By Connie · Last reviewed: April 2026 — pricing & tools verified · This article contains affiliate links. We may earn a commission at no extra cost to you if you sign up through our links.
New York's AI Ad Disclosure Law Takes Effect June 2026: What Every Marketer Must Know
April 7, 2026 · 8 min read
New York requires AI disclosures in advertisements starting June 2026. Any ad creative that was materially generated or altered by AI — images, video, voiceovers, written copy — must carry a visible label. Penalties run up to $5,000 per violation. Every marketer running national digital campaigns needs a compliance workflow in place before June.
Reported on April 7, 2026, New York is the first US state to mandate AI disclosures specifically in advertising. The law targets consumer-facing AI-generated ad creative and takes effect in June 2026 — giving marketers roughly two months to update workflows, audit existing campaigns, and implement disclosure labels.
This guide covers exactly what the law requires, which ad formats are covered, what penalties look like, and how to build a disclosure workflow that keeps your campaigns compliant without slowing down production.
What the Law Actually Requires
The New York AI advertising disclosure rule falls under the state's existing consumer protection framework. It does not create a new regulatory body — enforcement runs through the Attorney General's office, which already handles deceptive advertising under General Business Law.
The core requirement is simple: if AI materially generated or materially altered any consumer-facing element of a paid advertisement, that ad must disclose it. The disclosure must be clear and conspicuous — meaning it cannot be buried in fine print or displayed for less than a minimum dwell time.
What Counts as "Material" AI Use
The law distinguishes between AI as a production tool and AI as the content source. Using AI to resize an image, auto-crop for a platform, or run grammar checks does not trigger disclosure. What does trigger it:
- AI-generated product images or lifestyle photos (FLUX, Midjourney, Firefly, DALL-E, Imagen)
- AI-generated or AI-cloned voiceovers (ElevenLabs, Murf, Suno voice)
- AI-generated video or AI-altered video of real people (Runway, Kling, Veo, Sora)
- AI-written ad copy that was not substantially rewritten by a human
- Synthetic faces, AI avatars, or AI-generated talent in ads
What Is Explicitly Excluded
- AI used for ad targeting, audience segmentation, or bidding algorithms
- AI analytics, attribution, or performance reporting
- Spell-check, grammar correction, or minor copy editing by AI
- Background removal from real photographs
Scope: Which Ads Are Covered
| Ad Type | Covered? | Notes |
|---|---|---|
| Digital display (banner, native) | Yes | If creative is AI-generated |
| Video ads (YouTube, CTV, social) | Yes | Includes AI voiceover over real footage |
| Social media sponsored posts | Yes | Instagram, TikTok, Facebook, X |
| Search text ads (Google, Bing) | Partial | Only if copy was AI-generated and unedited |
| Email marketing | Yes | If targeting NY consumers |
| Print advertising | Yes | If distributed in New York |
| TV commercials | Yes | If aired in New York market |
| Organic social posts (non-paid) | No | Paid promotion only |
| AI targeting / bidding tools | No | Back-end tools explicitly excluded |
Penalties: How Serious Is This
Non-compliance is treated as a deceptive trade practice under New York General Business Law §349 and §350. Civil penalties:
- Per-violation fine: $500–$5,000
- Each non-compliant ad impression can be treated as a separate violation — running a non-disclosed AI ad to 100,000 New York users is theoretically 100,000 violations
- Injunctive relief: AG can seek court orders to pull non-compliant campaigns immediately
- Private right of action: consumers can sue for actual damages or $50, whichever is greater, plus attorney fees
In practice, enforcement will likely start with major advertisers and egregious cases — deepfake-adjacent ads, AI-generated celebrity likenesses, or ads that clearly deceive consumers about product authenticity. But the per-impression structure means large-scale digital campaigns carry serious theoretical exposure.
Happycapy can help you draft disclosure templates, audit campaign assets, and maintain a compliance log — all from one AI workspace.
