AI disclosure scenario
Do you need to disclose a synthetic spokesperson?
Direct answer
Usually yes when the spokesperson looks or sounds like a real person. Disclose the generated presenter and synthetic voice separately, use the destination platform's AI label, and review consent whenever the avatar or voice resembles an identifiable person. A label does not resolve likeness, testimonial or advertising-claim risks.
Applies when
- Article 50(4) requires disclosure when the avatar content constitutes a deep fake: it resembles an existing person or other real entity and would falsely appear authentic. A clearly fictional avatar may fall outside that definition, but the platform rule can still be broader.
- Disclosure does not resolve consent, likeness, testimonial or advertising-claim risk. A synthetic spokesperson must not be presented as a real customer, employee or expert unless that representation is independently truthful and authorised.
Doesn't apply when
- A clearly illustrated mascot that cannot reasonably be mistaken for a real person, subject to any broader client policy.
- AI used only to draft a script that is performed and recorded by a real, identified spokesperson.
- Minor retouching that does not alter what a real spokesperson said or did.
Worked decision
Example
A photorealistic AI presenter delivers a product recommendation in a TikTok ad using a synthetic voice.
Decision: Disclosure required
The presenter and voice appear human and are central to the endorsement. Apply TikTok's AI-generated label and avoid implying that a real person gave the recommendation.
Presenter and voice are AI-generated.
Next steps
Required action
- Record the generated or altered component, source asset, tool and destination before publishing.
- Apply the native platform label wherever the applicable platform workflow requires it.
- Use component-specific wording such as: “Presenter and voice are AI-generated.”
Publishing layer
Platform differences
TikTok
Label realistic generated image, audio and video. TikTok can prohibit some impersonation or false endorsement scenarios even if a disclosure is present.
Open platform guide →YouTube
Use the AI-use setting when a realistic generated presenter appears to say or do something that did not occur, or when the synthetic scene otherwise seems real.
Open platform guide →Instagram and Facebook
Meta can add AI info from technical indicators or publisher disclosure. Keep visible wording when the synthetic identity is material to the ad.
Open platform guide →Evidence
Source
Freshness and accountability
Last verified
Last reviewed 2026-09-12 · Current
Run this exact scenario
The content type, AI operation, context and starting platform for this scenario are preselected. Confirm the remaining facts.
Copy-ready language
Disclosure wording options
Presenter and voice are AI-generated.
AI-generated presenter. No real customer testimonial is depicted.
Related guides
“Doesn't apply” means no disclosure requirement is identified for the facts stated here. A client policy, advertising rule or another destination platform can still be stricter. This page is source-backed information, not legal advice.