Governor Josh Shapiro signed Senate Bill 806, the Artificial Intelligence Deceptive Advertising Act, on October 22, 2026. The disclosure can't sit in a platform panel — it has to be baked into the ad itself, in the same medium, the first time the synthetic content appears.
2026-10-24 · by Moe Ameen
On October 22, 2026, Pennsylvania Governor Josh Shapiro signed Senate Bill 806 — the Artificial Intelligence Deceptive Advertising Act — into law as Act 99 of 2026, one of 58 bills he signed that day. The bill's prime sponsor was State Senator Nick Pisciottano (D-Allegheny); it cleared the Senate 48-2 in June and the House 163-40 on October 7, 2026, with no amendments in the House. By its terms the law takes effect 60 days after signing — December 21, 2026 — which makes Pennsylvania one of the first US states to put AI-ad disclosure into statute rather than platform policy.
The rule is narrower than "label every AI ad." It bars an advertiser from using "synthetic content" — an image, audio clip, video, or text generated or significantly modified by AI — in a way that creates a false perception about the nature, origin, or authenticity of the goods or services being advertised, unless the ad carries a clear and conspicuous disclosure that it contains AI-generated or AI-altered content. A sponsor's own example: an AI-generated image that could lead a consumer to believe they are seeing the actual product must say AI was used. The format requirements are specific. The disclosure has to appear the first time the synthetic content is shown, be presented in the same medium as that content, and — where technically feasible — be permanent or "extraordinarily difficult to remove." Audio disclaimers must run at a volume and cadence a consumer can hear; on-screen disclaimers must stay up long enough to read. An ad whose only AI-touched element is text is exempt outright, so the duty is really about image, video, and audio.
Enforcement runs through the Pennsylvania Office of Attorney General and county district attorneys, with civil penalties capped at $3,000 per violation plus attorney fees and court costs. One change between the Senate floor version and the enacted text matters: the words "knowingly, recklessly or willfully" were dropped, so the disclosure duty does not turn on proving the advertiser intended to mislead. The tech-industry group NetChoice opposed the bill, arguing existing deception laws already cover the harm.
Pennsylvania joins a widening patchwork rather than a federal standard. New York's AI-advertising notice requirement took effect in June 2026, and California's SB 1050 — a different trigger aimed at AI "synthetic performers" in ads — takes effect January 1, 2027. Treat the figures above as a snapshot of the signing coverage; the controlling text is Act 99 of 2026 on the Pennsylvania General Assembly site, which is worth checking before building a compliance process around the law.
The specific thing SB 806 asks for — a disclosure in the same medium, at the first appearance, "extraordinarily difficult to remove" — is a production requirement wearing legal clothing. You cannot satisfy it with a platform's "AI info" checkbox, because that lives on one distribution surface and comes off the moment the asset moves. The compliant move is to render the label into the asset itself, and the only place that scales is the layer where the creative is generated in the first place. That is the part [Kompozy](/) handles that an ad-account toggle cannot.
Kompozy is a generation-and-publishing engine, so the disclosure rides with the asset by construction. A video ad generated as a [Persona Short](/glossary/persona-shorts) or Marketing Short carries on-screen text burned into the frame; an image or [Carousel](/glossary/hyperframes) ad rendered through HyperFrames composites the "contains AI-generated content" line into the pixels of every slide; an audio-bearing cut keeps the spoken note in the track. Because one source produces the whole batch of variants, you decide the disclosure once and it is baked into all fifty cuts instead of being re-added by hand in the ad manager — which is exactly what the per-violation penalty punishes you for missing. Then nothing ships blind: [Autopilot](/glossary/autopilot) fans the set across the eight social platforms plus blog and email, but each creative can pass a per-post review gate first — the natural single checkpoint to confirm the in-scope call and the legible label before December 21, not after an AG inquiry. Be clear on the boundary: Kompozy generates and distributes the creative and gives you the gate; it does not run your ad account or make the legal call on whether a given ad "creates a false perception" — that stays with you and, for high-stakes spends, your counsel. For the full breakdown of the law and the step-by-step compliance task, see [state AI advertising disclosure laws in 2026](/guides/state-ai-advertising-disclosure-laws) and [how to comply with state AI-ad disclosure laws](/how-to/comply-with-ai-ad-disclosure-laws).
A clear and conspicuous disclosure on an advertisement when its AI-generated or significantly AI-modified content creates a false perception about the nature, origin, or authenticity of the goods or services being sold. The label must state that the ad contains content generated or altered with artificial intelligence, appear the first time the synthetic content is shown, and be in the same medium as that content. It is not a rule to label every AI ad — the duty is tied to a misleading impression about the product.
Governor Josh Shapiro signed Senate Bill 806 into law on October 22, 2026 as Act 99 of 2026. By its terms it takes effect 60 days after signing — December 21, 2026. The bill had passed the Senate 48-2 in June 2026 and the House 163-40 on October 7, 2026.
Only the ones where the synthetic content creates a false perception about the product. Using AI to make an honest ad is not, by itself, the trigger, and an ad whose only AI-touched element is text is exempt outright. Because "creates a false perception" and "significantly modified" are subjective, the practical posture for most advertisers is that a clear label is cheaper than arguing the line — especially since the enacted law has no intent requirement.
Those are platform policies inside one ad system; SB 806 is a state statute enforced by the attorney general and county district attorneys, with civil penalties up to $3,000 per violation. They stack rather than substitute. Clearing Meta's "AI info" label or Google's "How this ad was made" panel does not satisfy the Pennsylvania statute, which also demands the disclosure be in the same medium and hard to remove — a platform panel alone would not meet that.