New York synthetic performer ad law and AI avatar ads
NY S8420A requires a conspicuous disclosure of synthetic performers in ads. What it covers, what it exempts, and how to burn a line onto an avatar ad.
New York bill S8420A requires a person who produces an advertisement for a commercial purpose, with actual knowledge that a synthetic performer appears in it, to disclose that conspicuously in the ad. Penalties are $1,000 for a first violation and $5,000 for each later one. Whether an AI avatar presenter counts depends on the definition below, but if it does, the ad needs a visible line.
This summarizes the bill text as read 2026-09-30 and is not legal advice.
What does the bill define as a synthetic performer?
A digitally created asset created, reproduced or modified by computer using generative AI or a software algorithm, designed to appear as a human performance when the performer is not recognizable as a real person. The disclosure must appear in the advertisement itself, across all media.
| Item | Text of the bill |
|---|---|
| Who must disclose | A business that produces or creates an ad for a commercial purpose, with actual knowledge |
| Penalty | $1,000 first violation; $5,000 each subsequent violation |
| Exempt | Audio ads; language translation uses; promotional materials for expressive works such as films, TV and video games where use matches the work |
| Effective | The 180th day after it became law; the snapshot's summary says it was signed Dec 11, 2025 |
Is an avatar video a synthetic performer?
Sume Avatar Videos turn a ready avatar into a "script-driven talking video", and an avatar can be created from a prompt, a profile or a reference image. A presenter made that way is generated, so check it against the definition above. If the avatar is a recognizable real person, the bill's wording is about performers who are not, so your question is different; take advice on that case.
How do I add a disclosure line to the ad?
Two routes. For an avatar video, the optional captions object burns styles into the clean final MP4 after generation, using the spoken script, which is a transcript of speech and not a disclosure. For a fixed line, send the finished MP4 to POST /v1/video-captions with cues: text plus start and end, which "burns exactly that copy at those times" with no speech-to-text. Set the cue to span the whole ad, and place it so it reads as conspicuous. See Video captions.
curl -X POST https://api.sume.com/v1/video-captions \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: synthetic-performer-001" \
-d '{
"video_url": "https://example.com/avatar-ad.mp4",
"style": "black-outline",
"cues": [
{ "text": "Synthetic performer in this ad", "start": 0, "end": 15 }
]
}'What about translated or audio-only ads?
The bill exempts audio ads and language translation uses, so a voice-only spot is outside the on-screen question. For comparison with another regime, see EU AI Act Article 50 labeling.
Sources
Related posts
More in Use cases
- Turn product photos into a silent video with Timeline static holds
Timeline 1.0 treats a still as a static hold, so product photos become a silent video with fades. Shopify lists MP4 up to 10 minutes; here is the request.
- Pull product stills from a video with the video-frames API
Video frames returns jpeg or png stills from a Sume-hosted clip at set seconds. Google lists 500 x 500 minimum, so extract from the clip before captions.
- Seedance 2.5 draft mode: a 480p-then-1080p loop on Sume
Higgsfield added Seedance 2.5 Draft Mode. On Sume there is no draft model: render at resolution 480p, pick a take, then re-request at 1080p.
- Seedream 5.0 pro bbox and point tags vs region edits on Sume
Seedream 5.0 pro edits a region from <bbox> and <point> tags on a 0-999 grid. Sume lists other Seedream ids; use mask_url on ChatGPT Image 2.5 for regions.
Written by Sume