In 2020, Selena Scola won a landmark class-action lawsuit against Facebook after spending eight and a half months working as a content moderator. She alleged that constant exposure to extreme graphic content – including sexual abuse and child endangerment – on the platform led to post-traumatic stress disorder.
Even six years later, she can no longer watch her favorite horror films. Violence in film or television destabilizes her. She experienced what she calls "digital death," an inability to use social media that led to strained social connections; she just recently began using platforms like Facebook again.
Scola said that more recently, she has become the focus of online abuse and has to preserve evidence of doxxing, sexual imagery and threats
against her.
"I once moderated harmful content to protect Facebook and Instagram users, now I am stuck reviewing abuse directed at me to protect myself," she said.
Since her time as a moderator, the internet has, some argue, become an even darker place.
"Previously, much of the harmful material I reviewed had been photographed, recorded or livestreamed by a person," Scola explained. "Generative AI now enables users to manufacture non-consensual intimate imagery and child sexual abuse material, including from ordinary photographs, and produce variations at scale."
The rise of "nudify" apps and AI-generative software has transformed how nonconsensual intimate imagery (NCII) and child sexual abuse material (CSAM) are created, distributed and removed online, and experts are sounding the alarm on the deepfake nude crisis. An analysis by Wired and Indicator, which cited USA TODAY's reporting on a Lancaster, Pennsylvania, deepfake scandal, found nearly 90 schools and 600 students across the world were impacted by AI-generated nudes between 2023 and April 2026. Research from the child protection group Thorn found that 1 in 8 teens personally know someone who has been targeted by fake nudes, and 1 in 17 said they had been a victim themselves.
Fingers have been pointed at the parents, the schools who failed to act on complaints and the teenagers who have created such images of their classmates. Victims have spoken about the very real and debilitating social and psychological impact of these images and what happens when they spread.
But advocates say there’s one crucial component all too often left out of the conversation – the tech companies which facilitate the creation and distribution of deepfake pornography. Experts and representatives from major tech companies like Meta say they are working together to prevent the spread the deepfake nudes, but some researchers and digital safety advocates argue that those same companies inadvertently enable the abuse through faulty moderation systems and dangerous advertisements.
It's true the issue is growing, and tech companies are, at times, struggling to keep up, explained Sean Litton, the former CEO of the Tech Coalition, an alliance of more than 60 technology companies focused on child safety. The companies are working in conjunction toward solutions, he added.
"Generative AI is really changing everything," Litton added. "Its ability to create imagery at speed and scale is changing the whole paradigm."
Tech companies are sharing 'signals' to take down repeat offenders
Deepfake abuse and online child sexual exploitation rarely happens on one platform. According to Litton, an offender may initially contact a child on a gaming platform before moving communication to a "confidential" messaging service, and might even use a third platform to exchange images. A payment platform may be involved, but the financial institution doesn’t have any information on what the transaction is about.
"No single company or no single part of the ecosphere can solve these problems on their own," Litton said. "Each one of these platforms is just seeing a little piece of this long type of grooming conversation."
To address repeat offenders and cross-platform abuse, the Tech Coalition launched Lantern in 2023, a system that allows participating companies to share intelligence across their network. Companies can distribute "signals" involving accounts, email addresses, URLs and other indicators that may help other platforms identify related abuse and remove accounts associated with those details across platforms. Litton said more than 2 million signals have been shared through Lantern, leading to more than 350,000 enforcement actions; 164,575 accounts and 163,112 URLs were banned across platforms, and 26,119 pieces of content were removed between 2023 and 2025 due to Lantern.
Enforcement is not automatic; it requires an independent investigation by every company in the chain. But what it allows for, Litton said, is each company to have the information needed to start that investigation.
Not all bad actors are adults grooming children. Often, teenagers are the ones making deepfake nudes of their peers. But Litton said the platforms treat all abusive imagery the same.
