's account is paused for repeatedly discussing violence.
- OpenAI said it made more than 200 law enforcement referrals in Q3 2026 for imminent threats of violence, and it will publish its first Community Safety Report in early 2027.
OpenAI said these safeguards would roll out over the following few weeks, so some may not be live on every account yet.
What parental controls let you do
Parental controls link your ChatGPT account to your teen's. Either of you can send the invite from Settings, then Parental controls. Once linked, you can manage these settings for your teen:
| Setting | What it does | Default |
|---|---|---|
| Reduce sensitive content | Extra safeguards against age-inappropriate content | On |
| Quiet hours | Times when your teen cannot use ChatGPT at all | Off |
| Study mode and study hours | New chats start in study mode, which guides instead of giving answers | Off |
| Improve the model for everyone | Whether your teen's chats can be used to train models | On |
| Reference saved memories | Whether ChatGPT personalises answers using memories | On where available |
| Voice Mode | Whether your teen can talk to ChatGPT by voice | On |
| Image generation | Whether your teen can create and edit images | On |
You may also get a safety notification by email, SMS or push in limited serious situations, for example when trained reviewers identify a serious self-harm concern, or when your teen's account is deactivated for conversations about violence.
What parental controls do NOT do
This is the easiest part to misunderstand, so I want to be very clear:
- You cannot read your teen's chats. OpenAI says parental controls do not give access to conversations, chat history or real-time activity, and even a safety notification does not include the chat.
- Your teen can unlink. Either of you can unlink at any time, though you are notified if your teen does it.
- It is not real-time monitoring. OpenAI says safety notifications may not catch every concern and do not replace professional care or emergency services.
- It ends at 18. When ChatGPT identifies the account as 18 or older, parental controls and teen protections are removed automatically.
So settings help, but they do not replace a conversation.
My 15 minute setup checklist for parents
- Make sure your teen has their own account, not yours. Your account has your memories, your history and no teen protections.
- Link accounts from Settings, then Parental controls, then Add family member.
- Keep Reduce sensitive content on.
- Turn on Quiet hours for school nights, for example from 10 pm to 7 am.
- Consider Study hours after school, so homework chats start in study mode and your teen has to think instead of copy.
- Turn off Improve the model for everyone if you do not want your teen's chats used for training.
- Have the "three rules" talk: never share passwords, Aadhaar, addresses or photos of friends; double check anything about health, money or law with a real person; and come to you if a chat ever feels wrong. If you want the privacy side in more depth, read my guide on whether your ChatGPT and Claude chats are private.
The conversation that matters more than settings
Teens will find workarounds for any setting. What they cannot work around is a parent who understands the tool. Sit with your teen and use ChatGPT together for 20 minutes. Ask it to explain a chapter they find hard, or to quiz them before an exam. You will learn how they use it, and they will see you are curious, not suspicious.
The same goes for teachers. If you teach, my guide on AI for teachers and colleges covers how to bring it into the classroom safely.
If you want to understand these tools well enough to guide your family, or your own team at work, that is exactly what my 1:1 programs at The Arthi AI Collective are built for. Explore the programs.