Pennsylvania sends the governor a bill against AI sexual deepfakes: enforcement will be the real test

By WGAL · October 23, 2026.

A Pennsylvania (U.S.) bill to toughen the response to sexually explicit deepfakes generated with artificial intelligence has cleared the legislature and awaits the governor's decision, according to local station WGAL. A deepfake is synthetic content, image or video, that realistically imitates a real person. The story's URL suggests the bill includes mandatory reporting obligations involving minors. We do not have the text of the article, so we cannot confirm its penalties, deadlines or definitions. Those details are what determine whether the law works.

Our take: this kind of measure tends to pass easily because the harm is concrete and identifiable, and almost nobody defends it. It is the AI regulation least at risk of sliding into panic: it punishes a documented abuse, not a hypothetical capability. In general, U.S. states have long been legislating on this issue on their own in the absence of a federal law, creating an uneven regulatory patchwork.

The key question is the same as with any AI safety commitment: what happens when it is breached. A law against deepfakes is only as good as its investigative capacity, the speed with which content is taken down and the real protection it gives victims. If the duty to report falls only on those who already know about a case, it will do little against material circulating on platforms and tools outside the jurisdiction.

In the short term, the ease of generating this content will continue to outpace the law's ability to prosecute it. In the long term, the same technology can improve the detection and traceability of synthetic content. But that requires governance that is evidence-based and enforced, not just laws passed to applause.