Impersonation, Deepfakes and Morphed Images: How to Get Them Taken Down in India

We get this call most weeks. A working professional discovers a fake LinkedIn or Instagram account in her name. A businessman sees a morphed photograph circulating on WhatsApp groups. A school-going daughter is the target of a deepfake clip generated from her holiday pictures.

The instinct is to fight back publicly. The right move is the opposite — quiet, fast, evidence-first action that gets the content removed and the impersonator identified.

What law actually applies

Impersonation online is prosecutable under Section 66C of the IT Act (identity theft) and Section 66D (cheating by personation using a computer resource), read with Section 318 BNS — formerly Section 420 IPC — for cheating, and Section 336/356 BNS for forgery and defamation respectively. Where the content is sexually explicit or morphed into intimate imagery, Sections 67, 67A and 67B of the IT Act add serious teeth, and Section 77 BNS (formerly 354C IPC) covers voyeurism.

The intermediary — Meta, X, YouTube, the dating app — has safe harbour under Section 79 of the IT Act only if it acts on a valid takedown notice expeditiously. The IT Rules 2021 give them 24 hours to remove non-consensual intimate imagery and 36 to 72 hours for most other categories.

Hour 0: preserve before anything else

Do not report the post first. Capture it first. Full-page screenshots with URL and timestamp visible. Profile URLs, account handles, follower counts, post IDs. Where possible, archive.org or archive.ph snapshots — they are independently retrievable later. If the content is on WhatsApp, do not delete the message; export the chat with media.

• Save the URL of the offending post and the offending profile separately.
• Note the timestamp of capture in IST.
• If video, download a copy and compute a SHA-256 hash before sharing — that hash anchors the chain of custody if you ever need it in court.
• Where the platform exposes a 'shareable' URL with an ID, the ID itself is more durable evidence than the rendered page.

Hour 1–4: the platform takedown

Every major platform has an expedited reporting flow for impersonation and non-consensual intimate imagery. They are not all equal. Meta's NCII flow is the fastest in our experience; X has improved; smaller platforms often need a formal legal notice quoting the IT Rules to move at all. We send a Section 79-grounded takedown the same day, copying the platform's India grievance officer.

Hour 4–24: the criminal track

File a complaint at cybercrime.gov.in (NCRP) and follow up with the local Cyber Crime Police Station — in Bangalore, that is the CID Cyber Crime cell at Carlton House and the city Cyber Crime PS at Madiwala. The FIR matters because it unlocks Section 91 BNSS notices to platforms, which is how we get the actual IP logs, signup details and payment trails behind the impersonator.

The civil track runs in parallel

John Doe injunctions — increasingly granted by Indian courts in deepfake matters — bind unknown defendants and intermediaries to pull down the content within hours. The Delhi High Court has been particularly progressive; Karnataka courts have been receptive in personality rights matters. We use the civil track when the platform is dragging feet or when the content is spreading faster than criminal process can move.

Common traps

• Reporting the post yourself before preserving evidence — many platforms remove the content before you have a record of it, leaving you unable to prosecute.
• Engaging the impersonator. Anything you say can and will be screenshotted out of context.
• Trusting cheap 'reputation cleanup' agencies. Most do nothing the platform's own flow does not, and some are themselves part of extortion networks.
• Waiting. Morphed imagery embeds in search results and reverse-image graphs within days — speed of removal correlates directly with the eventual size of your problem.
• Filing only the criminal complaint and assuming the police will get the content removed. They generally will not — that is a separate workstream you must drive.

What we do behind the scenes

On a typical impersonation matter we run four parallel tracks in the first day. We send platform-specific takedowns citing Rule 3(2)(b) of the IT Rules 2021 to the grievance officer, the resident grievance officer and the Grievance Appellate Committee where relevant. We file the NCRP complaint and personally walk the printout into the Cyber Crime cell to obtain an FIR. We prepare the Section 91 BNSS request that the investigating officer will issue to the platform — pre-drafted, with the exact post IDs, account handles and timestamps the platform will need. And where the matter is moving fast or the content is sexually charged, we move a writ petition or a civil suit for John Doe injunctive relief in parallel.

When deepfakes specifically are involved

Deepfake-specific matters carry an additional layer. The Ministry of Electronics and Information Technology has issued advisories under the IT Act treating synthetic media that misrepresents a person as a Rule 3(1)(b) violation, requiring intermediaries to remove it within 36 hours of being notified. We cite that advisory in every deepfake takedown we send. Where the deepfake is sexually explicit, Section 67A of the IT Act and Section 77 BNS apply with materially heavier punishment — up to seven years.

If you or a family member is dealing with impersonation, a fake profile, a morphed image or a deepfake right now, message us on +91 63634 69138. The first 24 hours decide how visible this becomes — and we move fast in matters where that window matters.