Pennsylvania AI Generated Ads Disclosure Law Explained: Labels From December 21, What Counts and What Is Exempt (USA Tech)

Illustrative image: the Philadelphia skyline — Pennsylvania's largest city, where the new advertising rules will apply. Image: Wikimedia Commons (August 2007 skyline photo).

TL;DR: Pennsylvania AI generated ads disclosure law explained: from 21 December 2026, Pennsylvania requires a clear label on advertisements whose AI-generated or heavily AI-altered images, video or audio could mislead shoppers about the product or service. It is not a ban on AI in advertising — and it is not a label on every AI-assisted ad. The trigger, the exemptions and the $3,000 penalty are narrower than the headlines suggest.

📌 KEY HIGHLIGHTS

• Signed 22 October 2026: Governor Josh Shapiro signed the Artificial Intelligence Deceptive Advertising Act (Senate Bill 806, now Act 99 of 2026).
• In force 21 December 2026: the law takes effect 60 days after signing, giving advertisers under two months to comply.
• What triggers a label: an ad that uses AI-generated or significantly AI-modified content in a way that creates a false consumer perception about the nature, origin or authenticity of the goods or services advertised.
• What is exempt: content that is evidently fictional, parody, satire, commentary or criticism — and, under the bill text, content limited to text.
• Enforcement: Pennsylvania's attorney general and district attorneys can seek injunctions and civil penalties of up to $3,000, plus legal fees and costs.
• National spillover: the rule applies regardless of where an ad was made, so an ad produced elsewhere that is expected to air in Pennsylvania needs a label too.

What happened?

Pennsylvania has become one of the first US states to put a disclosure rule on AI in advertising. Governor Josh Shapiro signed the Artificial Intelligence Deceptive Advertising Act on Thursday 22 October 2026, one of 58 bills signed that day. The measure began as Senate Bill 806, introduced by state Senator Nick Pisciottano, passed the state Senate 48–2 on 10 June and the House 163–40 on 7 October, and is now recorded as Act 99 of 2026.

The timing matters. The law takes effect on 21 December 2026 — 60 days after signing — which lands squarely in the holiday advertising season, when brands are running their heaviest campaigns. Advertisers do not get a long runway: the compliance clock is already running.

Pennsylvania is not alone in moving here. California's governor signed a law last month requiring ads using "synthetic" performers to disclose that the figures were AI-generated. Pennsylvania's version is broader in one way — it covers any misleading synthetic content, not just digital people — and narrower in another, as the next section explains.

Pennsylvania AI generated ads disclosure law explained: what triggers a label

The most misunderstood part of this law is its trigger. Several early reports described it as a blanket requirement to label all AI-generated ads. The statute is more targeted than that. A label is required only when an advertisement uses "synthetic content" — an image, video, audio clip or text significantly modified or generated by AI — in a way that creates a false consumer perception about the nature, origin or authenticity of the goods or services being advertised.

In plain terms: an AI-polished product shot that still shows the real product is a different case from an AI-generated image that makes shoppers believe they are looking at a photo of the actual product, a real person endorsing it, or a real place it was made. The law targets the second kind — the deceptive case — not ordinary AI assistance in production.

Question	What the law says
What needs a label?	An ad whose AI-generated or significantly AI-modified content creates a false perception about the nature, origin or authenticity of the advertised goods or services.
What must the label say?	That the advertisement contains content modified or generated using artificial intelligence — clear and conspicuous, in the same medium as the ad.
Where must it appear?	When the synthetic content is first presented; on screen long enough to read for video, at an audible volume and pace for audio, and permanent or very hard to remove where technically feasible.
What is exempt?	Content that is evidently fictional; parody, satire, commentary or criticism; and — under the bill text — content limited to text.
Who enforces it?	Pennsylvania's attorney general and district attorneys — injunctions and civil penalties of up to $3,000, plus fees and costs. No private right of action.
When does it start?	21 December 2026, 60 days after the 22 October signing.

Illustrative image: advertising signs in Times Square — the law covers TV spots and print flyers alike, from big-brand commercials to a local restaurant's leaflets. Image: Wikimedia Commons / Eden, Janine and Jim, CC BY.

Why the wording leaves room for argument

The law's open terms are its biggest weakness, and critics on both sides have spotted them. "Significantly modified" is subjective — nobody can say in advance how much AI editing crosses the line. "Evidently fictional" is undefined, and it is unclear what exactly counts as a "false consumer perception" or who decides. The sponsor's office did not immediately respond to requests for comment when the signing was reported.

Industry opposition came before the signing. The tech group NetChoice wrote to lawmakers arguing that existing federal and state deception laws already cover misleading ads, and that prescriptive format rules risk overwhelming short videos, small display ads and audio spots. Supporters, including the sponsor, frame it as common sense: as Senator Pisciottano put it when the bill passed, consumers deserve to know if AI is being used to create a false impression about a product or service.

