Bombay High Court moves against AI chatbots impersonating an actor
An unauthorised AI chatbot that talks as a real person is now a court matter in India. On 5 November 2026 the Bombay High Court granted actor Ananya Panday interim protection against the misuse of her name, image and identity, including deepfakes. Reports say the order also targets websites hosting fake AI chatbots that impersonate her.
What the court ordered
Justice Madhav Jamdar passed the order on Thursday, 5 November 2026. It is an ad-interim order, granted ex parte, which means the other side has not yet been heard and the case continues. According to the reports, the court found a prima facie case that the material violated her right to privacy, her right to publicity, her moral rights and her right to live with dignity.
Based on reporting by Indian Television, Bollywood Hungama and MediaNama, the order and the plea cover:
- Unauthorised use of her name, image, voice and signature online, and any false impression that she endorses a product.
- Deepfakes and AI-generated images and video clips, including sexually explicit content.
- Fake AI chatbots that impersonate her. Domain registrars and government authorities were asked to block websites hosting them.
- Product listings on marketplaces such as Amazon and Flipkart, and flagged content linked to Meta, X and Reddit, which intermediaries were directed to remove.
- Physical merchandise using her identity without permission, which the court ordered destroyed.
Part of a pattern
Panday's case is one of a growing series in India. A day earlier, the Delhi High Court gave actor Kriti Sanon interim protection over her name, image, likeness and voice. Her plea covered deepfakes, impersonating accounts, AI chatbots and synthetic voice models. That order left room for news, reviews, criticism, satire, parody and fan expression. Earlier, actors such as Anil Kapoor and Amitabh Bachchan won similar protection in Delhi.
Indian law has no single statute for this. Courts protect "personality rights" case by case, and each order names a specific person. A chatbot that copies someone is treated like any other misuse of their identity: it can create a false impression of endorsement, and it can say things the real person never said.
The same idea, in other places
- In the EU, AI-generated deepfakes and AI chatbots must be disclosed to people since 2 August 2026. What the AI Act requires.
- Denmark is debating a bill that would give everyone rights over realistic copies of their face and voice. Read our report.
- In the US, the NO FAKES Act on voice and likeness is still a bill. Where it stands.
This article is general information, not legal advice. The order is interim and may change once the other parties are heard.
What it means for you
You do not need to be famous for this to matter. Anyone whose words, photos and voice are online can be copied into a chatbot. A few practical steps:
- Search for your name together with "AI", "chatbot" and "voice" from time to time, and on character chat sites.
- Keep screenshots and links of any copy you find. Courts and platforms ask for the exact URLs.
- Use the reporting tools of the platform first. Most have a route for impersonation.
- If you build a twin yourself, keep control of where it appears and who it speaks for.
A twin you own
The difference between an impersonating bot and a digital twin is consent and control. On TalkMie the owner trains the twin on their own material, chooses whether it is private or public, chooses on which websites the chat window may appear and can download or delete anything they create. Talk to the founder's own twin to see how that looks, or create yours.
Questions
Is it illegal to make an AI chatbot of a celebrity in India?
India has no single law on this. Indian courts have repeatedly ordered such chatbots and deepfakes taken down when they use a person's identity without consent. The Ananya Panday order of 5 November 2026 is the latest example.
Is the Ananya Panday order final?
No. It is an ad-interim order granted ex parte. The other parties can still be heard and the court can change it.
How is a TalkMie twin different from an impersonating chatbot?
A TalkMie twin is created and trained by the person it represents. The owner decides whether it is public, where it can be embedded, and can delete it.
