Guides · Article 50 · EU AI Act
Virtual influencers and the EU AI Act
Virtual influencers — synthetic personas that post content and sometimes chat with followers — can touch more than one Article 50 paragraph at once, depending on what they do. This guide separates the interaction question (50(1)), the provider-side marking of generated media (50(2)), and the deployer-side deepfake question (50(4)), from the regulation text. It is informational only, not legal advice, and does not name any persona or brand as in or out of scope.
When the persona chats back — Article 50(1)
Where a virtual influencer is an AI system intended to interact directly with people — replying to comments or direct messages automatically — Article 50(1) can come into view. That paragraph is written on the provider of the system: it must be designed so people are informed they are interacting with an AI system, unless that is obvious to a reasonably well-informed, observant and circumspect person given the context. A persona presented as a real person raises the question more sharply than one openly labelled as virtual. Whether it clears the 'obvious' bar is fact-specific.
Generated images and video — Article 50(2)
The imagery and video that make up a virtual influencer's feed raise Article 50(2), the provider-side marking of synthetic image and video as detectable and artificially generated. As written, that marking duty sits on the provider of the generative system rather than on every account that reposts a frame. Whether a given pipeline falls inside 50(2), and whether an assistive-editing carve-out applies to any retouching step, are fact-specific questions.
Likenesses of real people — Article 50(4)
A wholly invented persona differs from content that depicts an identifiable real person. Where output constitutes a deep fake, Article 50(4) is written on the deployer of the system to disclose that the content has been artificially generated or manipulated, with the artistic-work modulation limiting that to disclosing the existence of generated content in a manner that does not hamper enjoyment where the content is part of an evidently creative work. Whether a given post is a 'deep fake' and whether a carve-out applies are fact-specific. Mapping which of the three paragraphs are in play, and whether you are provider-side or deployer-side for each, is the useful first step — confirm with qualified counsel.
Common questions
Does a virtual influencer have to disclose that it's AI?
It can touch several paragraphs: Article 50(1) where the persona interacts with people (provider-side), 50(2) for marking generated media (provider-side), and 50(4) where a post depicts a real person as a deep fake (deployer-side). Which apply, if any, are fact-specific — verify with qualified counsel.
See what may apply to your business
Answer seven quick questions for an automated, informational indication of which Article 50 obligations appear likely to apply — free, and not legal advice.