New York Synthetic Performer Advertising Law Research Note

New York Synthetic Performer Advertising Law Research Note

New York General Business Law Section 396-b requires a conspicuous disclosure in certain advertisements that use an AI-created or modified synthetic human performance. The law is narrower than a general AI advertising label.

Current legal text

The current codified statute defines a synthetic performer as a computer-created, reproduced, or modified asset produced through artificial intelligence or software that is intended to give the impression of a visual or audiovisual human performance when the performer is not recognizable as an identifiable natural performer.

The obligation applies to a covered person that conducts business involving property or services, produces or creates an advertisement for commercial purposes, and has actual knowledge that the advertisement includes a synthetic performer.

The advertisement must conspicuously disclose the use of a synthetic performer. The statute provides a civil penalty of $1,000 for a first violation and $5,000 for each subsequent violation.

The law includes exceptions for promotional material for expressive works when the synthetic use is consistent with the work, for audio advertisements, and when AI is used solely to translate the language of a human performer's performance. An advertising medium that publishes or disseminates the advertisement is excluded from the covered-person definition.

Enactment and effective date

The S8420A bill record shows that the measure was signed as Chapter 617 on December 11, 2025, with an effective date 180 days later.

Governor Hochul's June 9, 2026 announcement states that the law was in effect. The codified page reviewed on July 27, 2026 lists a most recent revision date of June 12, 2026.

The codified statute is the controlling public text for publication. The bill and governor's announcement provide enactment history and official explanation.

Operational questions

A creative system needs a flag for whether an asset depicts or modifies a human performance, whether the person is identifiable, whether the use is visual or audiovisual, whether it is a commercial advertisement, which entity created it, and whether an exception may apply.

The release workflow should preserve the source asset, generation or modification record, model or tool, reviewer, disclosure text, placement, and final rendered advertisement. The team should test whether the disclosure remains conspicuous on each placement and device.

The statute uses actual knowledge. A business should not design its workflow to avoid knowledge. Asset provenance, vendor representations, reviewer prompts, and records help determine what the company knew before release.

What the law does not resolve

The statute does not establish that a disclosed ad is truthful, non-deceptive, licensed, non-defamatory, or authorized by a person whose identity is used. Other advertising, publicity, contract, intellectual property, platform, and civil-rights rules may still apply.

The law expressly states that it does not limit rights under the New York Civil Rights Law and does not affect 47 U.S.C. Section 230.

Publication boundary

The public explainer should not say every AI-generated advertisement requires a New York label. The trigger depends on the definition of synthetic performer, the commercial advertisement, the covered entity, actual knowledge, and the exceptions.

It should use the current statute, include the enactment date and penalties, and tell readers to obtain legal review for a specific campaign. It is an editorial explainer, not legal advice.