From 20 February 2026, social media and messaging platforms operating in India including Meta’s Facebook and Instagram, X (formerly Twitter), YouTube, WhatsApp and Telegram will face significantly stricter obligations to counter harmful content such as AI‑generated deepfakes, impersonation posts, misinformation and non‑consensual imagery.

Under newly amended Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, these intermediaries must now remove flagged unlawful or harmful content within three hours of receiving a court order or government notice, a sharp reduction from the earlier 36‑hour deadline. The updated framework also mandates prominent labelling of AI‑generated content, regular user warnings, metadata tracking and expanded accountability, with penalties for non‑compliance under existing criminal laws.

While the government says the move prioritises user safety and trust in the digital ecosystem, digital rights advocates and some industry experts say the shortened timeline and obligations could create implementation challenges, risk over‑removal and raise concerns about free speech.

Sharpened Enforcement: What the New Rules Require

The Ministry of Electronics and Information Technology (MeitY) has formally notified amendments to the IT Rules, 2021, which will come into effect on 20 February 2026 through Gazette Notification G.S.R. 120(E). These changes represent one of India’s most comprehensive efforts to regulate synthetic and AI‑generated content on digital platforms. The updated rules define “synthetically generated information” including AI‑created or altered audio, visuals and video that appear real and treat harmful versions of such content as unlawful, similar to child sexual abuse material, impersonation, fake documents, obscene material or content tied to explosives or other illegalities.

Under the new framework:

• Platforms must remove unlawful or harmful content within three hours of receiving a lawful notice from courts or authorised officers. Earlier, intermediaries had up to 36 hours to act, a window critics said allowed harmful material to spread widely before removal.
• In certain sensitive cases such as non‑consensual intimate imagery some reports indicate timelines could be as short as two hours.
• All AI‑generated or altered content must be clearly labelled as such before publication, with associated metadata or persistent identifiers that cannot be removed or suppressed. This aims to improve transparency and help trace the origin of manufactured content.
• Platforms must ask users to declare whether uploaded content is synthetic, use automated detection tools to verify these claims, and deploy safeguards to curb the creation and spread of illegal synthetic material.
• Intermediaries are required to inform users every three months about the consequences of violating content norms, including possible account action or legal implications.

Government officials have emphasised that these obligations, including the compressed timeframe, are necessary to ensure that harmful content which can rapidly go viral is neutralised before it fuels deception, discrimination or real‑world harm. A senior MeitY official told reporters that the rules “place enhanced due‑diligence obligations” on platforms and that tech companies “certainly have the technical means to remove unlawful content much more quickly than before”.

Background, Debate and Industry Reaction

The tightening of social media regulation in India has evolved against a backdrop of rising concern over the rapid proliferation of AI‑generated material, including deepfake videos, cloned voices and altered imagery that can mislead, defame or manipulate public opinion. Earlier laws already required intermediaries to act on flagged content within 36 hours, but digital experts and policymakers increasingly found this window insufficient given how quickly harmful content can spread across large user networks.

The 2026 amendments mark a significant shift from a notice‑and‑takedown model towards what some analysts describe as a proactive governance regime on AI content. According to industry sources, this places a heavier compliance burden on platforms seeking to balance speedy removal of harmful content with careful legal or contextual assessment.

Digital rights advocates have raised concerns about the practicality and implications of such compressed deadlines. Some warn that three‑hour windows make meaningful human review nearly impossible, potentially driving platforms toward automated takedown systems that could sweep up lawful or creative content in the process. Others argue that smaller platforms and startups, lacking as much moderation infrastructure as global tech giants, could face disproportionate compliance pressures.

Critics nevertheless acknowledge the importance of addressing synthetic harms. Experts say mandatory labelling and transparency are positive steps, but urge careful calibration to avoid chilling free expression or disproportionately burdening creators and users who produce legitimate synthetic content such as satire or creative art when it is clearly identified.

Political responses have also surfaced. In the Uttar Pradesh Assembly, a regional legislator called for specific laws against deepfakes and AI misuse, with state leaders debating how the Centre’s directive should be implemented at the state level. Officials pointed to the new Union government orders as part of a broader legal framework that could dissuade technological abuse.

The Logical Indian’s Perspective

We at The Logical Indian recognise the very real harms posed by deepfakes, misleading AI content and non‑consensual imagery that undermine trust, dignity and safety online. Transparent, clear labelling and faster response times can play a meaningful role in reducing misinformation and shielding vulnerable people from abuse.

However, the push for speed must be balanced with principles of fairness, due process and freedom of expression. Overly compressed timelines such as three‑hour takedown mandates risk incentivising automated systems that may inadvertently remove lawful or legitimate speech before proper review, suppressing nuance and undermining public confidence in digital spaces.