Two people have been charged in the US with using AI to make non-consensual sexually explicit deepfake content, the first prosecutions to test whether the country’s deepfake crackdown has any teeth.
In 2025 the Take It Down Act was signed into US federal law, criminalizing sharing sexually explicit photos and videos without the consent of people depicted in the content, including AI-generated deepfakes. Responding to the arrests made following this law change, a senior FBI figure said that people who share illegal deepfakes will be pursued “with the full force of the law”.
The two men arrested in unconnected cases were Arturo Hernandez, 20, and Cornelius Shannon, 51. They were arrested in Texas and New Jersey respectively and face up to two years in prison for their alleged offences.
Hernandez is accused of publishing over 100 albums containing deepfake content online, including content depicting around 50 different women. US Attorney for the Eastern District of New York Joseph Nocella said in a statement that the albums “contain non-explicit images of identifiable individuals which morph into deepfake depictions of the individuals in various stages of undress or engaging in sexually explicit conduct.”
Shannon is accused of publishing at least 360 albums of content containing sexually explicit deepfakes depicting around 90 women. Both Hernandez and Shannon are accused of posting nonconsensual deepfakes of famous women and women who are not public figures.
FBI Assistant Director James C. Barnacle said: “This predatory conduct represents a disturbing abuse of technology that inflicts emotional harm on victims, violating their privacy, dignity, and security. The use of this emerging technology to victimize individuals is not innovative – it is criminal and will be pursued with the full force of the law.”
The Take It Down Act was signed into law following a tidal wave of cases of AI-created alleged nonconsensual deepfake porn being posted online, including on platforms such as the Mr Deepfakes site, which shut down around the same time the Act was signed.
As well as causing distress to those depicted in it without consent, non-consensual deepfakes can be used for blackmail, extortion and image-based abuse, the through-line in this long-running problem. Beyond bringing in the prospect of prison time for individual offenders, the Act compels platforms and sites that primarily feature user-generated content, such as social media platforms, to provide a system by which non-consensual adult and explicit content can be reported for removal.
Critics of the Act have warned it could create a takedown-on-demand landscape where platforms strip out almost anything flagged, on the safe assumption that over-removal carries less legal risk than leaving content up. That pattern lands hardest on legitimate adult creators and performers, who already bear the brunt of broad moderation sweeps. It’s the same dynamic we’ve flagged repeatedly, where measures aimed at bad actors end up policing consensual adult content instead, while the actual deepfake operators keep moving, as Apple and Google’s nudify app problem shows.
“Degraded and violated”
The arrests of Hernandez and Shannon are believed to be among the first to derive from the Take It Down Act, with authorities using the charges to send a message. Whether two prosecutions amount to enforcement or symbolism is the open question, given that nudify apps reportedly racked up 700 million downloads while sitting in the major app stores, but it has to start somewhere.
Nocella said the defendants “used cutting-edge digital technology to create images that degraded and violated victims across the United States.”
He added: “This case makes clear that posting deepfake pornography is not a victimless crime, and our Office will pursue the criminals who engage in this reprehensible conduct with all the legal resources that the federal government can bring to bear, including new authorities granted by Congress to address these emerging forms of psychological, reputational, and financial abuse.”
Authorities in other countries are cracking down hard on sexually explicit nonconsensual deepfake content too. In April a man was fined AUS $343,000 (US $225,000) for posting deepfake images depicting famous women on the now-defunct site Mr Deepfakes.
In Denmark, copyright laws are set to change to give people legal rights to their face, body features and voice, in a move designed to crack down on non-consensual deepfakes.
The UK is also criminalizing the creation of non-consensual deepfake porn, alongside making non-consensual sharing illegal, though we’ve noted the gray areas that remain. As they do in the US, offenders face up to two years in prison.
