For years, parents taught children a fairly straightforward rule about the internet:
Be careful what you post.
That is still good advice.
It is no longer enough.
Artificial intelligence has changed something fundamental about growing up online. A photograph no longer has to document something that actually happened. A voice recording does not necessarily prove someone said the words you hear. A video can be altered. A face can be placed onto another person’s body. A message can be written to sound remarkably like someone who never wrote it.
For children, that creates an entirely new category of digital risk.
A teenager may someday have to say:
“That’s me in the picture. But I never took that picture.”
Or:
“That sounds like my voice. But I never said that.”
Parents need to be prepared to believe them.
The new rules of digital safety aren’t only about protecting what children create and share.
Increasingly, they are about protecting children from things technology can create about them.
What Is a Deepfake?
A deepfake is synthetic or manipulated media created using artificial intelligence or related technologies to make someone appear to say or do something they did not actually say or do.
Deepfakes can involve:
- photographs;
- videos;
- voices;
- or combinations of different media.
Some are obvious jokes.
A friend’s face is placed into a movie scene.
A celebrity appears to sing a ridiculous song.
A teacher appears in a silly fake photograph.
But the same underlying technology can be used for much more harmful purposes.
A fake image can humiliate a child.
A cloned voice can be used in a scam.
A manipulated photograph can become part of bullying.
And an ordinary photograph of a child can potentially be transformed into sexualized material that the child never created.
That last category is particularly important for parents to understand.
A Child Doesn’t Have to Take an Intimate Photograph for One to Exist
Traditional conversations about online sexual safety often focused heavily on sexting.
Don’t take intimate pictures.
Don’t send them.
Assume anything you send can be saved.
Those lessons remain valuable.
But generative AI has created a disturbing new possibility.
Someone may be able to take an ordinary photograph—perhaps from social media, a school event, a sports team or another online source—and manipulate it into fabricated sexual imagery.
The victim never undressed.
Never posed.
Never sent an intimate photograph.
Yet an image may exist that appears to show otherwise.
For a teenager whose classmates receive it, the distinction between “real” and “fake” may not lessen the humiliation.
Their face is still attached to it.
Their peers may still see it.
The image may still spread.
The emotional harm can be very real even when the image is completely fabricated.
Deepfakes Can Turn Ordinary Bullying Into Something Different
Children have always found ways to embarrass one another.
A cruel rumor.
An altered yearbook photograph.
A humiliating note.
A mean joke.
AI changes the scale and realism of what can be created.
Imagine a student angry after a breakup.
Instead of spreading a rumor, the student creates a convincing fake image of the other person.
Someone makes a fabricated video showing a classmate doing something humiliating.
A fake voice recording appears to capture a student making a racist or threatening statement.
The content gets shared in a group chat.
By the time the child learns it exists, dozens or hundreds of people may have seen it.
Now the child faces two problems.
The content itself.
And proving that it isn’t real.
That second problem is something previous generations rarely had to consider.
“But Everyone Knows It’s Fake” Is Not Good Enough
Adults sometimes minimize manipulated content because it isn’t authentic.
But authenticity is only part of the harm.
Imagine someone distributes a convincing sexualized image of a 14-year-old.
The fact that the image was fabricated doesn’t necessarily undo:
- humiliation;
- harassment;
- bullying;
- social isolation;
- fear;
- reputational harm;
- anxiety;
- or the trauma of seeing one’s body or identity digitally manipulated.
A child may also fear that the image will follow them indefinitely.
Who saved it?
Where else was it posted?
Will someone find it years later?
Will everyone believe the explanation?
Parents should take fabricated content seriously.
Fake content can cause real harm.
Voice Cloning Creates Another Risk
Photographs aren’t the only thing AI can manipulate.
Voice-cloning technology can generate audio that resembles a real person.
That creates obvious possibilities for scams.
A parent receives a call.
The voice sounds like their child.
“Mom, I’m in trouble.”
“Dad, I need money.”
“Please don’t tell anyone.”
The emotional reaction can come before rational analysis.
Criminals have long relied on urgency and impersonation. AI can potentially make the impersonation more convincing.
Children can be targeted too.
A message that appears to come from a friend may request money, a password or a security code.
The lesson for families is simple:
A familiar voice is evidence. It is no longer necessarily proof.
When an unexpected request involves money, secrecy, credentials or an emergency, verify the situation through another method.
Families Need a Verification Plan
This is one of the most practical steps families can take.
Agree in advance on how you will verify an unusual or urgent request.
If a child supposedly calls asking for money, call them back through the number you already know.
