Article

Recently, in the case of Alakh Pandey v. John Doe & Ors. CS(COMM) 835/2026, the order granting ad interim injunction in the matter marks a major milestone in how Indian courts approach the intersection of AI-generated content, deepfakes, and personality rights. As generative AI tools make it frictionless to replicate voices, map faces, and create manipulated imagery or stickers, this judgment introduces vital structural boundaries for future litigation.

The suit was filed by prominent educator and Physics Wallah founder Alakh Pandey against various unidentified entities (“John Does”), online platforms, social media handles, coupon/sticker websites, and domain registrars. Represented by Senior Advocate J. Sai Deepak, the plaintiff argued that his name, image, likeness, voice, and unique persona elements were being misappropriated across the internet through unauthorized AI-generated content, stickers, GIFs, commercial coupon codes, and sexually explicit material. The plaintiff explicitly clarified that the action did not target the registered intellectual property or trademarks of Physics Wallah, but was strictly confined to the unauthorized commercial exploitation and degradation of his personal attributes.

Justice Anup Jairam Bhambhani raised a critical cautionary note, stating that claims seeking blanket protection over a public figure’s identity can be overbroad and susceptible tomisapplication. Citing the precedent in DM Entertainment Pvt. Ltd. v. Baby Gift House & Ors., the Court underscored that personality rights must not become a tool to suppress the dissemination of information concerning public wrongdoing and extinguish legitimate forms of creative and public discourse, such as caricature, lampooning, parody, and criticism, unless they cross into unauthorized commercial exploitation.

The Court balanced public accountability against personal dignity, noting that public figures cannot claim an absolute shield against critique, but are entitled to intervention against degrading, highly vulgar, or non-consensual commercial misuse.

Following a defendant-wise tabulated presentation of URLs submitted by the plaintiff, the Court restricted its ex-parte ad-interim injunction strictly to three defined categories where a clear prima facie case, balance of convenience, and risk of irreparable injury were established:

1. Sexually Vulgar Portrayals: Content depicting the plaintiff in a sexually explicit, vulgar, or deeply degrading manner (including AI-manipulated imagery/stickers).
2. Unauthorized Commercial Monetisation: Content exploiting his name, face, or attributes for commercial gain, selling products/coupons, or generating profit without a license.
3. Impersonation: Accounts or content fraudulently masquerading as the plaintiff to deceive the public into believing he endorsed or communicated the material.

The Court issued explicit directions to major social media platforms and intermediaries (such as Google, X, Telegram, and LinkedIn) regarding compliance and user privacy:

• URL Takedowns: Ordered the removal of specific offending URLs rather than taking down entire websites indiscriminately.
• Basic Subscriber Information (BSI): Directed intermediaries to disclose the identities, IP addresses, contact specifics, and payment/billing particulars of the operators behind the infringing accounts to allow proper service of legal notice.
• Statutory Neutrality under IT Act: Invoking Section 79 of the Information Technology Act, 2000, the Court stressed that Internet Service Providers (ISPs) and Domain Name Registrars (DNRs) must maintain strict statutory neutrality to retain “safe-harbour” protection. They cannot possess arbitrary discretion to censor or block websites merely at a litigant’s request.

To prevent defendants from evading the injunction by shifting domains, the Court established a tightly regulated verification protocol:

• Affidavit-Based Procedure: If the plaintiff discovers active mirror, alphanumeric, or redirect websites sharing identical infringing content, he may submit details via affidavit to the concerned ISP or DNR.
• Technical Verification vs. Adjudication: The Court clarified that ISPs/DNRs are not asked to legally judge copyright or personality infringement. Instead, their role is strictly limited to technically verifying whether the flagged domain is a true mirror or redirect of the already-enjoined defendant websites. If verified technically, the injunction automatically extends to block those mirror nodes.

In recent years, celebrity personality rights cases (such as those involving actors like Anil Kapoor or Amitabh Bachchan) often sought sweeping, blanket orders against any unauthorized use of their persona. The Delhi High Court explicitly cautioned that treating personality rights as absolute property can lead to misapplication.

Further, by refusing to issue a blanket ban on all memes or AI creations, the Court pragmatically carved out what generative AI misuse looks like. It targeted technology when it crosses into deepfake sexual vulgarity, fraudulent impersonation, and unauthorized commercial scaling (e.g., selling stickers, merch, or coupon codes using cloned likenesses).

Generative AI thrives on memes, parodies, and satirical commentary. The ruling ensures that creators, internet users, and critics cannot be silenced under the guise of personality rights.  The judgment establishes that unless a deepfake or AI creation amounts to direct commercialexploitation or severely degrades an individual (such as explicit non-consensual deepfakes), protected forms of expression like lampooning, caricature, and internet culture must remain untouched.

AI tools allow rogue actors to instantly spin up “mirror websites” or alternate social media handles hosting deepfakes once an original URL is taken down. The Court balanced this by forcing intermediaries (Google, X, Telegram, LinkedIn) to reveal Basic Subscriber Information (BSI). Unmasking the creators behind anonymous AI accounts is crucial because deepfakes are frequently deployed anonymously.

By allowing affidavit-based technical verification for mirror and redirect websites through ISPs and Domain Name Registrars, the Court provided a practical remedy to chase down fluid, AI-generated scam sites without requiring plaintiffs to file fresh lawsuits every time a deepfake domain shifts its URL.

Ultimately, this judgment shifts Indian jurisprudence away from treating a public figure’s persona as an impenetrable shield, steering it instead toward a balancedframework – one that gives victims of malicious deepfakes, vulgar AI manipulations, and commercial identity theft sharp legal teeth, while preserving the breathing room required for internet parody and free speech.