California SB 1050: AI Performers in Ads Must Be Labeled

LOS ANGELES — The smiling spokesperson telling you to buy the mattress, the voice narrating the car commercial, the impossibly perfect influencer demoing a skincare line: starting January 1, 2027, if they're built from generative AI, California law says you have to be told.

On September 16, 2026, Governor Gavin Newsom signed SB 1050 into law, a measure that takes effect January 1, 2027 and makes it an unlawful advertising practice to run an ad prominently featuring a "synthetic performer" — an AI-created figure, voice, or likeness so realistic you can't tell it isn't human — without a clear disclosure saying so. Newsom signed the bill at the Los Angeles headquarters of its sponsor, SAG-AFTRA, the 160,000-member performers' union that helped write it. The California Legislature passed the bill on the final day of the 2026 session, August 31.

The signing puts California alongside New York as the only states in the country to directly regulate AI-generated people in advertising. And it drew one of the sharpest fights yet between organized labor and the advertising business over how far AI disclosure rules should reach.

California Governor Gavin Newsom signed SB 1050, requiring disclosure of AI-generated performers in ads. Photo: Office of the Lieutenant Governor of California / Wikimedia Commons (public domain).

What SB 1050 actually requires

Authored by State Senator Angelique Ashby (D-Sacramento), the bill adds a new article to the California Business and Professions Code. The core rule is straightforward: it is unlawful for any person to create — and cause to be published in an advertising medium — an advertisement that prominently includes a synthetic performer without a "clear and conspicuous" disclosure that the performer is synthetic.

The law's definitions are doing the heavy lifting. A synthetic performer is a humanlike digital figure, voice, or representation created in whole or in part using generative AI that creates the impression of an audio, audiovisual, or visual performance of a human performer "who is not recognizable as any identifiable natural performer." In other words: a realistic but fictional person. A stylized robot or cartoon mascot isn't covered; a photorealistic AI spokeswoman is.

"Prominently" means the synthetic performer appears in the foreground demonstrating or illustrating the product or service, narrates the advertisement, or illustrates or reacts to the narration. And "clear and conspicuous" means a disclosure that is difficult to miss and easily understandable — presented so that a reasonable consumer would notice, read, and comprehend it, given the medium, format, and context. The bill suggests wording "substantially similar" to "this performance features a synthetic digital performer" or "no human performer is depicted."

Coverage is wide. An advertisement is any audio, video, or audiovisual message disseminated in any manner or by any means — including online platforms — intended, or reasonably expected, to induce the purchase of goods or services. Audio ads count, which matters for podcasts, streaming radio, and voice assistants.

There are carve-outs. The disclosure rule does not apply to advertisements for expressive works — movies, TV shows, streaming content, video games, documentaries — as long as the synthetic performer's use in the promo is consistent with its use in the work itself. Pure language translation of a human performer is exempt. And the bill explicitly says it does not restrict the creation or exhibition of synthetic content, nor does it regulate the expressive or informational content of an ad.

How it's enforced — and the takedown provision everyone is watching

Violations of SB 1050 count as violations of California's False Advertising Law (Business and Professions Code § 17500), enforceable through the Unfair Competition Law (§ 17200) — the same statutes that underpin much of California's consumer-protection litigation. The law also bans the continued use of an ad found to be in violation.

The most consequential piece may be the court-order removal mechanism. If a court issues an order finding that an ad violates the law, and the "advertising medium" — a broadcast station, cable operator, streaming service, online platform, digital ad network, or publisher that distributes or displays ads to California consumers — is served with that order and enough information to identify the ad, the medium must remove or disable access to it "as commercially reasonable and technically feasible," and stop accepting payment for it. Entities not capable of removing or disabling an ad are excluded from the definition.

That provision is exactly what alarmed legal observers. Santa Clara University law professor Eric Goldman told MediaPost that the publisher-removal requirement "appears inconsistent with Section 230 of the Communications Decency Act," the federal law shielding interactive services from liability for third-party content. Goldman noted that some judges have held Section 230 bars courts from ordering online publishers to remove content at all. "As a practical matter, I don't know what publishers would do if they received one of these court orders," he said. "But in the abstract, not knowing the situation of any individual publisher, it looks like a 230 violation."

Supporters answer that the regulated party is the advertiser, not intermediaries — platforms face obligations only after a court order, and the disclosure duty sits squarely on whoever created the synthetic spot.

Why Sacramento acted now

Newsom framed the bill as a transparency measure for a moment when generative tools can conjure a convincing spokesperson in minutes, for dollars rather than the cost of a production shoot.

