Deepfakes, Morphed Images and the Law in India

1. Certain morphed and impersonation content has a specific platform removal route

The Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules contain a specific grievance mechanism for certain highly personal forms of harmful content. Where a complaint concerns material that exposes an individual's private area, depicts nudity or sexual conduct, or is in the nature of electronic impersonation including artificially morphed images, Rule 3(2)(b) requires the intermediary to take all reasonable and practicable measures to remove or disable access to the content within twenty four hours of receiving the complaint from the individual or a person acting on their behalf.

The availability of this mechanism depends on the nature of the content and the intermediary involved, but for victims of morphed or impersonation material it can be an important immediate step alongside preservation of evidence.

2. The 2026 synthetic media framework

The Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 were amended in February 2026 to introduce an express framework for synthetically generated information. The amendments came into force on 20 February 2026 and introduced obligations concerning unlawful synthetically generated information, labelling and technical provenance, together with additional declaration and verification duties for significant social media intermediaries.

These intermediary obligations are important, but labelling does not make otherwise unlawful content lawful. A labelled synthetic image used for impersonation, fraud, privacy invasion or without consent sexual content may still engage substantive legal provisions.

3. Identity theft and personation

Where manipulated media is used to assume another person's identity, obtain credentials, solicit money or induce someone to act on a false belief, Sections 66C and 66D of the Information Technology Act may become relevant in appropriate cases involving identity theft or cheating by personation through computer resources.

Section 319 of the Bharatiya Nyaya Sanhita, 2023 addresses cheating by personation. Its application still depends on the statutory elements of cheating and personation being established; the mere existence of a deepfake does not automatically establish that offence.

4. Privacy and without consent intimate imagery

Section 66E of the Information Technology Act may be relevant where its statutory elements concerning an image of a person's private area, absence of consent and violation of privacy are satisfied. Sections 67, 67A and 67B separately address electronic publication or transmission of specified categories of obscene, sexually explicit and child sexual material.

Which provision applies to morphed or synthetic intimate imagery therefore depends on the content, the source material and the manner in which it was created, published or transmitted. The creation, publication, transmission, threat to publish and use of material to coerce or harass can each raise different legal issues.

5. Fraudulent voice and video impersonation

Deepfake voice calls and synthetic video can be used to imitate directors, employees, family members or public facing professionals. In a financial fraud scenario, the manipulated media may be only one component of a larger deception involving fraudulent instructions, compromised accounts, false electronic records or unauthorised transfers.

The immediate response should focus on both evidence and containment. Preserving the media is important, but so is preserving the surrounding transaction record. This may include phone numbers, email headers, payment instructions, account details, chat logs and access records.

6. Preserve evidence before reporting

When harmful synthetic or morphed content appears on a platform, prompt reporting may be important, but evidence should be secured before the content disappears wherever it is safe and lawful to do so.

• Record the full URL or account handle, not only a screenshot of the image.
• Capture the date and time at which the content was accessible.
• Preserve screenshots of the post, profile, captions, comments and messages connected with it.
• Retain the original file received, where available, instead of repeatedly resaving or editing it.
• Keep copies of platform complaints, acknowledgement emails and complaint reference numbers.

7. Platform, police and cybercrime escalation

Platform reporting, police or cybercrime reporting and other legal remedies serve different purposes. The appropriate route depends on the content, whether the material is still online, whether there is an immediate safety or fraud concern, and whether the available evidence identifies an account, device, payment trail or other person involved.

Victims should avoid destroying or materially altering the original evidence while attempting to disprove the content. A working copy can be used for reporting while originals and device level records are preserved separately where possible.

8. Not every synthetic image is unlawful

Synthetic media has legitimate uses in entertainment, accessibility, education, design and other fields. The legal risk depends on the circumstances in which the material is created and used. Relevant factors can include deception, consent, privacy, sexualisation, fraud, reputational harm, unlawful content and compliance with intermediary obligations.

'AI generated' is therefore a technical description, not a complete legal conclusion. The same technology can be used lawfully in one context and form part of unlawful conduct in another.

Preserve first. Remove and escalate second.

Deepfake and morphed image disputes should be approached as evidence problems as much as content problems. Identify what was created, who was impersonated, where it was published or transmitted, what was represented as true, what harm followed and which digital records can still be preserved. Those facts determine which legal and platform remedies are actually relevant.