IT Rules 2021 and Deepfake/AI Content Regulation

A convincing fake video can now trigger the same legal machinery as a fake press release. On 10 February 2026, the government amended India’s core internet rules specifically to target deepfakes and AI-generated content. This article covers the IT Rules, 2021 framework, that 2026 amendment, and the reporting tools built around it.

mcqquestion.com

Indian Polity · Polity0161

IT Rules 2021 & Deepfake Regulation

India’s intermediary framework, and its 2026 update for AI content

2000

The Information Technology Act lays the legal foundation for regulating digital content and conduct.

2021

IT Rules (Intermediary Guidelines and Digital Media Ethics Code) impose due-diligence duties on platforms.

10 Feb 2026

IT Rules amended specifically for Synthetically Generated Information (SGI): deepfakes and AI-generated content.

5 Aug 2026

Government discloses cyber-safety stats in Parliament: ₹132.93 crore released, 33 States/UTs with cyber labs.

Safe harbour is the lever

Platforms follow the IT Rules mainly to keep their Section 79 safe-harbour protection. Losing it exposes them directly to prosecution for user content.

📑 Contents

1Must Know
2Good to Know
3Test Yourself
4Great to Know
5Current Affairs

Must Know

India’s core framework for online content sits in two laws. The Information Technology Act, 2000 (IT Act) provides the legal base. The IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 (IT Rules) set the operating duties.
The IT Rules place due-diligence obligations on intermediaries, including social media platforms. They must make reasonable efforts to stop users from hosting or sharing obscene, harmful, or illegal content.
Intermediaries that fail this duty lose their Section 79 safe-harbour exemption under the IT Act. Section 79 normally shields platforms from liability for content posted by users. Without it, they become directly liable for prosecution. A dedicated tool, the ‘Sahyog’ Portal, now speeds up Section 79(3)(b) takedown notices for child-safety-related content specifically. See Polity0149 for that fuller coverage.
On 10 February 2026, the government amended the IT Rules specifically for harms from Synthetically Generated Information (SGI), meaning deepfakes and AI-generated content.
Under the amendment, intermediaries and large platforms must deploy reasonable technical measures. These measures must prevent unlawful AI-generated content: obscene, misleading, impersonating individuals, or harmful to children.

Good to Know

Platforms must report related offences to authorities where the law requires it, under the Bharatiya Nagarik Suraksha Sanhita, 2023 (BNSS) or the Protection of Children from Sexual Offences Act, 2012 (POCSO).
The National Cyber Crime Reporting Portal (NCRP), at cybercrime.gov.in, lets citizens report cyber crimes. Its “Report Women/Child Related Crime” module allows complaints without disclosing personal identity.
NCRP added a “Report and Check Suspect” module. It lets citizens flag suspicious website URLs, WhatsApp/Telegram handles, phone numbers, emails, SMS headers, deepfakes, and social media URLs.
Under the Cyber Crime Prevention against Women and Children (CCPWC) Scheme, the government released ₹132.93 crore to States and UTs. This funded cyber forensic-cum-training labs, junior cyber consultants, and training for law-enforcement, prosecutors, and judicial officers.
Cyber forensic-cum-training labs are now commissioned in 33 States/UTs. More than 24,600 law-enforcement personnel, judicial officers, and prosecutors have been trained.
Alongside Part-II’s due-diligence duties on intermediaries, Part-III of the IT Rules, 2021 covers publishers of online curated content (OTT platforms). It requires them not to transmit any content prohibited by law. (Source: PIB)

Test Yourself

Great to Know

The IT Rules framework applies to all internet users, not just children. It sits alongside a separate, narrower children-specific layer built from the DPDP Act’s child provisions and the PRAGYATA Guidelines. See Polity0149 — Children’s Online Safety in India for that layer in full.
The real enforcement lever here is economic, not direct censorship. Platforms police content mainly to protect their own Section 79 safe harbour, since losing it exposes them to liability for every piece of user content, not just the offending post.
This IT Rules track handles platform-side obligations. A separate institutional track, built around I4C and the NCRP, handles citizen-facing crime reporting and enforcement. See SciTech0030 — Cyber Crime and Cyber Security for that institutional history.
India’s AI-content rules are one strand of a wider governance push. The judiciary has its own separate AI framework, covering AI use inside courts rather than on public platforms (see Polity0164), while the IndiaAI Mission funds AI development itself, including its own safety safeguards (see SciTech0033).

Current Affairs

On 5 August 2026, MoS for Information & Broadcasting Dr. L. Murugan gave a written Lok Sabha reply detailing the government’s cyber-safety measures. He confirmed the ₹132.93 crore CCPWC disbursement, 33 States/UTs with cyber labs, over 24,600 personnel trained, and more than 2 lakh NCC/NSS/NYKS students given Cyber Hygiene training by I4C. (Source: PIB)
On 10 February 2026, the government amended the IT Rules, 2021 to address Synthetically Generated Information (SGI). This closed a gap where deepfakes and AI-generated content fell outside clear intermediary obligations.
I4C has released a “Handbook for Adolescents/Students” on preventing cybercrime, and a “Cyber Security Handbook for Mahila Suraksha” to raise awareness among women. Both sit under the government’s wider Information Security Education and Awareness (ISEA) project.
On 30 July 2026, MoS Jitin Prasada gave a Lok Sabha reply on the amended IT Rules. He detailed sharply cut takedown timelines. Content removal on valid intimation dropped from 36 hours to 3 hours. Grievance redressal dropped from 72 hours to 36 hours, and to 2 hours for sensitive matters like nudity or impersonation. He cited 13 Responsible AI projects approved for deepfake detection. These include IIT Jodhpur and IIT Madras’s Saakshya framework, and IIT Kharagpur’s real-time voice deepfake detector. He confirmed the ISEA project has run 6,650 cyber-awareness workshops, reaching over 11.37 lakh people. Users can appeal unresolved complaints to Grievance Appellate Committees at gac.gov.in. (Source: PIB)
On 5 August 2026, MoS L. Murugan gave another Lok Sabha reply, this time on film-piracy enforcement. Intermediaries were directed to disable 7,393 URLs for film piracy, including 4,996 Telegram channels and 1,263 websites. This used the Cinematograph (Amendment) Act, 2023’s Section 7(1B)(ii), alongside Section 79(3) of the IT Act. The Ministry authorised its Joint Secretary (Films) to issue these takedown notices directly. (Source: PIB)