Try Happycapy FreeRequired Disclosure Format
The law specifies that disclosures must be "clear and conspicuous." Guidance expected before June will provide specific size and placement rules, but the baseline standard is already clear from FTC case law applied to the NY AG:
- Text size: at least as large as the smallest body text in the ad
- Contrast: must be legible against the background (no white text on light background)
- Placement: visible without scrolling or user interaction; not hidden in expandable fine print
- Video: must appear for at least 3 seconds in a readable size, or be present throughout the ad
- Audio ads: verbal disclosure required, read at a pace a reasonable person can understand
Approved disclosure language (pre-approved phrasing reduces enforcement risk):
- "Created with AI"
- "AI-generated content"
- "This [image/video/voiceover] was generated using artificial intelligence"
How to Build a Compliance Workflow Before June
The two-month runway is tight for large advertising operations but manageable if you start now. Here is a practical four-step workflow:
Step 1: Asset Audit
Inventory every active campaign asset. Tag each creative element: human-produced, AI-generated, or AI-altered. Build a simple spreadsheet: campaign name, platform, asset type, AI tool used (if any), disclosure status. This audit also serves as your compliance record if enforcement comes.
Step 2: Update Creative Templates
Add a disclosure layer to every AI-inclusive ad template in your design system. For display ads, this is a text layer. For video, it is a title card or persistent lower-third. Brief your design and production teams on the format requirements before the next production cycle.
Step 3: Update Your AI Tool Workflow
When using tools like Midjourney, Firefly, ElevenLabs, or Runway in ad production, flag the output for disclosure at the point of creation — not at the point of publishing. The earlier in the workflow you tag it, the less likely it slips through.
Step 4: Geo-Target Your Disclosure Logic (If Applicable)
If you run national campaigns and want to apply disclosures only to New York impressions, most major DSPs and social platforms support geographic creative variants. You can serve a disclosed version to NY and a non-disclosed version elsewhere — though legal counsel generally recommends applying disclosure nationally to reduce operational complexity and liability exposure.
The Broader Context: Other States and Federal Direction
New York is not alone. As of April 2026, 78 state-level AI-related bills are active across 27 states. Several include advertising-specific provisions similar to New York's:
- California: AB 2602 (AI in commercial ads, including synthetic performers) already in effect since January 2026
- Tennessee: ELVIS Act covers AI voice and likeness in commercial use — already in effect
- Texas: HB 4337 requires AI labeling in political advertising — in effect since September 2025
- Federal: The FTC has issued guidance (not binding rules) on AI disclosure in advertising; formal rulemaking expected by Q3 2026
The pattern across all of these is consistent: consumer-facing AI-generated content in commercial contexts requires disclosure. If you build a disclosure-ready workflow now, you are effectively compliant for the full landscape, not just New York.
What This Means for AI Marketing Tools
The law does not ban AI in advertising. It requires transparency. For marketers using AI tools like Happycapy, Jasper, Copy.ai, Adobe Firefly, or ElevenLabs in ad production, the workflow change is minor — add a disclosure label. The productivity gains from AI-assisted production are unaffected.
What will change: AI ad generation platforms will likely add built-in disclosure metadata and disclosure label generation to their export workflows. Expect platform updates from Canva, Adobe Express, and major DSPs before June.
Use Happycapy's AI workspace to draft compliance policies, generate disclosure language, track asset inventories, and brief your team — all in one place. Pro plan starts at $17/mo.
Start Free on HappycapyFrequently Asked Questions
When does New York's AI ad disclosure law take effect?
June 2026. Advertisers running campaigns that target New York consumers or that are created with AI-generated content must include clear AI disclosures starting that month.
What counts as AI-generated content under the New York law?
The law covers AI-generated images, video, voiceovers, and written copy used in paid advertising. It does not cover AI tools used only for targeting, bidding, or analytics. Any consumer-facing ad creative that was materially generated or altered by AI requires disclosure.
What are the penalties for non-compliance?
Civil penalties range from $500 to $5,000 per violation under New York General Business Law. Each non-compliant ad impression can count as a separate violation. The AG can also seek injunctive relief to pull non-compliant campaigns.
Does the law apply to ads only shown in New York?
The law applies to any advertisement that is reasonably likely to be seen by New York consumers, regardless of where the advertiser is based. National digital campaigns that include New York in their targeting audience are covered.
Sources
Get the best AI tools tips — weekly
Honest reviews, tutorials, and Happycapy tips. No spam.