One more clarification worth making, because some coverage gets it wrong: several reports list text among the formats covered, but the final bill text expressly excludes content "limited to text". The disclosure rule is aimed at AI-generated or altered images, video and audio. A machine-written headline alone does not trigger it; a synthetic product photo or voiceover can.

Why it matters for USA readers — and for India

For American readers, the practical effect is likely national, not local. Because the rule follows the ad rather than the advertiser — it applies to any ad expected to air in Pennsylvania, wherever it was produced — national brands are unlikely to cut a Pennsylvania-only version of a commercial. Expect labels to appear in ads shown in other states too, and expect other states to copy the model. Pennsylvania is positioning itself as an early mover in a patchwork that advertisers will probably standardise around, rather than geofence.

For India, the contrast is instructive. India is still at the consultation stage: the government has said a consultation paper on AI legislation is coming in the coming weeks, covering safety, deepfakes and AI harms, as TechByCountry covered in its India AI legislation consultation paper explainer. Pennsylvania shows what the next step looks like once consultation ends — a short, targeted statute with a hard deadline and a modest penalty. India's debate will face the same drafting problem Pennsylvania has just exposed: writing a rule tight enough to catch deception without labelling every AI-assisted creative. The same week also saw Sweden restrict ChatGPT and Claude on government work computers, a reminder that AI governance is now moving on several fronts at once — workplace rules in Europe, advertising rules in the US, and a framework still being drafted in India. The UK, meanwhile, has gone the institution-building route with its £100 million AI biosecurity lab.

Should you care? ✅ / ⏳

✅ Yes — if you advertise, create ads, or buy from them: from 21 December, a misleading AI-generated product shot or voice in Pennsylvania needs a label, and the rule follows ads made anywhere. ⏳ Watch the grey areas: "significantly modified" and "false perception" are untested, so the first enforcement cases — not the statute — will define the real boundary. 🔍 Honest caveat: the $3,000 penalty cap is small enough that some brands may treat it as a cost of doing business; the real pressure is likely reputational, and the disclosure standard itself, once it becomes the national default.

FAQs

Does Pennsylvania require labels on all AI-generated ads?
No. The law is not a blanket labelling rule. A label is required only when AI-generated or significantly AI-modified content creates a false consumer perception about the nature, origin or authenticity of the goods or services advertised.

When does the Pennsylvania AI ad law take effect?
On 21 December 2026, 60 days after Governor Josh Shapiro signed the Artificial Intelligence Deceptive Advertising Act (Senate Bill 806, Act 99 of 2026) on 22 October 2026.

What ads are exempt from the Pennsylvania AI disclosure law?
Content that is evidently fictional, parody, satire, commentary or criticism is excluded, and the bill text also excludes content limited to text. The rule targets misleading AI images, video and audio.

What happens if an advertiser breaks the law?
Pennsylvania's attorney general and district attorneys can seek injunctions and civil penalties of up to $3,000, plus legal fees and costs. There is no private right of action for individual consumers.

Does the law apply to ads made outside Pennsylvania?
Yes, if they are expected to air in Pennsylvania. The rule applies regardless of where an advertisement was produced, so a commercial cut in another state needs a label if it will run in Pennsylvania.

Takeaway: Pennsylvania has not banned AI in advertising — it has banned hiding it when it misleads. From 21 December, the question for every campaign is not "did you use AI?" but "could a shopper be fooled by it?"

Follow USA Tech on TechByCountry for tomorrow's update.

📸 Get tech in the last 24 hours on Instagram — follow @_techinlast24hrs for daily posts (9 AM–9 PM) and reels at 12:30 PM & 8:30 PM.

Sources

• The Verge — Pennsylvania will require labels on AI-generated ads (22–23 Oct 2026, as syndicated) — signing, scope, exemptions and open terms.
• Silicon Noon — Pennsylvania's AI ad-label law starts Dec. 21, fines capped at $3,000 — Act 99 of 2026, Senate 48–2 / House 163–40 votes, 60-day clock.
• Daily AI Brief — Pennsylvania Enacts AI Disclosure Law for Deceptive Ads (Loeb & Loeb analysis) — disclosure format, text-only exclusion, $3,000 penalty, no private right of action.
• Let's Data Science — Pennsylvania Enacts AI Disclosure Law for Misleading Ads — General Assembly record, Act 99, effective 21 December.
• MediaPost — Pennsylvania Passes Bill Requiring Disclosure Of Misleading AI In Ads (8 Oct 2026) — sponsor quote, NetChoice opposition, California comparison.