What an AI avatar may not claim on TikTok Shop
TikTok Shop prohibits AI that impersonates real people or invents doctors and experts to endorse products. An avatar can present, but it can't be a fake expert.
An AI spokesperson on TikTok Shop must not impersonate a real person and must not be given a fabricated professional identity, such as a doctor or other expert, to endorse a product. A synthetic face or voice also triggers disclosure. A plainly labeled avatar that presents accurate product information is a different thing from a fake expert.
The rules that bite avatars
| Rule | What the page says |
|---|---|
| Impersonation | Prohibited: impersonating real people, or fabricating professional identities (doctors, experts) to endorse products |
| False impressions | Prohibited: AI that misleads or creates false impressions about products, people or facts |
| Unrealistic results | Prohibited: for example instant hair growth or cavity removal |
| Disclosure | Required for synthetic faces, voices and digital humans |
| Method | On-screen text, watermark or note, or the AI-generated content toggle |
What that means for a script
Sume's avatar video API turns a ready avatar and a script into a talking video. A script, or a multi-scene video_inputs plan, has to land at 4 to 60 seconds. What the avatar says is your text, so the policy is a scripting problem more than a rendering one.
- Introduce the presenter as what it is. Do not write "As a dermatologist..." for an avatar that is not one.
- Do not borrow a real person's name, title or likeness for the avatar.
- Keep claims to what the listing supports; no medical effect, no instant result.
- Add the disclosure in the file or at posting time.
Script review before rendering
Review the script before you render, because an avatar render cannot take a claim back. Search the text for titles (Dr., nurse, pharmacist, licensed), for "clinically" or "proven" language, and for any named person. Each hit needs a source you can show, or it comes out.
The same review catches results claims. The policy lists unrealistic results such as instant hair growth and cavity removal as prohibited. A script that says what the product is, what it costs and how to use it is the safe core.
Practical limits
The avatar API accepts aspect_ratio 1:1, 3:4, 9:16, 4:3 or 16:9 (default 9:16), a resolution of 720p, and quality of standard, plus (the default) or max. Optional inline captions can burn the spoken script into the final MP4, which is also a place to carry a short disclosure line. If the caption stage fails, the avatar job can still succeed with a clean video and captions.status=failed, so check that field before you publish.
Enforcement on the policy page runs from removal or restricted visibility and warnings to permanent bans and a Creator Health Rating deduction.
Sources
Related posts
More in Sume Avatar 1.0
- UGC-style ad batch: ten 12-second hooks on Avatar 1.0 for $29.40
Ten 12-second UGC-style hook variants cost about $29.40 on Sume Avatar 1.0 plus, $22.08 on standard. The request, captions and checks.
- UGC-style avatar ad: hook, silent demo beat and CTA in video_inputs
Build a three-scene UGC avatar ad with POST /v1/avatar-1.0/talking-video: a spoken hook, a silent demo beat, a spoken CTA, inside the 4 to 60 second window.
- What Sume Avatar 1.0 Does and Does Not Do vs Live Avatars
A plain list of what Sume Avatar 1.0 renders (scripted 4-60 s clips) and what it does not do (real-time conversation), set beside Tavus Griffin's live model.
- Can I use Tavus Griffin-Lite yet? Preview status and what to ship
Tavus Griffin-Lite is a research preview for select trusted testers, not open to customers. Here is what the page says, and what you can build on Sume today.
Written by Sume