A federal appeals court has ruled that Supreme Court precedent from 2002 still protects the private, in-home possession of AI-generated child sexual abuse material when no real child is depicted — even as two judges warned that today’s image generators have outrun the assumptions behind that precedent.
The Seventh Circuit on Aug. 25 affirmed a Wisconsin district court’s decision to dismiss one of four federal charges against Steven Anderegg. Investigators said devices seized from his home contained hundreds of AI-generated sexually explicit images of children. Anderegg was charged with producing and distributing the material, transferring it to a minor under 16, and possessing it. Only the possession count was thrown out on First Amendment grounds. The other three charges remain.
The Justice Department appealed, arguing that the government should be able to ban possession of obscene virtual child sexual abuse material even when no actual child was used to create it. The appeals court refused.
The legal framework is older than generative AI. The government generally cannot criminalize private possession of obscene material inside a home, under Stanley v. Georgia (1969). It can ban possession of child sexual abuse material involving real children, because creating that material exploits actual victims (New York v. Ferber; Osborne v. Ohio). In 2002, however, Ashcroft v. Free Speech Coalition struck down restrictions on sexually explicit images of fictional children created without using real minors. The Court held that the government’s interest in protecting children from exploitation does not apply in the same way when no real child is involved.
Judge John Z. Lee, writing for a panel that also included Judges Doris Pryor and Joshua Kolar, said that distinction still controls. Because the images at issue did not depict an actual child, the statutes that reach real-child material did not apply, and the government’s arguments for treating virtual images as unprotected had already been rejected in Free Speech Coalition. The court therefore held 18 U.S.C. § 1466A(b)(1) unconstitutional as applied to Anderegg’s private possession.
Lee and Kolar also filed a concurrence that reads as a warning to the Supreme Court. Free Speech Coalition is nearly 25 years old, Lee wrote, “and the image-generation technology available today was likely unimaginable back then.” Modern systems can produce images “virtually indistinguishable” from material depicting the abuse of real children. “Given the relentless advancement in artificial intelligence models, we have some concerns about the lines these cases draw, but we are not free to redraw them ourselves.”
Hans von Spakovsky, a senior fellow at Advancing American Freedom, told Fox News Digital the panel “had no choice but to uphold the dismissal under the First Amendment” because of that 2002 ruling. He agreed with the concurrence that the justices may need to revisit the issue. “Unfortunately, AI-generated images are now virtually indistinguishable from the real thing and some of these images can be created based on real child sexual abuse material,” he said. “Technology may have overtaken and overwhelmed the reasoning used in those prior decisions.”
Free Speech Coalition itself drew a line the government still can use: “morphing” images of real children into sexually explicit poses is not protected, because real children are involved. Justice Department lawyers have also argued that realistic AI images can stall investigations by making it harder to identify actual victims.
Lee and Kolar flagged a further issue the government did not fully develop in this case: research finding child sexual abuse images in a major dataset used to train image-generating models. That could, they suggested, create a link between supposedly virtual material and the exploitation of real children. They stressed that the record was insufficient to decide the point.
Justice Clarence Thomas anticipated the problem in 2002. If technology made virtual depictions indistinguishable from images of real abuse and thereby thwarted prosecution of unlawful material, he wrote, “the Government may well have a compelling interest in barring or otherwise regulating some narrow category of ‘lawful speech’” in order to enforce laws against material produced through the abuse of real children.
The Seventh Circuit’s message is narrow and blunt: under existing Supreme Court law, private possession of obscene virtual images that depict no real child remains protected. Production, distribution, and sending such material to a minor do not. Whether that line still makes sense is a question the court said only the Supreme Court can answer.
Be the first to comment