Someone used AI to make a deepfake video of me. What are the 2026 laws against this?

Yes — malicious deepfakes are unlawful on multiple counts. A 'deepfake' is AI-generated synthetic media (image, video, audio) that depicts a real person doing or saying something they did not. Where the deepfake is used to impersonate, defame, sexualise, or extort a real person, it triggers liability under:

(1) Bharatiya Nyaya Sanhita (BNS), 2023 — multiple sections (see next);
(2) Information Technology Act, 2000 — Section 66D (cheating by personation using computer resource), 66E (privacy violation), 67/67A (obscene content);
(3) IT Rules, 2021 — platform takedown obligations;
(4) MeitY's 2023-24 advisories impose stricter labelling and removal obligations on AI/synthetic media platforms.

Consensual deepfakes (e.g., for parody, satire, art with clear labelling) generally remain lawful as protected speech under Article 19(1)(a), subject to defamation, IP and obscenity carve-outs.

Depending on the nature and intent of the deepfake, multiple provisions apply (often in combination):

(1) Section 319 BNS — cheating by personation. Up to 3 years on conviction;
(2) Section 336(4) BNS — forgery for the purpose of harming reputation. Up to 7 years;
(3) Section 356 BNS — defamation. Up to 2 years;
(4) Section 79 BNS — insulting a woman's modesty. Up to 3 years;
(5) Section 77 BNS — voyeurism, where the deepfake is sexualised. 3-7 years;
(6) Section 351 BNS — criminal intimidation, where the deepfake is used as a threat;
(7) Section 66D IT Act — cheating by personation using a computer resource. Up to 3 years;
(8) Section 66E IT Act — privacy violation by capture/transmission of images of a private area;
(9) Sections 67 and 67A IT Act — publishing or transmitting obscene or sexually explicit material. 3-7 years.

Strict timelines apply under the IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021:

(1) Impersonation content (including deepfakes) reported by a user — 24 hours for removal;
(2) Non-consensual intimate deepfakes — 24 hours for removal;
(3) Generally unlawful content per court/government order — 36 hours.

Further, MeitY's November 2023 and March 2024 advisories have warned all major platforms (Meta, X, Google, YouTube, Telegram, OpenAI) that hosting or generating deepfakes without due-diligence labelling and removal will cost them their 'safe harbour' under Section 79 of the IT Act.

If the platform misses these deadlines, escalate to the Grievance Appellate Committee at gac.gov.in within 30 days, or to MeitY directly.

Step 1 — Capture evidence first. Take screenshots, save the URL, note the account/profile that posted it, and download the file if technically possible. Deepfakes are deleted quickly once reported.

Step 2 — Use the platform's deepfake / impersonation reporting flow. Most major platforms have dedicated category options. Note the report reference number.

Step 3 — Email the Grievance Officer with the URL, screenshots, your ID proof showing the resemblance is to you, and a formal takedown demand citing Rule 3(1) of the IT Rules, 2021.

Step 4 — File a complaint on the National Cyber Crime Reporting Portal under 'Synthetic Media / Deepfake', or call 1930.

Step 5 — Lodge an FIR at the local cybercrime cell. Use the Zero FIR procedure if necessary.

Step 6 — Engage a reputable, specialised cyber-law lawyer if the deepfake is widely shared or being re-uploaded — see the next section.

Yes — and this is the most effective remedy when content keeps reappearing. The Delhi and Bombay High Courts have repeatedly granted John Doe ('Ashok Kumar') dynamic injunctions in deepfake and image-abuse cases. These orders:

(1) Bind all platforms (named and unnamed) to remove the specified content;
(2) Continue to apply as new copies are identified — your lawyer can simply forward new URLs to the platforms invoking the same order;
(3) Direct platforms to preserve account data of the uploader for potential criminal investigation;
(4) Can require platforms to identify the original uploader.

The procedure: a reputable, specialised cyber-law lawyer files a civil suit in the High Court with the evidence, the names of major intermediaries as defendants, and an ex parte interim injunction application. Most High Courts grant interim orders within 1-3 days for non-consensual deepfakes.

For broader content takedown advice, see our picture takedown guide.