A federal appeals court has determined that private possession of AI-generated child pornography is shielded by the First Amendment when no real child is used in creating the images, a ruling that underscores how far the law lags behind rapidly advancing technology.
The Seventh Circuit Court of Appeals upheld the dismissal of a possession charge against Wisconsin resident Steven Anderegg, who investigators say had hundreds of AI-generated sexually explicit images depicting children stored on his devices. According to Gateway Pundit, Anderegg still faces three additional counts tied to allegedly producing and distributing the material and transferring it to a minor under the age of 16.
The Justice Department had appealed the dismissal, insisting that the government must be allowed to outlaw possession of all obscene virtual child sexual abuse material, regardless of whether a real child was directly harmed. The appellate panel, however, said its hands were tied by the Supreme Courts 2002 ruling in Ashcroft v. Free Speech Coalition, which invalidated broad restrictions on sexually explicit depictions of fictional children where no actual minors were exploited.
That nearly quarter-century-old precedent left two of the judges openly uneasy about applying it to todays AI-driven imagery. Indeed, in Free Speech Coalition, the Supreme Court addressed the scope of First Amendment protections for virtual CSAM [child sexual abuse material], but that was nearly twenty five years ago, and the image generation technology available today was likely unimaginable back then, Judge John Lee wrote.
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. The result is a deeply troubling legal distinction in which possession of child sexual abuse material involving real children remains criminal because identifiable victims were abused, while wholly artificial depictions enjoy constitutional protection when privately held at home.
Judges Lee and Joshua Kolar pointed to research uncovering child sexual abuse images embedded in a major dataset used to train image-generating AI systems. That evidence suggests that supposedly virtual material may still be tainted by the exploitation of real children, even if the final product appears synthetic.
The judges acknowledged that this concern was not sufficiently developed in Andereggs case to justify a different outcome at this stage. Justice Clarence Thomas foresaw this dilemma as early as 2002, warning that technological progress could render virtual depictions indistinguishable from images involving actual victims, leaving courts struggling to protect children without eroding core free-speech protections.
For now, the ruling stands as a stark reminder that constitutional doctrine has not caught up with AI, and it will remain in force unless the Supreme Court agrees to revisit Ashcroft v. Free Speech Coalition and draw new lines for an era in which digital fabrications can be weaponized against societys most vulnerable.
Login