Artificial intelligence is changing the way businesses create advertisements. A company can now use an AI-generated spokesperson, build a video without hiring a human performer, and adapt the same campaign for several markets.
That convenience also creates a practical question: when does a business need to tell the audience that the person in the advertisement is not real?
New York now requires a disclosure in certain commercial advertisements that use AI-generated “synthetic performers.” For California businesses, the law matters because digital campaigns do not always stay within one state. An advertisement created in California may still reach New York through social media, streaming services, online advertising, or a national campaign.
The short answer: If a business knowingly produces or creates a commercial advertisement that contains a qualifying synthetic performer, New York law requires a conspicuous disclosure in the advertisement. The law took effect June 9, 2026. It does not ban AI-generated performers, and it does not require one exact disclosure sentence.
What Does New York’s AI Advertising Law Require?
Under New York General Business Law Section 396-b, a person or business that produces or creates a commercial advertisement must conspicuously disclose when the advertisement contains a synthetic performer, provided that person or business has actual knowledge of the synthetic performer’s use.
A synthetic performer is a computer-created or computer-modified asset that appears to be a human performing on screen, but is not recognizable as an identifiable real person. A realistic AI spokesperson, virtual salesperson, or humanlike figure in a product demonstration may fit the definition.
The law does not require the business to use a particular phrase. Language such as “This person is AI-generated” may be clear, but the statute’s requirement is that the disclosure be conspicuous. In practice, the disclosure should be easy to notice, read, and understand in the format where the advertisement appears.
The law includes several important exceptions:
- Audio-only advertisements are excluded.
- Advertisements are excluded when AI is used only to translate the language of a human performer.
- Certain promotions for films, television programs, streaming content, documentaries, video games, and similar expressive works are excluded when the synthetic performer is used consistently with the underlying work.
The statute also provides civil penalties for violations. More importantly for day-to-day operations, it creates a reason for businesses to understand how advertising assets are created before they are approved and distributed.
If a Marketing Agency Creates the Ad, Who May Be Responsible?
This is where the wording of the law matters. Section 396-b does not say that the client business is always responsible for an agency’s mistake. It also does not say that hiring an agency automatically shifts all responsibility to the agency.
The law focuses on a person or business that produces or creates the advertisement and has actual knowledge that a synthetic performer is present. Based on that language:
- A marketing agency that knowingly creates an advertisement with a synthetic performer may fall directly within the law.
- A client business may also fall within the law if it participated in producing or creating the advertisement and actually knew that the asset contained a synthetic performer.
- Depending on the facts, more than one party may have a role in producing or creating the same campaign. The statute does not provide a simple rule making only one party responsible.
What if the business truly did not know? The statute’s actual-knowledge requirement may be important. A business that did not know an agency-created asset used a synthetic performer may have a different position from a business that approved the AI use or received information showing how the asset was made. But lack of knowledge should not be treated as an automatic safe harbor. The result may depend on the business’s role in creating the advertisement, the communications between the parties, the contract, and other laws that may apply.
Practical takeaway: Do not leave AI use unaddressed in the agency relationship. Require the agency to identify AI-generated people before approval, document that disclosure, and confirm who will add and test any required label.
Why a New York Law Matters to California Businesses
A California business does not need a New York office for its advertising to reach New York. A social platform may optimize delivery across state lines. A streaming campaign may be national. A franchisor, retailer, or outside agency may reuse the same creative asset in several markets.
When New York first enacted this law, we explained why California businesses should pay attention to the broader regulatory trend. Read our earlier analysis of New York’s AI advertising disclosure law and what it may signal for California businesses.
The practical issue is not simply where the advertisement was created. Businesses should understand where the campaign will appear, who controls distribution, and whether later edits or platform changes could remove or obscure the disclosure.
This issue deserves closer review when a business:
- Runs national or New York-facing social, streaming, video, or display campaigns.
- Uses an agency or production company to create realistic AI people or avatars.
- Receives finished content without information about how the people shown were created.
- Allows employees or vendors to publish AI-assisted content without a defined approval process.
How Is California’s AI Transparency Law Different?
California takes a different approach. The California AI Transparency Act, which became operative on August 2, 2026, focuses on large providers of generative-AI systems, such as the companies behind widely used tools from OpenAI and Anthropic. Covered providers must offer ways to identify content created or altered by their systems, allow users to add a visible AI-generated label, and include digital information showing where the content came from.
New York’s law is different because it speaks directly to businesses that knowingly produce or create commercial advertisements using synthetic performers. California businesses may therefore need to consider both the tools an AI provider must make available and the separate disclosure decisions the advertiser must make.
Businesses should also remember that AI does not replace existing advertising rules. The Federal Trade Commission continues to require advertising to be truthful, nondeceptive, fair, and supported by evidence. An AI label does not correct an inaccurate product claim, misleading testimonial, or unsupported promise.
A Practical AI Advertising Review
The goal is not to make every campaign harder to approve. It is to create a repeatable process so marketing teams, agencies, and business owners know what information must be reviewed before an advertisement is published. Here are six practical steps businesses can take to strengthen oversight, support compliance, and better protect their interests before publishing AI-assisted advertising.