"Whether it’s being published or attempted to publish by a minor or an adult, they’re going to treat it the same way. They're going to take action on that account, and they're going to report it to the National Center for Missing and Exploited Children," he said.
Other tools like stopNCII.org work by generating and assigning a unique "hash," sometimes referred to as a "digital fingerprint," to the intimate images or videos. Participating platforms will then look for matches to the hash and remove any images within their systems, and stopNCII.org will periodically search for resurfaced content.
To help prevent images from being shared in the first place, experts recommend implementing "deterrence messaging." These pop-ups can be triggered by certain keywords or tags that are associated with deepfake or child sexual abuse images. A 2025 research report from the Center for Democracy and Technology found that individuals exposed to nonconsensual intimate images are more likely to engage in sharing behaviors. Flagging harmful keywords, sharing information on the impacts of nonconsensual intimate imagery and adding an additional step to proceed with those potentially harmful searches can make it more difficult to find, create or share this imagery.
Still, researchers are finding hundreds of ads linked to 'nudify' apps and child sexual abuse material
Across the board, Americans are increasingly fed up with AI and Big Tech. According to an August Pew Research Center survey, a majority of adults under 30 (55%) now say they’re more concerned than excited about AI, and a YouGov Poll in July found that 7 in 10 adults believe technology companies have too much power.
While tech companies point to new safety initiatives, some researchers say those same platforms can also enable deepfake abuse.
Katie Paul, director of the Tech Transparency Project, said it doesn’t boil down to being unprepared. Rather, her research suggests that tech companies are "incentivized to not address these issues."
"The blame is getting placed on individuals, when really the only ones who stand to benefit from this content are the companies that are failing to effectively address it," Paul said, adding there are many loopholes her organization has found that allow for this type of content to appear on popular apps.
In September, the organization found over 300 ads on Facebook and Instagram with child sexual abuse material, some of which included images of real children. The vast majority of these ads directed people to AI apps from China, many of which were capable of "nudifying" people.
Paul said that these big tech companies have "far more resources" than the parents, schools, kids or counselors who often get tasked with handling the deepfake nude crisis. Even researchers like herself aren’t properly trained to process the amount of child sexual abuse material in their findings.
"It’s incredibly upsetting and incredibly psychologically taxing," she said. "We’re a small nonprofit that is not trained to do this. We're doing it because it's important and we've come across this, but we're talking about a multi-billion dollar corporation that has all the resources in the world to effectively address this and is choosing not to."
But companies reject the notion that they are ignoring the problem.
According to Meta, many of these ads had already been removed before the Tech Transparency Project flagged them, most had fewer than 200 impressions, and the total ad spend was under $5,000.
Cindy Southworth, Meta's head of women's safety, said the company has strict rules against nudify apps and has been taking steps to fight them. She noted that the company removed more than 344,000 advertisements related to nudify apps during a recent three-month period and has pursued legal action against some app operators. The company sued a Hong-Kong based company that was attempting to advertise these apps, and sent cease and desist letters to dozens of others. Meta was also a founding member of the Tech Coalition, and has shared over 5,500 URLs — seemingly benign links that forward to a handful of harmful apps — with industry partners to help block the same offenders across multiple platforms, Southworth added.
"They are pivoting so quickly. So we need to pivot just as quickly to keep up with them, or get ahead," Southworth said. "This is an industry-wide issue, so we’re really committed to figuring out how we can help other companies prevent the spread (of non-consensual intimate imagery) and prevent nudify apps."
Litton said the deepfake problem isn’t going away anytime soon, but he has hope for a brighter future. Ahead of his interview with USA TODAY in August, he announced that he would be stepping down as the CEO of the Tech Coalition (Dr. Erin Saltman has stepped into the role as of Sept. 1). On his way out, he wants to see initiatives like Lantern spread even further.
"We need to be talking to each other. We need to find our way forward," he said. "As these things are developing so rapidly, we need to work together."
This article originally appeared on USA TODAY: Deepfake nudes are flooding social media. Here's what's being done about it
