Deepfakes of Your Kid: What They Are, How to Lock the Photos, and How to Take Them Down
By Natalie Gibson, Founder and CEOLast updated:
How to lock kids' photos, talk about deepfakes, and use official Take It Down steps if an image is already out.

Update, October 2026: app store parental controls don't block nudify apps
On October 19, 2026, the Tech Transparency Project published research showing that teens can easily find nudify and face-swap apps in the Apple App Store and Google Play. Test accounts set up as 13-year-olds could download 135 of these apps. Even with parental controls turned on, 24 were still available on Apple and 51 on Google. That's because many of the apps were inaccurately rated as suitable for minors. Apps that the stores later removed kept working on the test phones that already had them installed.
What this means for parents: app store age ratings and download controls alone won't keep these apps off a child's phone. Check which apps are already installed, not just new downloads, and talk with your kids about why making or sharing fake images of real people is harmful.
What is a deepfake?
A deepfake is a fake image, video or voice clone made with AI that shows a real person saying or doing something they never did. The name comes from "deep learning," the type of AI used to make them. The goal is to look real enough that people believe it.
Five kinds matter most for parents:
- Face swaps. Your kid's face is placed onto someone else's body in a photo or video.
- Nudify images. An app takes a clothed photo and generates a fake nude. These are the most common deepfakes aimed at kids and teens.
- Voice clones. A few seconds of audio is used to make a fake voice that says anything. Scammers use these for "your kid is in trouble" calls.
- Lip-sync videos. A real video is changed so a person's mouth matches words they never said.
- Fully generated images. AI creates a new picture from text, sometimes guided by a real child's photo so the result looks like them.
Not every deepfake is harmful. Movies use the same technology, and kids may make silly face-swap videos of themselves. The danger starts when someone uses a real child's face or voice without permission, especially to embarrass, threaten, or sexualize them.
How deepfakes are made, and why kids are targets
Making a deepfake used to take skill and powerful computers. Now it takes a phone, a free app, and one clear photo. The AI studies the face, then maps it onto new images or video in minutes. The more photos and video of a person that exist online, the more convincing the fake.
Kids are easy targets for two reasons. Their photos are everywhere: parent posts, team pages, yearbooks, and school sites. Offenders often use photos of real children taken from social media. And the people making them are often other kids, who don't see it as a crime.
The numbers are climbing fast:
- NCMEC received 67,000 reports of AI-generated child abuse material in 2024, and 485,000 in the first half of 2025.
- The Internet Watch Foundation found 3,443 AI-generated child abuse videos in 2025, up from 13 the year before. 97% of AI-generated abuse images depicted girls.
- 15% of high school students say they know of sexual deepfakes of classmates at their school.
How to spot a deepfake
Honest answer: it's getting harder every month, and you shouldn't rely on your eyes alone. Still, these signs can help:
- Faces that don't fit. Blurry edges around the hairline or jaw, skin that looks too smooth, or lighting on the face that doesn't match the room.
- Odd details. Warped hands, jewelry that melts into skin, background text that doesn't spell anything.
- Video glitches. Lips slightly out of sync, too little blinking, or a face that flickers when the head turns.
- Voices that sound flat. Strange pauses, no breathing sounds, or emotion that doesn't match the words.
- The story is off. It showed up from a stranger, a new account, or with a demand for money or more photos.
The best test isn't technical. Ask your kid what happened, and believe them. If they say "that's not me," treat it as a fake and move to the takedown steps below.
Is it illegal to make a deepfake of a kid?
Yes. Making or sharing a sexual deepfake of a real child is a crime under federal law and in almost every state.
- Federal law. The TAKE IT DOWN Act bans publishing intimate images of minors, real or AI-made, without consent. Since May 19, 2026, platforms must remove reported images and known copies within 48 hours.
- State law. As of August 2026, 46 states have laws against AI-generated child abuse material. Many treat it the same as real abuse images.
- Schools. Recent Title IX changes count AI-generated intimate images as sexual harassment, which gives schools a duty to respond.
