# Are AI influencers legal? US, EU, UK and India Are AI influencers legal? Yes, in the US, EU, UK and India. Each one has different rules for AI labels, advertising and platforms. See what applies to you. - Canonical: https://aiofmcity.com/blog/are-ai-influencers-legal - Topic: EU compliance (https://aiofmcity.com/topics/eu-compliance) - Author: Gus - Published: 2026-09-14 - Updated: 2026-10-27 --- ## Key points - Running an AI influencer is legal in the US, EU, UK and India. The rules are about labelling and about misleading people. None of them bans the persona itself. - Of the four, only the EU puts the labelling rule directly on you as the person publishing. It has applied since 2 August 2026. - In the US and UK, the rules that matter are advertising rules. They apply to virtual influencers exactly as they apply to people. - India works through the platforms. Since 20 February 2026 they have to label synthetic media, and the largest ones have to ask you to declare it. - Legal does not mean allowed. Each platform adds its own rules, and those rules decide whether your account survives. --- Yes. You can legally create and run an AI influencer in the United States, the European Union, the United Kingdom and India. None of the four bans a generated character with a public account. None of them asks for a licence to run one. All four have rules for how you present the content to the people who see it. Each one does it in a different way. A persona that follows the rules in one of them can still break the rules in another, so you need to know the differences. This page is a dated summary of published rules. It is not legal advice. ## AI influencer rules by country | Country or region | Legal to run? | What applies | Who is responsible | | --- | --- | --- | --- | | **European Union** | Yes | You have to label AI-generated content (AI Act Art. 50), since 2 August 2026 | You, as the person publishing | | **United States** | Yes | No federal labelling law. Advertising and endorsement rules apply when the persona promotes something | You, and the brand paying you | | **United Kingdom** | Yes | Advertising codes, applied the same way with or without AI | The advertiser, and you as the publisher of the ad | | **India** | Yes | Platforms have to label synthetic media and ask users to declare it, since 20 February 2026 | Mainly the platform. You declare | ## European Union: you have to label AI content The EU is the strictest of the four. It is the only one where you, the person publishing, have to label the content. That applies whether you sell anything or not. > Since 2 August 2026, deployers publishing AI-generated or manipulated image, audio or video content that resembles real persons, objects, places or events and would falsely appear authentic must disclose that it is artificially generated. Content generated and published before 2 August 2026 is not covered retroactively. > > — Source: EU AI Act Art. 50(4), read 2026-09-09 — https://eur-lex.europa.eu/eli/reg/2024/1689/oj > The Commission's guidelines state the duty applies without deceptive intent, and also where no real person is depicted. > > — Source: European Commission guidelines on Article 50 transparency obligations, read 2026-09-09 — https://digital-strategy.ec.europa.eu/en/policies/guidelines-ai-transparency-obligations The second source is the important one for a persona. You cannot argue "she is not a real person". The guidelines say clearly that the rule applies anyway. The rule also depends on where people can see the content. If you live outside the EU and have an audience inside it, you are not automatically outside the rule. [EU compliance for AI personas](https://aiofmcity.com/guides/eu-compliance-for-ai-personas) shows what the label has to look like and where you have to put it. ## United States: no AI labelling law, but FTC endorsement rules apply No federal law says you have to label AI-generated content. California's AI Transparency Act applies to the companies that make AI tools. It does not apply to the people who use them. [Do I have to disclose AI content in the US](https://aiofmcity.com/blog/do-i-have-to-disclose-ai-content-us) covers both in detail. The rules that do apply are older and broader than any AI law. > The Federal Trade Commission's updated Endorsement Guides changed the definition of "endorsements" to clarify the extent to which it includes fake reviews, virtual influencers, and tags in social media, and added specific guidance for influencers on when and how to disclose material connections. > > — Source: FTC — Updated Endorsement Guides (June 2023), read 2023-06-29 — https://www.ftc.gov/news-events/news/press-releases/2023/06/federal-trade-commission-announces-updated-advertising-guides-combat-deceptive-reviews-endorsements For you this means: as soon as your persona promotes a product that it was paid for or got for free, it counts as an endorser. The same rules apply as for a human creator. You have to tell people that you were paid or got the product for free. A generated persona is not outside the Guides. The FTC added virtual influencers to them on purpose. A paid post needs an ad label on top of the AI label: [AI influencer brand deals](https://aiofmcity.com/blog/ai-influencer-brand-deals-disclosure). ## United Kingdom: the advertising rules apply The UK has no special AI labelling law for this. The advertising codes apply, and the regulator has said how it treats AI. > The Advertising Standards Authority says that making clear an influencer is AI-generated "could well help to negate an otherwise misleading impression", but that "disclosure alone is very unlikely to mitigate the harm caused by a fundamentally misleading message", and that "the same rules apply in the same way, regardless of whether AI is used or not". > > — Source: ASA — Disclosure of AI in advertising, read 2025-05-29 — https://www.asa.org.uk/news/disclosure-of-ai-in-advertising-striking-the-balance-between-creativity-and-responsibility.html This means two things. An AI label helps when the only problem would be