New York’s law isn’t the beginning of this story. It’s the latest chapter.
The AI Synthetic Performers Disclosure Law (SB-8420A), effective June 9, 2026, requires conspicuous on-screen disclosure for AI-generated synthetic human-like performers in commercial advertisements shown to New York consumers. Civil penalties run from $1,000 (first violation) to $5,000 (subsequent violations), according to legal analysis from Hunton Andrews Kurth. The primary government source is currently non-resolving. All specific legal claims in this section are attributed to Hunton Andrews Kurth.
Two exemptions apply per that analysis: expressive works promotions (film, TV, video games) where synthetic performer use is consistent with the work, and audio-only advertisements. The law uses the word “conspicuous” without defining it. No state guidance has specified format, font size, or placement. The standard defaults to what a reasonable viewer would notice.
That’s the New York brief. The deeper question is what New York tells compliance teams about where this is going.
The State-Level Pattern
State AI synthetic content disclosure laws share a structural logic: they respond to the same problem, AI-generated human likenesses in commercial and political contexts, with similar but not identical obligations. The differences in scope, penalty, and exemption structure are where compliance complexity lives.
New York’s SB-8420A focuses specifically on commercial advertising. The disclosure obligation applies when synthetic performers appear in ads shown to New York consumers. The law’s scope is commercial use; it isn’t a general deepfake prohibition.
California has enacted AI-related disclosure requirements in both the political advertising context (synthetic media in political ads require disclosure) and, under separate statutes, in the digital replica and voice cloning context. California’s framework extends into employment, the state requires consent for digital replicas of workers, addressing a distinct use case from the advertising context NY’s law covers.
Texas has enacted a prohibition on synthetic political media (deepfakes in elections), framed as a prohibition rather than a disclosure requirement. The Texas approach addresses the same underlying technology but through a different legal mechanism and with a narrower scope.
Unanswered Questions
- Where does composited imagery, digitally altered human talent combined with synthetic elements, fall under the 'AI-generated' definition in each state's law?
- Does geo-targeting to exclude New York users satisfy the law's scope, or does any New York consumer reachable through the distribution channel trigger the obligation?
- How does the expressive works exemption apply to branded entertainment content, sponsored content for a streaming series, for example, that promotes a brand inside a fictional work?
What these laws share: they all start from the recognition that AI-generated human likenesses require some form of disclosure or limitation in high-stakes contexts. What they don’t share: the specific contexts covered, the disclosure format required, the penalty structure, and the exemption framework. A national advertising campaign can simultaneously comply with New York’s commercial disclosure requirement, fall outside California’s political advertising rule, and be entirely unaffected by Texas’s political media prohibition, while still potentially violating state laws the compliance team hasn’t mapped yet.
The Federal Gap
The Copyright Office has been explicit about the problem. The USCO’s position, addressed in prior pipeline coverage, is that deepfakes require congressional action, the existing copyright framework doesn’t provide adequate protection against unauthorized synthetic replicas of real people, and state-level patchwork creates inconsistent protection across jurisdictions.
Congress hasn’t acted. The absence of federal synthetic content disclosure legislation isn’t an oversight, it reflects the same pre-emption dynamics visible in the broader federal AI governance debate. A federal synthetic content disclosure law would pre-empt state variations, providing a single compliance floor. It would also reduce state flexibility to address jurisdiction-specific harms. The political economy of pre-emption makes federal action on synthetic content disclosure no more straightforward than federal action on broader AI governance.
The bipartisan federal AI bill circulating includes a workforce title and a pre-emption clause for state AI development and safety laws. Whether a synthetic content disclosure provision would fall within that pre-emption scope is unknown, the bill hasn’t been formally introduced and the scope of “AI development and safety” laws isn’t defined in publicly available draft text.
The Compliance Posture for National Advertisers
Don’t expect federal harmonization on a useful timeline. The Copyright Office’s recommendation is years from any legislation, and the federal AI governance debate has more urgent priority items consuming congressional attention.
The minimum defensible posture for national advertisers is a state-by-state disclosure mapping exercise. That means: identify every state with an enacted AI synthetic content disclosure obligation affecting commercial advertising; map your active campaigns against those obligations; document where exemptions apply and why; and build disclosure capability into your production and trafficking workflows rather than treating it as a one-time compliance task.
The NY law is the compliance floor for this week. It won’t be the ceiling. States with pending legislation, and there are several, are watching New York’s enforcement posture to calibrate their own frameworks. The pattern suggests a trajectory: more states, broader scope, and eventually federal action that either harmonizes or displaces what states have built.
What national advertising campaigns need right now is an AI content audit protocol that can be applied consistently across jurisdictions. That means: a documented definition of what counts as an AI-generated synthetic performer in your production workflows; a process for flagging campaigns that meet the definition; a geo-targeting or distribution-mapping capability to identify which state disclosure requirements apply; and a disclosure format that satisfies “conspicuous” under any state’s standard, not just New York’s.
What to Watch
The real question compliance teams should be asking isn’t “Do we comply with the NY law?”, that answer is yes, and it needs to happen before Monday. The question is whether your compliance infrastructure is built to handle the next state that enacts a disclosure requirement, and the one after that, without requiring a new project each time.
What’s Coming
Several states have synthetic content disclosure bills in various stages of legislative consideration. The specific states and bill numbers aren’t confirmed in available sources ; this section doesn’t name them to avoid fabricating details. What the pattern indicates: the legislative momentum is toward more disclosure requirements, not fewer.
The federal copyright debate is the longer arc. Congressional action on deepfakes, if it comes, will likely address the political and non-commercial contexts first, the electoral deepfake problem is politically easier to legislate against than the commercial advertising context because it has broader bipartisan support. Commercial advertising disclosure will follow, not lead.
The Copyright Office’s recommendation is the reference point. When Congress acts on the USCO’s deepfake guidance, the commercial advertising context will be in scope. The disclosure obligations that brands are building state by state right now will either be harmonized under a federal standard or found to be insufficient. Either outcome argues for building compliance infrastructure that can adapt, not one-time compliance fixes for each new state law.
New York’s June 9 effective date is the immediate deadline. The compliance infrastructure question is the durable one.