What about the August 2026 court ruling? You may have seen headlines that a federal appeals court said AI abuse images are legal. That's not what it said. In United States v. Anderegg, the Seventh Circuit ruled one man couldn't be charged for privately possessing AI images that showed no real child. His charges for making and sharing images, and for sending them to a minor, are still pending. The ruling applies only in Illinois, Indiana, and Wisconsin, and it doesn't cover fakes made from a real child's photo.
Lock the photos tonight
Public photos of your kid, with their name and school in the same place, are the raw material. You do not have to disappear them. You do have to stop handing out the combo.
Your posts. Set old kid photos to friends-only or take them down. Turn off public tagging. Ask grandparents and the team mom to stop posting full-name captions on public photos.
School and sports. Ask if the yearbook, booster page, or league site is public. Opt out of public galleries if you can.
Do not upload their face to random AI. Headshot-to-cartoon apps, make-them-an-action-figure tools, and free face-swap sites are the easy path. If the tool is not one you chose, and you cannot see what it keeps, do not feed it a photo of your child.
Voice. A few seconds of a public video can be enough to clone a voice. Pick a family code word. Never post it. If a call says your kid is in trouble and needs money, hang up and call them yourself.
What do you need to tell your kids?
If someone makes a fake of you, come to me first. You are not in trouble.
Do not make one of a classmate, even as a joke. Creating or sharing an intimate image of a minor is a federal crime. Age of the kid who pressed the button does not make it a prank.
Do not send nudes. Do not save other kids nudes. If someone sends you one, show me. Do not forward it.
If a deepfake is already out, what do you do?
Do this in order. Do not skip to messaging the other parent.
1. Take the shame off the table. Say it is not their fault. A fake documents nothing they did.
2. Do not forward the image. Do not download it to keep a copy. Do not send it to the group chat to prove it. Write down the URL, the username, the date, and where you saw it.
3. Ask the platform to take it down. Use the app report flow. Under the federal Take It Down Act, covered platforms must give you a way to ask, and they must remove the image and known identical copies within 48 hours of a valid request. That includes real images and AI forgeries. The FTC parent-facing steps are here.
4. If the platform does not take it down, report the platform. TakeItDown.ftc.gov is for platforms that ignore a valid request, have no removal process, or have a broken one. It is not the first place you file. Ask the app first.
5. If your child is under 18, tell NCMEC. Report through the CyberTipline. If the file is already on a device in your house, use NCMEC Take It Down on that device. It fingerprints the file without uploading the photo. Do not download or send the image just so you can use the tool: takeitdown.ncmec.org.
6. Call the police, and tell the FBI. Local police, plus tips.fbi.gov or 1-800-CALL-FBI. Sharing, or threatening to share, an intimate image of a minor is a federal crime.
7. Tell the school in writing if classmates or a school device are involved. Email the principal or counselor so there is a record.
8. Do not pay. Do not send more photos. Do not give anyone access to a phone or a bank account. Block them.
Why we wrote this
We built HeyOtto to give kids a safer alternative to a general-purpose AI because kids were already using AI.
On HeyOtto, a parent creates the account. You can read the chats. Otto is the assistant / tool, not a friend or companion. We do not use kids conversations to train models, and we do not sell them.
HeyOtto now offers photo upload, but only on Pro and only after a parent turns it on for a child. Photo uploads stay off until you opt a child in, every photo is screened before Otto sees it, and photos are never used to train AI. Our first priority is still the safety of children. Read more about photo safety for kids on HeyOtto.
If you want that setup, start with the parent's guide to AI. When you are ready, try HeyOtto free.
This is a how-to, not legal advice. If your kid is in crisis, call or text 988. For the longer parent Q and A this was pulled out of, see Parenting in the Age of AI.
Note: This guide was written with the help of Otto's AI writing tool. Was edited and updated by the HeyOtto team.
FAQ
Frequently Asked Questions
Common questions about this topic, answered.
Use NCMEC Take It Down, then report the file on the platform that is hosting it.
No. Uploading a child face to an AI image tool can create another copy you do not control.
Take It Down is the official NCMEC process for intimate image abuse, including deepfakes of minors.
Related Articles
Ready to Give Your Child a Safe AI Experience?
Try HeyOtto today and see the difference parental peace of mind makes.