"Californians deserve to know when the person selling them something isn't a person at all. Here in California, we're building on our nation-leading AI regulations and putting commonsense guardrails in place to protect consumers, performers, and promote stronger transparency. As AI transforms the entertainment industry, we're protecting workers while ensuring California's creative economy can continue to grow and thrive." — Gov. Gavin Newsom, September 16, 2026

Sen. Ashby put the labor case bluntly:

"I'm grateful to Governor Newsom for signing SB 1050. This bill provides important protections for the creative economy, who are the living heartbeat of California. The bill requires that synthetic figures be labeled as such, thus protecting consumers from false advertising and disallowing AI to replace real people — real people who work every day to bring stories and brands to life in the production of all entertainment across the great Golden State." — State Sen. Angelique Ashby (D-Sacramento), September 16, 2026

SAG-AFTRA, which sponsored the legislation, called the law a transparency win for consumers and performers alike:

"We thank Governor Newsom for signing SB 1050 and Senator Ashby for her leadership in authoring this important legislation," said SAG-AFTRA National Executive Director & Chief Negotiator Duncan Crabtree-Ireland. "SAG-AFTRA was proud to help craft the language of the bill, which gives consumers greater transparency by making clear when they are seeing synthetics rather than human performers. This is an important step in the broader effort to protect consumers from deceptive AI practices, and to ensure that AI technology isn't used at the expense of human creativity and consent."

The union's fingerprints are all over the bill's official legislative history. In the Assembly committee analysis, SAG-AFTRA argued the disclosure must be "clear, conspicuous, and understandable to a consumer," and warned that synthetic performers in advertising both drain jobs from creative performers and risk deceiving viewers "who may have no idea they are not seeing or hearing a human being." E-commerce companies, the analysis notes, are increasingly turning to generative AI for marketing content — a trend law firm Greenspoon Marder flagged as the law's likely first enforcement frontier.

SAG-AFTRA Plaza, the union's national headquarters on Wilshire Boulevard in Los Angeles, where Newsom signed SB 1050. Photo: Shaunti Griffin / Wikimedia Commons (CC BY-SA 4.0).

The ad industry tried to kill it

The bill faced organized opposition from the start. On September 4, four major ad-industry groups — the Association of National Advertisers (ANA), the American Association of Advertising Agencies (4As), the American Advertising Federation (AAF), and the Digital Advertising Alliance — sent Newsom a letter urging him to veto the legislation, according to MediaPost, which reported the lobbying on September 18.

Their language was unusually pointed:

"Rather than targeting deceptive conduct, the bill establishes an overbroad mandatory labeling regime, exposes businesses to opportunistic litigation, and places severe burdens on ordinary commercial speech." — ANA, 4As, AAF, and Digital Advertising Alliance, September 4, 2026 letter to Gov. Newsom

The groups argued the disclosure requirement applies "regardless of whether a synthetic performer is likely to deceive a consumer regarding a material claim," and "mandates disclaimers even when a synthetic performer is used for purely illustrative, background or thematic purposes." Requiring labels where no consumer is actually misled, they wrote, creates "disclosure overload" that "dilutes consumer awareness" and "ultimately undermin[es] transparency where it would be more genuinely helpful to consumers." They also raised First Amendment concerns about compelling speech from advertisers and intermediaries.

The veto bid failed. But the arguments map closely onto the ones the 4As made against New York's version, when the group warned the law would work by "injecting compliance uncertainty into the advertising process, burdening brands (and their agencies) who advertise in New York and undermining creative and technological innovation." The industry's playbook — comply, but keep fighting in court and in other statehouses — is now the baseline expectation for California too.

New York did it first. California did it differently.

California is the second state to mandate synthetic-performer disclosure in ads. New York's law, signed by Gov. Kathy Hochul in December 2025, took effect in June 2026. But the two states built different machines, and companies can't assume one compliance plan covers both.

New York requires a conspicuous disclosure whenever an advertiser with actual knowledge uses a synthetic performer in an ad, and enforces it through statutory civil penalties: $1,000 for a first violation, $5,000 for subsequent violations. It also carves out audio-only advertisements entirely, along with promos for expressive works and ads where AI was used solely for language translation.

California takes the other fork. It applies only when the synthetic performer is "prominently" featured — a narrower trigger — but expressly includes AI-generated voices, so audio ads are in scope here. Instead of fixed statutory fines, enforcement runs through California's False Advertising Law and Unfair Competition Law, a familiar framework that gives both public prosecutors and qualifying private litigants a path to sue. As Greenspoon Marder's analysis put it, the two states "take notably different approaches," and businesses "should not assume that compliance in one state will satisfy the requirements of the other."

The bigger fight this fits into

SB 1050 is the latest front in a labor war over synthetic humans that has been escalating since the 2023 Hollywood strikes, when digital-replica protections were a central issue in SAG-AFTRA's negotiations with the studios. In May 2025, the union filed an unfair-labor-practice charge with the National Labor Relations Board against Llama Productions, the Epic Games subsidiary behind Fortnite, after the game rolled out an AI-generated Darth Vader voiced with a replica of the late James Earl Jones's voice. Jones's estate had given permission — but the union's grievance was about the work itself: "Fortnite's signatory company, Llama Productions, chose to replace the work of human performers with AI technology," SAG-AFTRA said, "without providing any notice of their intent to do this and without bargaining with us over appropriate terms."

