Session 1 of 8
See it. Try it. Understand it.
Ask your child: "What do you think AI is? Have you heard the word before?" Let them answer freely โ no right or wrong here. Just listen.
Now say: "Let's go ask the AI the same question and see what it says."
Open ChatGPT (chatgpt.com) or Claude (claude.ai) on your device. You type โ your child watches.
Read the answer together. Ask your child: "Does that make sense? Did it say anything surprising?"
Read together. Ask: "Do you already know any of these facts? Are any of them surprising?"
This one is theirs. They see that they can ask about anything they want.
Ask this:
If AI gets it right (it should be 20), try a harder one: ask it to count all the letters in a long sentence. AI often miscounts. Show your child that even a very clever tool can make simple mistakes. This is the most important thing they will learn today.
AI conversations are stored by the companies that make them. Your child's real name, school, or any personal detail should never go into a chat. If the AI asks 'what is your name?' โ use a made-up name. Always.
Talked to an AI for the very first time
Today was about first contact โ seeing what AI is by using it, not reading about it. The most important moment is the 'find a mistake' activity: it plants the seed that AI is not infallible.
Watch for signs of over-excitement ('it knows EVERYTHING!') โ gently redirect toward the mistake activity. Also watch for the opposite โ dismissiveness. Both are normal at this age.
At this first session, introduce the privacy rule clearly and early. Before you start, tell your child: 'We never put our real name into this. If it asks, we say a made-up name.' Make it a game, not a warning.