that someone takes the persona for a real person. It does not save an ad that misleads people about the product. One example is a skincare result that the product cannot produce. That ad stays misleading, however clearly you label the AI. ## India: the platform labels AI content and you declare it India does it a third way. Since February 2026, the rules apply mainly to the platforms. Your part as a user is to tell them the truth. > Amendments to India's IT Rules, notified on 10 February 2026 and effective from 20 February 2026, require intermediaries to label non-prohibited synthetically generated content clearly and prominently and not to allow those labels to be removed. Significant social media intermediaries must also require users to declare whether uploaded content is synthetically generated, and verify those declarations with technical measures rather than relying on the user's word alone. > > — Source: Freshfields — India targets deepfakes and AI-generated content (IT Rules 2026), read 2026-02-20 — https://www.freshfields.com/en/our-thinking/blogs/technology-quotient/india-targets-deepfakes-and-ai-generated-content-key-changes-under-meitys-2026-102mjwn For you that means two things. Answer the upload declaration honestly. And expect the platform to label the content whatever you answer. [Is an AI influencer legal in India](https://aiofmcity.com/blog/ai-influencer-legal-india) covers what counts as synthetic, the deadlines for taking content down and the draft rule that was dropped. ### What the sources do not say: Finding a country that bans AI influencers outright The first thing to check for an article with this title is whether any country bans the activity itself. This site searched for one and did not find it. Every result described a labelling rule, an advertising rule or a consumer-protection rule applied to AI content. None described a ban on running a generated persona. A search cannot reach everything, so this does not prove that no such law exists anywhere. For that reason the answer here covers only the four places named in the title. It does not say "legal everywhere". One secondary report described a South Korean labelling rule for AI-generated advertising. It could not be traced to a primary source, so it is not in this article. ## Where people get into trouble The persona itself is rarely the legal risk in this business. The risk comes from three things, and each of them is a problem in all four places. **A real person's likeness.** This means a generated face made to look like someone who exists, or a real face placed onto generated images. Both raise questions of consent and personality rights (a person's right to their own image) before any AI rule even applies. Platforms act on it too. > Fanvue requires disclosure of AI-generated media — in the profile bio, the content caption, or as a watermark — and its AI-content policy prohibits AI-generated media depicting any real person other than the account owner, including anyone who resembles or is intended to resemble someone under 18. A prompt or disclaimer cannot be used to establish age; visual appearance is what is judged. > > — Source: Fanvue Help — is AI content allowed on Fanvue, read 2026-09-08 — https://help.fanvue.com/en/articles/9538738-is-ai-content-allowed-on-fanvue **Anything that looks underage.** No country and no platform has a grey area here. What counts is how the person looks. The age you state does not count. **Undisclosed promotion.** If your persona takes money or free products to post about them, that is advertising. You have to label it the same way a human influencer does. ## Platform rules come on top of the law Every answer above is about the law. Whether your account survives depends on the platform it runs on, because each platform adds its own rules. Two examples: Fanvue officially allows AI creators, under conditions you can read. Instagram limits the reach of profiles that do not carry its AI label. The [label matrix](https://aiofmcity.com/ai-label-matrix) puts the platform rules side by side. The [platform rules](https://aiofmcity.com/platforms) section has a page for each platform. For most people, four habits cover nearly all of these rules at once: 1. Say that the persona is AI-generated in the bio and on the media. 2. Set each platform's own AI label. 3. Keep the provenance metadata that your generator puts into the files. 4. Label paid promotion the way any creator would. If your audience is in the EU, read [EU AI Act Article 50 for persona operators](https://aiofmcity.com/blog/eu-ai-act-article-50-persona-operators) next, because the EU has the strictest rules of the four. If your real question is whether the business itself is sound, read [is AI OFM legit](https://aiofmcity.com/blog/is-ai-ofm-legit). ## Frequently asked ### Is it illegal to make an AI influencer? No, not in the US, the EU, the UK or India. This site also searched for a country that bans the activity itself and did not find one. The law has rules for how you present the content: an AI label in the EU, advertising and endorsement rules in the US and UK, and platform labelling in India. ### Do I have to say an AI influencer is AI-generated? In the EU, yes. Since 2 August 2026, Article 50 of the AI Act says you have to label photorealistic AI-generated content that resembles real people. This also applies when no real person is shown. In other countries the rule comes from advertising law, when the persona promotes something. Each platform also has its own rules, in every country. ### Can an AI influencer do paid brand deals? Yes. The normal advertising rules apply. In 2023 the FTC changed its definition of endorsements to include virtual influencers. The UK advertising regulator says its rules apply the same way whether AI was used or not. You have to mark a paid post as paid, the same way a human creator does. ### Is this legal advice? No. It is a dated summary of published rules, and every claim names its source. If a decision depends on your own situation, ask a lawyer in the country concerned.