Openly AI influencer: why and how to disclose it

Should you say it is AI?

An AI influencer, or virtual influencer, is a character whose face, voice or videos are produced with artificial intelligence tools. The technology is not the problem. The problem starts when the audience believes it is following a real person.

This article brings together the rules of the platforms, of the European Union and of Switzerland, with their official sources. It is not legal advice: for your own situation, check with a lawyer or the competent authority.

What TikTok, YouTube and Meta ask for

We reread the official pages of the three platforms on 6 October 2026, in their French versions. They can change: open the link before you publish.

TikTok adds a limit that does not depend on the label. Even when labelled, AI content must not use the likeness of a minor, nor that of an adult who is not a public figure without their permission, nor falsely show a public figure in certain contexts, for example endorsing something. An AI persona must therefore have a face that belongs to no one.

What does the European Union say?

The European regulation on artificial intelligence (Regulation (EU) 2024/1689, the "AI Act") has an article devoted to transparency, Article 50. Among other things, it provides that:

• providers of tools that generate image, audio, video or text mark their outputs in a machine-readable format, so that they are detectable as AI-generated or manipulated;
• those who use an AI system to produce a deep fake disclose that the content has been artificially generated or manipulated;
• the information is given in a clear and distinguishable manner, at the latest at the time of the first interaction or exposure.

According to the European Commission, the regulation entered into force on 1 August 2024 and became applicable on 2 August 2026, with a few exceptions. The Commission states that it has been enforcing the new transparency requirements since that date. A simplification regulation (the "AI omnibus"), in force since 27 July 2026, changed part of the timeline: check the dates that concern you on the Commission's page. Sources: text of the regulation on EUR-Lex, Article 50 on the Commission's AI Act Service Desk and the Commission's page on the regulatory framework.

Switzerland is not a member of the European Union. If part of your audience is in the EU, check whether this regulation concerns you. Whether an entirely fictional character falls within the definition of a deep fake depends on the case: we do not settle that here. Our choice is simpler: we say it in every case.

Is an AI influencer legal in Switzerland?

This section covers Swiss law only. To our knowledge, no Swiss text prohibits creating a character with AI as such. We are not lawyers and the law evolves: have your project checked. The general rules apply, and two of them matter for a brand.

The Swiss Unfair Competition Act (UCA)

Its Article 2 sets the principle:

"Any conduct or business practice that is misleading or which otherwise violates the principle of good faith such that it influences the relationship between competitors or between suppliers and customers is unfair and unlawful."

Article 3, paragraph 1, letter b targets anyone who provides "incorrect or misleading information about themselves, their business, [...] their goods, works or services". Source: Unfair Competition Act on Fedlex (status as of 1 January 2025; the English text is a translation published by the Swiss Confederation for information, only the German, French and Italian versions are legally binding). Making people believe that a real person recommends a product when that person does not exist therefore raises a real question of fairness. Only a court can say whether a specific case is unfair.

The rules of the Swiss Fairness Commission

The Swiss Fairness Commission (Schweizerische Lauterkeitskommission) publishes rules on fairness in commercial communication. We read them in French and German; what follows is our own summary in English. Its rule B.15 says that commercial communication is unfair when it is not clearly recognisable as such. The same rule applies to accounts and profiles on social media: if the commercial nature is not obvious, the relationship with the third party must be disclosed, in particular where there is sponsorship, payment or a benefit in kind. Its paragraph 3 adds that it is unfair to use false identities, false accounts or false profiles. Source: Rules of the Commission, version of 1 January 2025 (PDF, in German).

In that document, we found no rule specific to artificial intelligence. To our knowledge, the Commission has not said whether a character that is AI-generated and declared as such falls under that paragraph 3: one more reason to declare it clearly. An AI persona that advertises must therefore flag two separate things: that it is advertising, and that the character is AI-generated.

How to say it, in practice

A useful disclosure is one people see without looking for it. Here is the order we recommend.

1. In the account bio. One simple line, for example "AI-generated persona". It is the first place a new visitor looks.
2. On every realistic piece of content. Turn on the platform's label (see the table). It stays attached to the video when it travels.
3. On the website. A readable sentence next to the character's introduction, and a reminder in the footer.
4. In advertising. Flag the partnership in addition to the AI disclosure. They are two different pieces of information.
5. In messages. If someone asks "are you a real person?", the answer is no, straight away.
6. Also say what is real. Who is behind it, which company, which products. An invented character does not allow invented results.

Three things to avoid: the disclosure written in tiny print at the bottom of a page, the vague word that says nothing ("digital", "augmented"), and a personal story told as a lived testimonial in order to sell.

How Katia Ryze applies it on this site

Katia Ryze is the face of the Ryze ecosystem. She is an AI-generated persona. Here is what the site displays today, nothing more and nothing less.

• On the home page, under Katia's introduction, a note says that Katia is an AI-generated persona and declared as such, and that the Ryze ecosystem, the software and the results are real and built by the Ryze team.
• On the Profile page, the same sentence appears in the section that introduces her.
• In the footer, on every page: "Katia Ryze is an AI-generated persona. Some visuals on this site are AI-generated."
• In this journal, every article repeats it at the top of the page.

The life story you read on the home page is the character's story. What is real can be checked elsewhere: the products of the ecosystem are online and presented in the portfolio.

What still needs improving: on the home page, the disclosure sits below the story, not on the first screen. We prefer to note it here rather than pretend everything is perfect.

Frequently asked questions

Read next: Creating an AI persona for a brand or a personal brand.