New York Synthetic Performer Law: What AI Advertising Disclosure Requires

New York's synthetic performer law is a narrow law with a broad operating lesson: AI-generated people in advertising are now a compliance workflow issue.

The law went into effect on June 9, 2026, according to Associated Press reporting. AP reports that advertisements in New York that use AI-generated people in place of actors must clearly disclose the use of a "synthetic performer." The same report says penalties start at $1,000 for a first violation and rise to $5,000 for later violations.

For marketing teams, this should not be handled as a last-minute label check. The real work is operational: identify synthetic performers early, approve disclosure language, test every format, and retain evidence that the final ad was properly labeled.

What the Law Is Really Targeting

AP reports that synthetic performers are defined under New York law as digitally created media that appear as real people. In a marketing context, that means the law is aimed at ads where a viewer might reasonably believe the person in the creative is a real actor, customer, employee, expert, influencer, or spokesperson.

That definition matters because many AI creative workflows blur the line. A campaign might start with a generated avatar, then use video editing, voice synthesis, motion tools, or image models to create something that looks like a real person. The compliance question is not whether the creative team thinks of the person as an "avatar." The question is whether the ad presents a person-like figure to consumers without a clear label.

The reported carve-outs are also important. AP says the law does not apply to audio-only ads, ads where AI is used solely for language translation, or certain ads for movies, TV, streaming content, video games, and similar works that feature synthetic performers throughout the underlying work.

Those carve-outs should not become a reason to skip review. A single creative asset can move across channels quickly. A video made for a product page can become a paid social ad. A generated character made for a campaign concept can become a customer-like spokesperson. The use case can change after the asset is created.

Why This Matters Outside New York

Even companies that are not based in New York may reach New York residents through national digital campaigns. Paid social, programmatic display, search, connected TV, influencer campaigns, and landing pages are often distributed broadly unless the media plan intentionally excludes a geography.

The bigger trend is also clear. Regulators and lawmakers are paying more attention to synthetic media, endorsements, dark patterns, consumer deception, and AI claims. The FTC's Disclosures 101 for Social Media Influencers is not about synthetic performers specifically, but its disclosure principle is useful by analogy: disclosures should be hard to miss and easy to understand. FINRA's 2026 Communications with the Public guidance similarly emphasizes communications that are fair, balanced, and not misleading as technology and communication practices evolve.

The common risk is consumer misunderstanding. If an AI-generated person looks like a real customer, adviser, patient, employee, investor, user, or expert, the creative can imply a real-world experience or credential that does not exist.

The Disclosure Needs to Survive the Format

The easiest mistake is assuming the disclosure is solved once a designer adds a label to the first export.

In practice, labels disappear. A vertical crop can remove a lower-third. A platform preview can cover text with interface chrome. A dark label can fail against a dark background. A caption can be truncated behind "more." A translated version can lose the disclosure entirely. A media buyer can upload the wrong asset.

The review standard should be format-aware. The disclosure should appear in the creative itself when the synthetic performer appears, use plain language such as "AI-generated synthetic performer" or "This ad uses an AI-generated performer," and remain visible across desktop, mobile, cropped social placements, muted video, and translated versions.

Do not rely on metadata, alt text, a platform disclosure tool, a terms page, or a campaign note by itself. Those may be useful supporting records, but they are not a substitute for a clear consumer-facing disclosure.

Build Synthetic Performer Review Into Intake

The right place to catch this risk is not final legal review. It is creative intake.

Every campaign intake workflow should ask whether the asset includes a generated or materially altered person, whether the person is intended to look real, whether a real person's likeness or voice was used, whether the person is presented as a customer or expert, and where the ad will run.

Those answers determine the review path. A stylized illustration may be low risk. A realistic generated customer in a testimonial-style video is not. A synthetic person in a national paid campaign should be reviewed as if New York is in scope unless the media plan clearly excludes it.

The approval record should then preserve the final creative, the disclosure copy, placement screenshots, any tool or prompt notes needed to classify the asset, and proof that the live ad kept the label.

Review moment	What to decide	Evidence to keep
Intake	Does the asset include a generated or materially altered person?	Creator response, asset notes, source files, tool context.
Classification	Could a viewer believe the person is real or acting in a real role?	Risk category and reviewer rationale.
Distribution	Will the ad run in New York or nationally?	Media plan, targeting settings, platform exports.
Disclosure approval	What label appears, where, and in which formats?	Approved disclosure text and rendered previews.
Publication check	Did the live placement preserve the label?	Live URL, screenshot, platform preview, or final ad export.

This is the kind of control Luthor is designed to support. The system should identify synthetic performer risk at intake, route the asset to the right reviewer, preserve the disclosure decision, and keep the final evidence tied to the campaign.

Agencies and Vendors Need Clear Rules

Brands will not always create the synthetic performer themselves. Agencies, production vendors, freelance creators, localization teams, and platform tools may introduce AI-generated people before the compliance team sees the asset.

That means vendor requirements need to change. Agencies should disclose AI use, identify synthetic performers, preserve source and rights information, show whether a real person's likeness or voice was used, and hand over the final disclosure evidence with the creative. If a vendor cannot identify whether an asset uses synthetic performers, the brand is taking on risk without visibility.

This overlaps with broader third-party AI risk in marketing compliance. The brand may outsource production, but it still needs a defensible review record.

What Regulated Teams Should Do Now

The immediate response should be practical. Add a synthetic performer question to creative intake. Update disclosure standards. Require agencies to disclose AI-generated people. Review current campaigns that use realistic AI people. Test labels across the placements where ads actually run. Store the final ad, disclosure placement, approval, and publication evidence in the same record.

That workflow will help with New York's law. It will also prepare the team for the broader direction of AI advertising regulation.

The lesson is not "AI creative is too risky." The lesson is that AI creative needs the same kind of controlled workflow that regulated marketing teams already use for claims, testimonials, disclosures, and approvals.

Sources and Further Reading

• Associated Press: Ads in New York must now label AI-generated "synthetic performers"
• FTC: Disclosures 101 for Social Media Influencers
• FINRA 2026 Annual Regulatory Oversight Report: Communications with the Public
• C2PA Specifications

FAQ

What is New York's synthetic performer law?

It is a New York advertising disclosure law that requires covered ads using AI-generated synthetic performers to clearly disclose that use. AP reports that it went into effect on June 9, 2026.

What is a synthetic performer?

AP reports that New York law defines synthetic performers as digitally created media that appear as real people. Marketing teams should review AI-generated actors, avatars, spokespersons, customers, experts, and influencer-style characters under this standard.

What are the penalties?

AP reports that covered ads without conspicuous disclosure may face a $1,000 penalty for a first violation and $5,000 for subsequent violations.

Does this apply outside New York?

The law is New York-specific, but national digital campaigns can reach New York unless geo-targeted otherwise. Brands should treat national AI-person campaigns as potentially in scope.

How should marketers comply?

Add a synthetic performer check to intake, approve plain-language disclosure copy, test all ad formats, and retain a review record showing the final creative, disclosure placement, approval, and live placement.