The National Center on Sexual Exploitation says it’s time for the nation’s highest court to close a legal gap that currently lets people possess AI-generated child sexual abuse material, so long as the images depict no real child.
“While most Americans would agree that AI-generated child sexual abuse material of a virtual child is heinous and should be illegal, the recent ruling is consistent with case precedent stemming from the 2002 Ashcroft v. FSC ruling which holds that ‘virtual CSAM’ is protected by the First Amendment,” said Dani Pinter, Chief Legal Officer and Director of the Law Center at the National Center on Sexual Exploitation, known as NCOSE in a press release.
NCOSE released its statement Tuesday, days after the 7th Circuit Court of Appeals ruled that private possession of such material is protected speech, as long as it doesn’t depict an identifiable real child. The case, U.S. v. Anderegg, involved a Wisconsin man found with hundreds of AI-generated sexually explicit images of children; the Aug. 25 ruling threw out only his possession charge, not the other counts against him.
Judge John Lee, writing for the panel, admitted the court’s hands were tied by 24-year-old precedent, even though “the image-generation technology available today was likely unimaginable back then.” He and Judge Joshua Kolar all but begged the Supreme Court to take a fresh look at the question now that generative AI has entered the picture.
Pinter noted that Congress already tried to patch part of the hole Ashcroft left open, amending the law to criminalize virtual CSAM that depicts an identifiable child. What she argues the 2002 justices never saw coming was AI trained on the abuse of actual kids.
“That said, we know that AI models are trained with real children, and there have been reports that some models have even been trained with CSAM depicting real children being abused. This presents a very different set of facts and legal considerations that were not contemplated by the 2002 Supreme Court. Americans would certainly welcome the Supreme Court revisiting this critical issue at its earliest opportunity,” Pinter said in her statement.
Related: NCOSE Just Put Mark Zuckerberg on Its 2026 Dirty Dozen List
Beyond the legal question, NCOSE argues the material is actively dangerous. Pinter said predators use AI-generated CSAM to groom victims, pointing to reports from the National Center for Missing & Exploited Children that children and their families have been sextorted with such images. Because the material is so cheap and easy to generate, she warns, demand for it will only grow.
“AI technology is rapidly changing, and given the industry has no meaningful guardrails, it is impossible to trust AI companies to completely safeguard children,” Pinter said, calling on AI developers to make sure their models are never trained on CSAM of real children or footage of adult sexual abuse.
Founded in 1962, NCOSE has spent decades exposing the links between pornography, prostitution, sex trafficking and child sexual abuse, and its Law Center regularly files briefs urging the Supreme Court to take up cases like this one. Movieguide® has tracked the wider reckoning over tech companies and children’s safety this year, from TikTok and ByteDance’s $400 million settlement over child-privacy violations to Meta’s $17 billion settlement over apps built to hook young users.
AI moves faster than any court calendar, and the technology behind an image generator doesn’t pause to ask whether Congress or the Supreme Court has caught up to it yet. Until the law does catch up, NCOSE’s message to families is a familiar one dressed in new circumstances: know what’s on the device in your child’s pocket, and don’t assume the law is already protecting them from what AI can now create in seconds.
Read Next: Tim Tebow Speaks at INTERPOL to Fight Child Sex Abuse
Questions or comments? Please write to us here.