Contact another family member.
Ask a question a stranger would have difficulty answering.
Some families may choose a private family verification word or phrase.
The exact system matters less than having one.
AI scams are most powerful when they create panic.
A plan creates a pause.
And a pause can prevent a terrible decision.
AI Makes Online Identity Harder to Verify
In another part of digital safety, parents traditionally told children:
Ask for a photograph.
Video chat.
See whether the person is real.
Those steps can still provide useful information.
But artificial intelligence complicates them.
Images can be generated.
Photographs can be manipulated.
Voices can be cloned.
Video technology continues to improve.
Children therefore need a more sophisticated understanding of online identity.
Verification should not depend on one photograph, one voice recording or one piece of digital evidence.
More importantly, children should maintain appropriate boundaries even when they believe someone’s identity has been verified.
Don’t share passwords.
Don’t send intimate material.
Don’t reveal precise location information unnecessarily.
Don’t arrange secret meetings.
Don’t allow someone to pressure you into hiding a relationship.
Good boundaries remain valuable even when technology makes identity harder to establish.
AI Can Make Scams More Convincing
Scams used to contain clues.
Awkward language.
Strange grammar.
A poorly designed website.
An email that clearly didn’t sound like the person supposedly sending it.
Generative AI can help eliminate some of those clues.
A scam message can be polished.
It can be personalized.
It can imitate a particular tone.
It can potentially incorporate information already available publicly about a child or family.
That means families should stop relying primarily on whether a message looks professional or sounds believable.
A better question is:
What is this person asking me to do?
Send money?
Provide a password?
Share a security code?
Click an unfamiliar link?
Send a photograph?
Keep something secret?
Act immediately?
Those behavioral warning signs remain useful even when the message itself becomes more convincing.
AI Can Be Used for Impersonation
A fake account used to require someone to steal photographs and write the messages themselves.
AI can make impersonation easier to scale.
Someone may create fake profile images.
Generate realistic-looking posts.
Produce convincing messages.
Imitate someone’s writing style.
Create fabricated audio.
The purpose could be bullying, fraud, harassment, sexual exploitation or reputational harm.
For children, this means identity theft can take on a very personal form.
Someone isn’t necessarily stealing their identity to open a bank account.
They may be stealing it to become them online.
Injured and Have Questions?
Call us or request a case review by our experts.
What Should Parents Teach Children About Photographs Now?
Parents shouldn’t tell children never to post a photograph.
That isn’t realistic for most families.
Instead, teach children that posting creates a degree of exposure.
Ask:
Is the account public?
Who can download or screenshot this?
Does the photograph reveal where you are?
Does it show your school?
Does it identify where you live?
Are you comfortable with people outside your intended audience potentially seeing it?
And now add another question:
How would you feel if someone manipulated this image?
That is an uncomfortable conversation.
But it reflects the environment children are entering.
What If Someone Creates a Fake Sexual Image of My Child?
Take it seriously.
Do not dismiss it because the image is fabricated.
Start by supporting your child.
They may feel violated, humiliated, frightened or furious.
Then preserve information showing what happened.
That may include:
- screenshots;
- usernames;
- profile information;
- links;
- messages;
- dates and times;
- information about where the material was posted;
- threats accompanying the material;
- and reports made to platforms or schools.
Be cautious about repeatedly copying, forwarding or redistributing sexualized material involving a minor.
Report the content through the platform where it appears.
Depending on the circumstances, the school, law enforcement, child-protection organizations or other appropriate authorities may need to become involved.
If someone is using the fabricated image to demand money, additional images or another action from the child, the situation may also involve sextortion.
What If the Fake Content Is Spreading at School?
Online harm and school life can become inseparable remarkably quickly.
A fabricated image created at home can be on dozens of students’ phones before first period.
Parents should document what is occurring.
Who received it?
Where was it shared?
Are students continuing to distribute it?
Is the child being threatened or bullied?
Has school administration been notified?
Schools may have policies and legal responsibilities involving bullying, harassment, sexual misconduct and student safety.
The specific response will depend on the circumstances.
But parents should resist the idea that because something was created “online,” it exists outside the child’s real-world environment.
If it affects the child’s ability to attend school safely, the harm is no longer confined to a screen.
Teach Children Not to Create Deepfakes of Other Kids
Digital safety isn’t only about protecting your child from other people.
It is also about teaching children not to harm others.
A teenager may view an AI-generated image as a joke.
“It’s obviously fake.”
“I was just messing around.”
“I didn’t even take a real picture.”
