Pennsylvania has become one of the first US states to force advertisers to come clean about AI. Governor Josh Shapiro signed the Artificial Intelligence Deceptive Advertising Act into law on Thursday, requiring any ad “significantly modified or generated by artificial intelligence” to carry a disclosure in the same format as the ad itself.
The rule covers everything from big-budget TV commercials to a local restaurant’s flyer — and because it applies regardless of where an ad was created, viewers in other states could start seeing the labels too, if those ads are also set to air in Pennsylvania.
What Pennsylvania’s AI Ad Label Law Requires
The law applies to ads in any format: still images, video, audio or text. If AI significantly modified or generated the content, the disclosure must appear in that same format — so a video ad needs an on-screen or spoken label, not a footnote buried somewhere else.
That breadth is the point. AI-generated imagery and voice have become cheap and convincing enough that a polished ad no longer proves a product, a person or a claim is real. Pennsylvania’s answer is transparency: tell the viewer when a machine made — or heavily altered — what they are looking at.
The Loopholes in Pennsylvania’s AI Advertising Act
But the law leaves plenty of room for argument. “Significantly modified” is subjective, and the bill carves out parody, satire, commentary, criticism and anything “evidently fictional” — without clearly defining what counts as evidently fictional.
The wording also suggests the disclosure duty applies mainly to synthetic content used “in a manner that creates a false consumer perception” about a product or service. That raises fresh questions: who decides what counts as false, and how is it enforced? The office of State Senator Nick Pisciottano, who introduced the bill, did not immediately respond to a request for comment.
Still, Pennsylvania’s move matters beyond its borders. As one of the first states to legislate AI ad disclosures, it sets a template other states may copy — and national advertisers may find it simpler to label everything than to cut a separate Pennsylvania version of every campaign. For consumers, the takeaway is simple: the next slick ad you see may soon have to admit whether a human actually made it.
Also Read: OpenAI Now Trains Its Frontier AI Models With No Internet Access After Rogue Agents Hacked Real Companies | Block’s Dessa AI Talks Like an Awkward New Friend — Jack Dorsey’s True Speech-to-Speech Model Runs on a MacBook
Source: The Verge AI — read the full story there.
