Teaching "Generative AI" to Explorer mode (ages 5–7)

Hook & Warm-Up

Gather students on the rug or at their desks. Read the app's hook line the way Pixel would say it — bright, curious, a little bit amazed:

"ChatGPT writes stories! A.I. makes pictures! A.I. even makes music! 🎵 But HOW? It doesn't have imagination — it mixes up patterns from things it learned!"

Before opening the app, spend two or three minutes on a warm-up. Ask: "Has anyone ever seen a computer or a phone draw a picture, or make up a story, or play a song it wrote all by itself?" Let a few students share — most will have seen this somewhere, even if they didn't know a name for it. Then ask the question that sets up the whole lesson, and don't answer it yet: "Where do you think the computer GOT that picture or that story from? Did it dream it up all by itself, the way you do when you imagine something?" Take three or four guesses without correcting anyone — you'll circle back to this exact question at the end of the lesson, and it's a nice moment for students to notice their own thinking changed.

You might also hold up (or point to) something a student drew earlier in the week and ask, "When you drew this, where did the idea come from?" This gives you a real comparison to come back to later, between a human idea and an AI "remix."

Main Activity

Open the lesson and go through its three scenes together as a read-aloud. For this age band you are narrating and translating every card, not letting students read alone — the goal is a simple, correct picture in their heads, not the vocabulary.

Scene 1 — How It Creates ✨: Walk through each card with a simple comparison and a small physical gesture, since acting things out helps this age group remember:

• Writing (📝): "When AI writes a story, it's playing a guessing game — 'what word comes next, what word comes next' — over and over, super duper fast, thousands of times, until it has a whole story!" Try it live: say the start of a sentence ("The dog ran to the...") and have students shout out the next word. "That's kind of what AI is doing — just millions of times faster, and it learned its guesses from reading a giant, giant pile of stories."
• Drawing (🎨): "AI drawing starts as messy fuzzy dots — like TV static, or a scrambled-up puzzle. Then, little by little, it cleans up the fuzz until a picture appears, kind of like a photo becoming clear underwater." Scrunch your hands together tight like static, then slowly open them out flat like a picture appearing — have students copy the gesture with you.
• Music (🎵): "AI listened to millions and millions of songs, and it learned patterns — like which sounds usually go together, the way peanut butter usually goes with jelly. Then it mixes those patterns into a brand-new song. It's remixing, not writing a song because it feels something."
• Movies (🎬): "A video is really just LOTS of pictures shown one after another, really fast — like a flip-book! AI makes each little picture, one at a time, then lines them all up so it looks like it's moving."

Expect a sharp student to ask, "so did the AI make that up, or did it copy it?" — that is exactly the right question, and worth celebrating out loud. Answer: "A little bit of both, and also a little bit of neither! It didn't copy one single thing exactly, but it also didn't dream it up out of nowhere the way you would. It mixed together little patterns from lots and lots of things it saw before — like making a smoothie out of fruit it already had, instead of growing brand-new fruit."

If time allows, act out one more comparison to make the "learning from lots of examples" idea stick: "Imagine I showed you ONE picture of a dog and asked you to draw a totally new kind of dog. Hard, right? Now imagine I showed you a THOUSAND pictures of dogs first. Suddenly you'd have lots of ideas to mix — floppy ears from one, a curly tail from another, spots from a third. AI needs to see a giant, giant pile of examples before it can remix anything at all — way more than a thousand, more like millions and millions."

Scene 2 — Strengths 💪: Keep this part quick and celebratory. AI can make a whole picture or story in seconds (⚡); it can mix silly, wild ideas together — like "an astronaut riding a horse on the moon in a rainbow spacesuit" (💡); anyone can make art with it, even without years of drawing lessons (🌍); and if you don't love what it made, you can just ask again and again for free (🔄). Go around the circle and have each student say one silly picture idea out loud that they'd want AI to try drawing — this makes the "combining ideas" strength concrete, fun, and gets everyone talking. If a student's idea is especially silly ("a shark eating pizza in outer space wearing sunglasses"), write it on the board and come back to it in the Wrap-Up extension activity.

Scene 3 — Limitations ⚠️: Slow down for this part — it's the most important few minutes of the lesson. "Sometimes AI makes up things that SOUND true but aren't true at all — like it's guessing confidently even when it's wrong. That's called a hallucination. That's exactly why we always check with a grown-up, a teacher, or a real book before believing something AI told us." Then: "AI can also make a fake video that makes it LOOK like a real person said or did something — even if that person never did it! That's called a deepfake, and it's exactly why we should never believe a video is 100% true just because we saw it with our own eyes — we check with a trusted grown-up first, every time." Finish gently: "And even though AI can write about being happy, or sad, or excited, it doesn't actually FEEL any of that — it's just matching patterns of words that usually go together. It doesn't truly understand feelings, the way you do."

Discussion

• "If AI doesn't really imagine things the way you do, where do you think it gets its pictures and stories from?"
• "Can you think of something YOU could imagine that you don't think AI could ever truly dream up on its own? Why not?"
• "Why do you think it's important to ask a grown-up before believing everything a video or a picture shows us?"
• "If AI writes a whole story in one second, is that really 'AI's story' — or is it something else? What do you think, and why?"
• "Remember the smoothie idea — AI mixes patterns like fruit it already has. What's something only YOU could make, that isn't a remix of things you've already seen?"
• "AI needs to see LOTS of examples before it can make something new. Do you need to see lots of examples before you can draw something new, or can you sometimes draw something with no examples at all?"

Quiz Walkthrough

Wrap-Up & Extension

Close with: "AI can make cool pictures, stories, and songs super fast — but it's mixing patterns it already knows, not truly imagining. YOUR imagination — the ideas that come from inside YOU, that nobody has ever seen before — is something AI can't do. That's something special about being you!" If a student shared a silly idea earlier (like the shark eating pizza in space), point back to it here: "That idea came from your own brain, in one second, with no examples needed. That's something AI genuinely cannot do."

Extension activity — Human vs. Remix Drawing: Give each student the same silly prompt (for example, "a cat wearing a hat, riding a bicycle, on the moon"). Have them draw it themselves for five minutes, using only their own imagination. Then, either using a teacher-driven device to generate an AI picture from the exact same prompt, or by showing a pre-made printed example if live generation isn't available, compare the two pictures side by side as a class. Ask: "What's different between your picture and the AI's picture? What did YOU add that came only from your own imagination, that nobody told the computer?" Let a few students hold up their drawings and share one thing they're proud of that the AI picture doesn't have.

To stretch this even further, hang both pictures — the human one and the AI one — side by side on the wall, and revisit them the next day during another lesson. Ask students if they can still tell which is which and why. This turns the 5–15 minute core lesson into a full, hands-on class period, and it reinforces the lesson's central point — remix versus true imagination — without any technical vocabulary at all.