AI avatar disclosure
Applies from 2026-08-02 · Last reviewed 2026-09-10 · Current
AI avatars — virtual presenters, synthetic models and generated spokespersons — need disclosure when they look realistic enough to be mistaken for a real person. They fall under the EU AI Act’s deepfake disclosure duty from 2 August 2026 and under TikTok, YouTube and Instagram AI labels. If the avatar resembles an identifiable person, add consent and human review. If it is evidently virtual, disclosure is still best practice for audience trust.
Who is affected
- Agencies scaling presenter-led content with avatars.
- E-commerce teams replacing photo models with virtual models.
- Brands testing synthetic spokespersons.
What triggers the rule
- Realistic avatar presenting or endorsing products.
- Virtual model indistinguishable from photography.
- Avatar resembling a real person.
What does not trigger it / exceptions
- Clearly stylised, animated or cartoon avatars.
Scenario table
| Scenario | Outcome | Why |
|---|---|---|
| Photoreal AI presenter for product videos | Disclosure required | Realistic synthetic person; platform labels plus specific wording. |
| Avatar resembling a famous person | Human review required | Likeness rights and deception risk; do not publish without legal review. |
| Stylised 3D mascot | No disclosure identified | Evidently fictional; record AI use in the manifest anyway. |
Required action
- Label avatar content on the platform.
- Disclose that the presenter is AI-generated.
- Record the avatar tool and approval trail.
Practical marketing examples
- “Presenter is an AI-generated avatar.”
- “Model is virtual. Clothing is real.”
Relation to platform policy
EU AI Act Article 50(4) requires deployers to disclose realistic AI-generated or manipulated image, audio and video resembling real persons, objects, places or events. Use the platform’s native label and keep your own evidence record.
Run the free check — result, wording and sources in under a minute.
Primary sources
Related pages
Methodology: Disclara answers are generated from a versioned rule set citing primary sources, reviewed by humans. Information on this page is not legal advice. See sources & methodology and rule changelog.