Skip to content
← All posts

· Updated · 5 min read

How to Teach Your Kid to Spot a Deepfake (Ages 7–12)

By . Founder of ThinkaBit: a dad based in the Netherlands who has worked in tech for 25 years and now works with AI.

Part of our guide: How to teach your kids to spot deepfakes and AI fakes

Share:

Your child is growing up in the first generation that can’t fully trust its own eyes. Deepfakes — AI-generated videos, images, and voices designed to look real — now spread across the apps kids use every day. The good news: spotting them is a skill, and kids are great at learning skills when you make it a game.

Here’s a simple, parent-friendly way to teach it.

First, explain what a deepfake actually is

Keep it concrete: “A deepfake is a video or picture that a computer made up to look real — like a costume for a face or a voice.” The key idea for a 7–12 year old is that looking real and being real are not the same thing. AI can copy a face or a voice from lots of examples, then make it say or do things that never happened.

Why kids are especially at risk

Research has found that a large majority of children and teens can’t reliably tell AI-made content from the real thing — and many believe convincing-but-false “facts” they see online. Kids tend to trust what looks polished, and deepfakes are designed to look polished. That’s exactly why the skill has to be taught, not assumed.

5 clues kids can learn to spot a fake

  • Weird eyes and blinking. Deepfaked faces often blink oddly, stare, or have eyes that don’t quite line up.
  • Blurry or wobbly edges. Look where the face meets the hair, ears, or glasses — fakes often smear or shimmer there.
  • Lips out of sync. If the mouth doesn’t match the words, be suspicious.
  • Too perfect, or slightly “off.” Strange shadows, extra fingers, melted backgrounds, or skin that’s too smooth are giveaways of AI images.
  • It’s trying to make you react fast. Shocking “you won’t believe this!” content is built to be shared before you think. That urgency is itself a red flag.

Teach the one trick that beats most fakes: pause and verify

The single most powerful habit isn’t a visual clue — it’s a pause. Teach your child to ask three questions before believing or sharing anything: Who made this? Why? Is there proof from a source I trust?

Then show them reverse image search: you can search with a picture (not words) to find where it really came from. A “today’s flood!” photo that turns out to be ten years old is instantly debunked. It feels like detective magic to kids — and it works.

Make it a daily habit, not a lecture

Skills stick through repetition, not a one-time talk. Turn it into a quick game: show a photo or headline at dinner and play “Fake or Fact?” Let them explain their reasoning. Over a few weeks, spotting fakes becomes second nature — the same way a daily streak builds any habit.

For still pictures there are extra tells worth knowing (hands, writing, melting backgrounds), which we cover in how to spot AI-generated images.

What to do when your child has been fooled

It will happen. It happens to adults every day. What matters is how the moment goes, because that decides whether your child tells you next time.

  • Stay curious, not cross. “Ooh, that one was good. What made it look real?” keeps the conversation open. Embarrassment shuts it down.
  • Look for the clue together. Go back to the video and hunt for the giveaway side by side. Finding it yourselves teaches more than being told.
  • Praise the check, not the catch. A child who says “I wasn’t sure, so I asked” did exactly the right thing, even if they believed it at first.
  • Undo the share if there was one. If they passed it on, help them delete it or send a quick “turns out this was fake” message. That’s a grown-up habit worth modelling.

If you want words for that first conversation, we wrote a short script in how to talk to your child about AI-generated videos.

Questions parents ask

What age should I start? Around 7 is a good time for the simple version: “pictures and videos can be made up by a computer.” By 10 to 12, kids can handle reverse image search and talk about why someone would make a fake.

Can’t an app or AI detector just tell us what’s fake? Detection tools exist, but none is reliable enough to hand the job over to. The clues and the pause work on any video, on any app, with no tool needed.

Won’t this make my child distrust everything? Not if you frame it as checking, not doubting. The goal is “let’s find out”, not “nothing is real”. Most of what they see is fine; the skill is knowing how to tell.

Do they need to understand how deepfakes are made? A little helps. Once a child knows that AI learns by copying patterns from lots of examples, fakes stop feeling like magic. Our guide on how to explain AI to a child has a simple way in.


That daily habit is exactly what we built ThinkaBit around — a playful app for kids 7–12 with a daily “Fake or Fact” challenge, plus games that teach how AI works, how to stay safe online, and how to think critically. It’s free to start: download it on the App Store or Google Play.

Share: