Discuss the adequacy of India's IT Rules, 2021 (as amended in 2026) in regulating AI-generated deepfakes. Suggest measures to balance innovation with accountability.

Q. Discuss the adequacy of India's IT Rules, 2021 (as amended in 2026) in regulating AI-generated deepfakes. Suggest measures to balance innovation with accountability. (15 marks, 250-350 words)

The IT (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026, in force from 20 February 2026, coin the term "Synthetically Generated Information" (SGI) to bring deepfake audio, video and text within binding law [2]. The framework marks a decisive shift from advisories to enforceable duties, yet remains partially adequate.

Strengths of the amended framework - Compressed timelines: removal of unlawful content on valid notice cut from 36 hours to 3 hours; general grievance redressal from 72 to 36 hours, and nudity/impersonation cases to 2 hours [1]. - Platform accountability: non-compliance costs the intermediary its Section 79 safe harbour, exposing it to prosecution under Sections 66C, 66D, 67/67A of the IT Act, 2000 [1]. - Transparency over prohibition: mandatory labelling and permanent traceable metadata with a unique identifier, which intermediaries cannot allow to be stripped — regulating use, not innovation itself [2]. - Electoral integrity: AI-altered campaign content must carry "AI-Generated"/"Synthetic Content" tags with originator disclosure [3].

Limits on adequacy - Rests on delegated legislation under a 2000-era statute never drafted for generative AI; proportionality of 3-hour windows versus natural justice for creators remains contested [1]. - Compliance capacity: 3-hour and 2-hour windows strain smaller intermediaries, risking over-removal of lawful speech [1]. - Detection is reactive — enforcement follows notice, while virality precedes takedown; cross-border and open-source models fall outside effective reach [2].

Way forward - Enact the proposed Digital India Act with a statutory, technology-neutral AI chapter. - Adopt graded obligations by platform size, with tiered timelines for small intermediaries. - Mandate interoperable provenance/watermarking standards and independent audits of detection systems. - Build a fast, appellate grievance route plus digital literacy campaigns so users can spot labelled synthetic media.

The amended Rules give India a workable first-generation deepfake regime — accountability without banning AI. Anchoring it in a dedicated statute, proportionate compliance and provenance technology would secure both Article 21 privacy and India's innovation ambitions.

(~320 words)

Sources: 1. Government Strengthens Regulatory Framework to Address AI-Generated Deepfakes — Press Information Bureau, MeitY (6 August 2026) — 36→3 hour takedown, 72→36 and 24→2 hour grievance timelines, Section 79 safe-harbour withdrawal, IT Act offence provisions 2. Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026 — MeitY — SGI definition, labelling and permanent metadata/unique identifier obligations, effective 20 February 2026 3. FAQs / Explanatory Note on the IT Amendment Rules on Synthetically Generated Information — MeitY (22 October 2025) — labelling of AI-generated political campaign content and originator disclosure