Newsom's signature also lands inside a dense thicket of California AI regulation. The September 16 press release highlighted a flurry of related measures: SB 813, making California the first state to require AI assessments by independent verification organizations; AB 1405, requiring third-party audits of AI systems; the 2024 California AI Transparency Act on AI watermarking and detection tools, strengthened in 2025 for large online platforms; and 2025's SB 53, the Transparency in Frontier Artificial Intelligence Act. That 2024 package included AB 2602, requiring contracts to specify uses of AI-generated digital replicas of a performer's voice or likeness with the performer professionally represented, and AB 1836, barring commercial use of deceased performers' digital replicas without estate consent. SB 1050 extends the logic from Hollywood contracts to the ad breaks in between.

What this means for advertisers, agencies, and platforms

For national brands, the practical fallout starts with inventory. Marketing teams will need a searchable catalog of every digital asset that might contain a synthetic performer — including AI voiceovers — and vendor contracts that require agencies, studios, and influencers to disclose AI use in deliverables, with representations, warranties, and indemnities to match. That's the advice coming from firms like Greenspoon Marder, and it applies doubly to companies that treated AI spokesperson generation as an internal production shortcut.

Agencies face the trickiest position. New York's actual-knowledge standard at least gives an advertiser cover when an agency slips a synthetic figure into a spot quietly. California's structure — unlawful for any person to "create and cause to be published" a non-compliant ad, enforced through the UCL's broad standing — leaves more room for private litigants to test who in the production chain is on the hook.

Platforms and ad networks, meanwhile, should read the court-order removal provision closely. The Section 230 question Goldman flagged won't be answered by the statute's text — it'll be answered the first time a California court orders a publisher to pull a non-compliant ad and the publisher objects. Expect that litigation. The law carries a severability clause, meaning a court could strike the removal provision without taking the whole statute down.

And the compliance map is only getting bigger. The Transparency Coalition, which advocates for AI disclosure legislation, called California's signing a template other states will copy. If they do, the state-by-state divergence — prominent-feature triggers versus actual-knowledge standards, fixed fines versus UCL litigation — becomes a national advertiser's full-time job.

FAQ: California's synthetic-performer ad law

What is a "synthetic performer" under SB 1050?

A humanlike digital figure, voice, or representation created in whole or in part using generative AI, that creates the impression of a performance by a human who isn't recognizable as any real person. Think AI-generated spokespeople, presenters, and voiceovers — not cartoon mascots or stylized robots.

What disclosure does the law require?

A clear and conspicuous statement that the performer is synthetic, using wording substantially similar to "this performance features a synthetic digital performer" or "no human performer is depicted." It must be difficult to miss and understandable to a reasonable consumer, given the medium and format.

Does it cover audio ads?

Yes. Unlike New York's law, which exempts audio-only ads, California's definition of "advertisement" expressly includes audio messages — podcasts, streaming radio, and voice-assistant spots are all in scope.

Who can be held liable?

The duty sits on anyone who creates and causes a non-compliant ad to be published. Violations are treated as false-advertising violations enforceable under California's Unfair Competition Law, which both public prosecutors and qualifying private parties can invoke. Advertising mediums (platforms, publishers, ad networks) must remove ads only after being served with a court order identifying the violation.

What are the exemptions?

Ads promoting expressive works (films, TV, games, streaming content) where the synthetic performer's use matches the work itself; ads where AI was used solely for language translation of a human performer; and the bill doesn't restrict creating or exhibiting synthetic content generally.

When does the law take effect?

January 1, 2027. The bill carries no urgency clause, so California's standard effective date applies — advertisers have roughly three and a half months from the September 16 signing to audit their creative and update vendor contracts.

How does this differ from New York's law?

New York triggers on any synthetic-performer use with actual knowledge, uses fixed civil penalties ($1,000/$5,000), and exempts audio-only ads. California triggers only on "prominent" use, includes AI voices, and enforces through the False Advertising Law and UCL instead of fixed fines.

Sources

• Office of Governor Gavin Newsom, "Governor Newsom signs new law to protect workers, require disclosures on AI-generated advertising," September 16, 2026 — gov.ca.gov
• MediaPost, "Gov. Newsom Signs Law Requiring Disclosure Of AI-Generated Performers In Ads," September 18, 2026 — mediapost.com
• Greenspoon Marder LLP, "California SB 1050 Targets AI-Generated Synthetic Performers," by Andrew Jablon and Aliza Malouf — gmlaw.com
• Assembly Committee on Arts, Entertainment, Sports, and Tourism / Judiciary — SB 1050 (Ashby) bill analysis, June 2026 — billtexts.s3.amazonaws.com
• Transparency Coalition, "California joins New York in requiring disclosure of synthetic performers in ads," September 16, 2026 — transparencycoalition.ai
• AFP via The Straits Times, "New California law mandates AI performer disclosure in ads," September 17, 2026 — straitstimes.com
• afaqs / AP, "New York mandates disclosure of AI-generated performers in ads," June 2026 — afaqs.com
• Eurogamer, "SAG-AFTRA files unfair labour practice charge against Fortnite maker Epic, following AI-generated Darth Vader voice lines," May 2025 — eurogamer.net