- Identify AI-generated people. List current and planned advertisements that use AI-generated or AI-modified spokespeople, avatars, testimonials, or humanlike demonstrations.
- Confirm how the asset was created. Ask employees, agencies, and vendors which tools were used and whether a real person’s image, voice, or performance contributed to the result.
- Map the campaign’s reach. Document the states, platforms, websites, streaming placements, and partner channels where the advertisement may appear.
- Plan the disclosure early. Decide who will draft, place, and test the disclosure. Check mobile presentation, contrast, screen time, cropping, and later edits.
- Review agency and vendor contracts. Address AI use, required notice, rights and permissions, compliance responsibilities, recordkeeping, and responsibility for corrections.
- Assign approval and keep records. Identify who approves AI-assisted advertising and preserve the asset history, vendor disclosures, geographic plan, and final approved version.
Questions to Ask a Marketing Agency or AI Vendor
Before approving an AI-assisted advertisement, businesses should ask their agency or vendor a few direct questions about how the content was created, where it will appear, and who is responsible for any required disclosure.
- Does this advertisement contain a realistic person who was generated or materially modified using AI?
- Is any real person identifiable, and what permissions or licenses support that use?
- Where will the advertisement be distributed, and can automated placement extend its reach?
- Who will decide whether a disclosure is required, and who will add and test it?
- Will the disclosure remain visible after resizing, cropping, translation, editing, or platform compression?
- What records will the agency provide about the asset’s creation, approval, and revision history?
How Koegle Law Group Can Help
AI advertising does not need to become another disconnected compliance project. With the right process, business owners can give their marketing teams room to use new tools while creating clear expectations around review, disclosure, documentation, and vendor responsibility.
Koegle Law Group partners with California businesses to turn changing legal requirements into practical systems. Through our Business Advice and Counseling services, we can help review agency and technology agreements, clarify AI-use and disclosure responsibilities, and develop an approval process that fits the way your business actually operates.
Our Employment Advice and Counseling team can also help businesses create workplace AI policies and train managers and employees on when marketing content requires additional review. When digital advertising raises privacy, tracking, or technology-vendor concerns, our Cyber and Privacy Liability Defense practice can help identify the broader issues that may need attention.
The goal is not simply to add a label to one advertisement. It is to help your team ask better questions, make informed decisions, and build a process that can adapt as the technology and the law continue to change.
Contact Koegle Law Group to schedule a consultation and discuss how AI-assisted advertising fits into your business’s compliance and vendor-management practices.
Frequently Asked Questions
Is the marketing agency or the client business responsible for an undisclosed AI performer?
Potentially either or both, depending on the facts. The law applies to a person or business that produces or creates the advertisement and has actual knowledge that a synthetic performer is present. An agency that knowingly creates the asset may be covered. A client business may also be covered if it participated in producing or creating the advertisement and knew about the synthetic performer.
What if the client business did not know the agency used AI?
Actual knowledge is part of the New York statute. A business that truly did not know may be in a different position, but the answer depends on its role in creating the advertisement, what the agency communicated, the contract, and any other applicable law. Businesses should require written disclosure of AI use rather than relying on assumptions.
Does New York require exact wording for the disclosure?
No. Section 396-b requires a conspicuous disclosure but does not prescribe one exact sentence. The wording and presentation should be evaluated for the specific advertisement and medium.
Does the law apply to audio advertisements or AI translation?
Audio-only advertisements are excluded. The law also excludes advertising when AI is used solely to translate the language of a human performer.
Is an AI disclosure enough to make the advertisement compliant?
Not necessarily. The advertisement must still comply with other applicable rules involving truthful claims, testimonials, intellectual property, privacy, contracts, and industry-specific requirements.
Government Sources and Further Reading
New York General Business Law Section 396-b — Current statutory text covering synthetic performers, actual knowledge, disclosure, exceptions, and penalties.
New York Senate 2025 legislative summary — Official summary (pg. 70) describing the law as applying when a person with actual knowledge uses a synthetic performer in commercial advertising.
New York Governor’s June 9, 2026 announcement — Official confirmation that the law is in effect.
California AI Transparency Act — California Business and Professions Code Sections 22757 through 22757.6.
Federal Trade Commission advertising guide — Federal guidance on truthful, nondeceptive, fair, and substantiated advertising.
Disclaimer: This article is intended for general informational purposes only and does not constitute legal advice. The information contained herein may not reflect the most current legal developments and is not guaranteed to be correct, complete, or up to date. Nothing in this article should be construed as creating an attorney-client relationship. Employers and HR professionals should consult with competent employment counsel regarding their specific facts and circumstances before taking any action.
About Koegle Law Group, APC
Koegle Law Group, APC is a boutique employment law and business litigation firm serving employers and business owners. Our attorneys advise clients on wage and hour compliance, PAGA defense, class action defense, employment counseling, and general business matters. For a confidential consultation, contact us at:
Koegle Law Group, APC – www.koeglelaw.com; info@koeglelaw.com
[Contact Us] | [Schedule a Strategy Call] | [Subscribe to Our Newsletter]
