One third of American teenagers say a sexualized AI-generated image of them has been shared without their consent. More than half admit they have created one of someone else. These are not edge cases. This is the baseline.
The Numbers Are Worse Than Anyone Expected
George Mason University researchers surveyed 557 teenagers and the results came back ugly. According to the New York Post, 33% of respondents, 185 kids, said a sexual AI image of them had been shared without their consent. And 55%, or 308 teens, admitted to using nudification software to create deepfake images themselves.
Chad M.S. Steel, the digital forensics researcher at George Mason who led the study, told the Post he had originally expected around 30% of teens to report using this software. He was "really shocked" by what he found, particularly how often these images were being spread around without the target's knowledge or agreement.
Hany Farid, a digital forensics professor at UC Berkeley, put it plainly. "Anyone with a single image of themselves online" is now at risk. "It is disturbingly easy," he said, pointing to the proliferation of "many free and easy-to-use apps and services" that can produce these images in minutes. Free. Easy. Available right now to any 13-year-old with a phone.
A 13-Year-Old Got Expelled. The Boy Who Made the Image Did Not.
Joseph Daniels watched this play out in real time with his daughter last August. She was 13 years old when an "extremely explicit" deepfake image of her started circulating around her school. When she saw it being passed around on a school bus, she snapped and slapped the boy she believed created it.
She was expelled. She had to transfer schools. The New York Post reports Daniels is now pursuing federal litigation against school officials for failing to intervene before the situation got to that point.
"She's faced anxiety, issues with depression, having to switch schools, it's taken a big toll on her," Daniels told the Post. He also issued a warning to other parents: get up to date on what's on your kids' devices, because the apps that exist now are nothing like what previous generations had to contend with. He's not wrong about that. The gap between what parents understand and what teenagers are actually doing online has never been wider or more dangerous.
This Is Not New, and the System Still Has No Real Answer
New Jersey mother Dorota Mani has been living this fight since October 2023, when her then-14-year-old daughter Francesca had a deepfake nude circulated at her high school. She has spent years since as an advocate for stronger legal protections and told the Post on Tuesday that the country still has not caught up.
"The technology used to create fake nude images has become widely accessible, while awareness, safeguards and legal protections have struggled to keep up," Mani said. That is an understatement so restrained it barely registers. The tools to destroy a teenager's life are free and require no technical skill. The legal scaffolding to punish the people who use them is, in most places, still being argued about in committee.
Mani is calling for the law to simply make this illegal, full stop. "Regardless of who creates or shares these images, the harm is real, and it should be illegal." She also argued schools need to be teaching digital citizenship, consent, and ethics around technology use. Those are good ideas. They are also ideas that should have been embedded in curricula about five years ago.
Why the Numbers Are This High
Steel offered some context for why the survey results came back so extreme. Speaking to the Post, he described this generation as "GenAI natives." Since they got their first phones, AI tools have been baked in. Camera filters, virtual try-on apps, image modification features, they are all normal, everyday interactions for these kids.
"Those are things that they grew up with and are native to, which I think greatly influenced the numbers," Steel said. The leap from a face-swap filter to a nudification app is, for a teenager who has spent years modifying images as casual entertainment, not as psychologically large as adults might assume. That is not an excuse. It is an explanation that should be informing both policy and parenting right now.
The Dingo Take
Let's be direct about what this survey is actually describing. We have built an environment where children can use free, publicly available software to produce sexual images of their classmates, distribute them instantly, and face consequences that are, at best, inconsistent and, at worst, nonexistent. The victim in the Daniels case got expelled. Sit with that. The girl whose face was put on someone else's naked body by a classmate lost her school, developed anxiety and depression, and had to start over somewhere else. The system responded to this crisis by punishing the person it hurt.
The legal framework is a mess. Some states have passed laws specifically targeting non-consensual deepfake pornography. Federal protections remain inadequate. And while lawmakers spend years drafting, debating, and watering down legislation, the apps keep getting better, faster, and easier to use. Farid at Berkeley is not being hyperbolic when he says anyone with a single photo online is at risk. That is literally everyone. Every teenager with a social media account. Every kid who has ever been photographed at a school event.
Parents being told to check their kids' phones is not the solution, though it is a start. Schools being asked to teach digital ethics is not the solution either, though they should absolutely be doing it. The solution is treating the creation and distribution of non-consensual sexual imagery, including AI-generated imagery, as the serious crime that it is, and building enforcement mechanisms with actual teeth. Right now, the technology is moving at full speed and the law is doing a light jog in the wrong direction. Somebody needs to close that gap before another 13-year-old pays for it.
