Themes
19 November 2026
AI bots and kids: where to start?
Understanding a brain-teaser of a math problem or asking a question about a favorite animal: your kid probably has big plans for a chatbot. You (hopfully) have some questions. What are they allowed to do already? What stays private? This article is about searching and chatting with AI chatbots. These checks will help you get started together.
1. Check the minimum age
The rules differ from chatbot to chatbot: 13 is usually the minimum age, but not always. Not yet 13? Then you can explore the chatbot together. You ask the questions and look at the answers first. Then discuss them together. Your kid does not chat independently. And yes, this is different from googling, because a chatbot generates its own answer and keeps the conversation going. But you can now encounter AI answers when googling too, so you need to check either way.
2. Ask which AI your child already knows
Maybe you mainly know ChatGPT, while your child comes across Gemini, Copilot, Meta AI or My AI from Snapchat. Try asking: “Which chatbots do you know? And what do you use them for?” Curiosity will get you further than an interrogation – and it is a great way to kick off a few good agreements.
3. Private stays private
What your kid types may be stored and used to train the AI. Good news: a chatbot does not need to know everything to be able to help. So agree together that these details stay private:
-
Last name, address, school or location;
-
Passwords;
-
Personal photos;
-
Private messages and screenshots of conversations;
-
Stories about problems at home, health or other sensitive topics;
-
Personal information about other people.
Make it concrete: “Come up with a fractions exercise” works perfectly without a name, school or photo of the worksheet. And if your child has a question about friendship, there is no need to paste in a private conversation with a classmate.
4. Take a look at the privacy settings together
ChatGPT:
-
Open your profile menu. Using the app? Open the sidebar and tap your profile.
-
Go to Settings (gear icon) → Data Controls.
-
Turn off “Improve the model for everyone.”
New conversations will then not be used to train the AI. They will still remain in your chat history.
Other chatbots:
Good to know: deleting chats is not the same as turning off AI training. And conversations may still be stored even if they are not used for AI training. So: private stays private.
ChatGPT for teens: a tool that provides extra protection for teens aged 13 to 18. It is automatically enabled for accounts that qualify. So you do not have to turn on a “teen mode” yourself. Want to manage settings too? You can link your account to your child’s account through parental controls. Note: a tool like this is never a guarantee of a safe conversation.
5. Sounds good. But is it true?
AI can say something with complete confidence and still be totally wrong. Make checking a habit: “Where did this come from? Can we find it somewhere else too?” Make sure you actually open any source that is cited.
6. A chatbot is not a friend
Does a chatbot still feel like a friend to your kid? Ask what they like about those conversations. Laughing it off or downplaying is forbidden. Explain that a chatbot can respond with understanding, but it does not have feelings and is not a real buddy. So agree on this: use AI for ideas or explanations. Is something bothering you? Then talk to someone you trust. You, for example. Discuss who else your child can turn to.
So actually, it’s very simple
-
Only use AI that you are allowed to use.
-
Keep private information private.
-
Keep thinking for yourself and go to a real person if something’s up