Skip to content
Bot Tamers

How to explain deepfakes to kids (ages 8 to 12)

By the Bot Tamers team. Last updated .

Your child has probably already seen a picture or video made with AI, whether or not anyone told them. You do not need to be a tech expert to explain it. Here is what a deepfake is, how to talk about it at 8 to 12, a few checking habits, and what to do if something they see upsets them.

What a deepfake is, in plain words

Netsafe, an online safety organization, uses "synthetic media" as the umbrella term for "any content (images, video, audio, text) that's been artificially created or altered using technology, especially AI." A deepfake is one kind: AI is used to make "highly realistic but fake content that makes it look like a real person said or did something they didn't."

Netsafe lists what kids might come across, including "face-swapping videos," "cloned voices" and "entirely AI-generated people or scenes."

Not every AI-made picture is meant to trick anyone. A cartoon dog that talks is AI-made but harmless. The trouble starts when something fake is passed around as real.

How to explain it at this age

Keep it short and true. For example: "Computers can now make photos, videos and voice recordings of things that never happened, and they can seem completely real. So we don't decide something is real just because it looks real. We check where it came from."

Netsafe suggests "approaching the topic with curiosity (instead of fear)," and it is fine to say you are learning too. A few of its questions to open the talk:

  • "Have you ever seen something online that looked real but wasn't?"
  • "What clues might help you figure out if something is real or fake?"
  • "Who could you go to if something felt wrong online?"

HealthyChildren.org, from the American Academy of Pediatrics, suggests turning it into a game: "make a game out of spotting things that seem real vs. those that appear to be fake."

Simple habits for checking

These are habits, not a test that always works. Three questions a child can remember:

  1. Who posted it? A known news outlet or the person or place in the picture? Or an account you have never heard of? HealthyChildren.org suggests asking, "What reasons do we have to trust (or distrust) the sender?"
  2. Does another trusted source show it? If something big really happened, more than one trusted place will usually show it. Netsafe suggests checking "together using trusted sources or reverse image searches."
  3. Does something look or sound off? Strange hands, scrambled writing, a voice that does not match the lips.

Be honest with your child about the third one: looks alone are not a reliable test. Better Internet for Kids, a European Commission site, notes that the clues "may not exist in deepfake content of the near future." In a 2022 study in the journal PNAS, researchers found that AI-made faces had become "indistinguishable" from real ones for the people they tested. So if a picture looks fine, that tells you very little. Who posted it, and whether trusted sources agree, tell you more.

And when you cannot tell: do not share it. Netsafe's advice is to encourage kids "not to share or react to something if they're unsure it's real."

Our guide on teaching kids to fact-check AI uses the same habit for AI answers in words, with a ten-minute activity.

If your child sees something upsetting

Sometimes a fake is meant to scare or embarrass, or shows someone your child knows. Netsafe's advice:

  • Stay calm and listen. "Try to stay calm, listen without judgement, and focus on understanding what they've seen and how it's affected them."
  • Look at it together. Netsafe suggests saying, "Let's look at this together and figure out if it's real or not."
  • Report it. Use the reporting tools on the app or site where it appeared.
  • Save evidence if it matters. Netsafe suggests "saving evidence if the content could cause harm or involve someone they know."
  • Let them know they can come to you. Netsafe suggests: "If anything like this ever happens, you can always talk to me."

The American Academy of Child and Adolescent Psychiatry (AACAP) notes deepfakes "can be used to create new forms of cyberbullying," and advises talking with a trained mental health professional if you have concerns. For parents of older kids, Common Sense Media has a separate guide on talking with teens about deepfakes and explicit content.

For more on keeping these talks going, see how to talk to your child about AI. If you homeschool, this makes a good weekly topic in our AI curriculum for homeschool plan.

How Bot Tamers teaches this

In the ages 8 to 10 lesson "Is it real or AI-made?", your child scrolls four posts in a pretend video app and first guesses real or AI-made just by looking. Two are real (a rare blue lobster and snow in the Sahara) and two are AI-made. Then they use detective tools on each post: who posted it, whether other trusted places show it, and whether there is a label or a clue. The rule they leave with: when they can't tell if a post is real, they stop, keep it to themselves, and check with a trusted adult. Then your child asks you: "Have you ever seen something online that turned out to be fake?"

There is no live chatbot: Bit, the robot in every lesson, is scripted. Your parent dashboard marks the skill once your child has shown they can do it. See every skill on the curriculum page, and how we handle privacy on our safety page.

Kids do Bot Tamers lessons on our website, on a computer, iPad or phone. 7 days free, then $19.99/month or $199.99/year. Renews automatically until canceled. Cancel anytime in Settings in your parent area, then Manage subscription. Cancel before the trial ends and you won't be charged. You can create a parent account on our website now, and add each child to it.

Sources