Those explanations may miss the point entirely.
Teach children:
Do not create sexualized images of classmates.
Do not fabricate humiliating content about other people.
Do not clone someone’s voice to embarrass them.
Do not create fake accounts pretending to be someone else.
Do not share manipulated sexual material because everyone else is sharing it.
Technology making something easy does not make doing it harmless.
This may be one of the most important ethical lessons of the AI era.
How Do You Teach a Child to Spot AI?
Be careful with this question.
There are clues that can sometimes suggest manipulated content.
Images may contain inconsistencies.
Audio may sound unusual.
Details may not line up.
A source may be suspicious.
But AI-generated media is improving quickly.
We should not give children false confidence that they can always identify a fake by staring at someone’s hands or listening for a strange voice.
Instead, teach verification habits.
Who posted this?
Where did it originate?
Is there another source?
Does the person supposedly involved confirm it?
Is someone using the content to demand an immediate action?
Could this have been manipulated?
The future of digital literacy may depend less on teaching children to spot every fake and more on teaching them not to trust important claims without verification.
The New Rule: Seeing Is Not Necessarily Believing
For generations, photographs and recordings carried special authority.
“There’s a picture.”
“I heard the recording.”
“It’s on video.”
Those statements felt like proof.
Children growing up with generative AI need a different instinct.
Not cynicism.
Verification.
A photograph can be real.
A video can be authentic.
A voice recording may be exactly what it appears to be.
But the existence of digital media alone can no longer settle the question.
That is an enormous cultural shift.
Children who understand it early will be better prepared for the world they are entering.
What Parents Can Do Today
Talk About Deepfakes Before Your Child Encounters One
Don’t wait for a crisis.
Show children age-appropriate examples of manipulated media and explain what the technology can do.
Establish a Family Verification System
Decide how family members will confirm unusual requests involving money, emergencies or sensitive information.
Review Privacy Settings
Limit unnecessary public access to photographs, personal information and recordings.
Teach Boundaries, Not Just Detection
Your child does not need to identify every fake if they already know never to provide passwords, sexual images, money or precise personal information because an online stranger requests it.
Take Fabricated Content Seriously
If your child says an image, video or recording is fake, investigate before assuming the digital evidence tells the whole story.
Preserve Evidence of Abuse
Document harmful posts, accounts, threats and distribution before material disappears.
Don’t Redistribute Harmful Content
Parents trying to document a problem can inadvertently spread the very material they want removed. Be especially careful with sexualized content involving children.
Teach Your Child the Other Side of the Rule
Don’t create harmful AI content involving someone else.
Don’t share it.
Don’t help it spread.
Digital citizenship includes responsibility for what technology makes possible.
Childhood Is Entering a World Where Evidence Can Be Manufactured
Every generation of parents has had to teach children not to believe everything they hear.
Then came the internet, and we taught them not to believe everything they read.
Now we have reached something different.
Children cannot automatically believe everything they see.
Or everything they hear.
That sounds frightening.
It can also become a form of resilience.
We can teach children to pause.
Verify.
Ask where something came from.
Question an urgent request.
Protect personal information.
Tell an adult when someone threatens them.
And understand that a fabricated image says something about the person who created it—not about the child depicted in it.
Technology will continue getting better at creating convincing versions of reality.
Our children need to get better at asking whether that reality is true.
When AI-Generated Content Causes Serious Harm to a Child
AI can make a joke.
It can also become a tool for exploitation.
A child’s photograph may be manipulated into sexualized material. A fake identity may be used to build trust. Fabricated content may become leverage for sextortion. A child’s voice or likeness may be impersonated. Deepfakes may be distributed through social-media platforms and cause devastating emotional or reputational harm.
When that happens, families may have questions about more than the person who created the material.
How did the content spread?
Was it repeatedly reported?
How did the platform respond?
Were protections available for minors?
Did recommendation or sharing systems amplify the material?
Did a company know that its products or services could be misused in ways that placed children at risk?
Those questions can matter when determining what legal options may exist.
At Rafferty Domnick Cunningham & Yaffa, our attorneys represent children and families confronting serious harms involving social media, online platforms and emerging technologies. We investigate how harmful digital material was created, distributed and amplified, as well as the role that platform design and safety practices may have played in the harm a child experienced.
If your child has been seriously harmed through a deepfake, AI-generated sexual imagery, online impersonation, sextortion or another form of technology-facilitated exploitation, contact Rafferty Domnick Cunningham & Yaffa to discuss what happened and learn more about your family’s legal options.
