India has fired a starting pistol on deepfake governance. On 10 February 2026, the Ministry of Electronics & Information Technology issued Gazette notification G.S.R. 120(E). Consequently, the Information Technology (Intermediary Guidelines & Digital Media Ethics Code) Rules now include a formal definition for synthetically generated information. Moreover, the amendment compresses lawful takedown windows from 36 hours to three. The change takes effect on 20 February, giving platforms barely ten days to react. Such urgency signals New Delhi’s tougher AI Policy stance amid rising election misinformation fears. Stakeholders across industry, civil society, and legal practice are scrambling to decode duties. Meanwhile, global platforms face expanded due-diligence checks, labeling mandates, and real-time moderation pressures. This article unpacks the new framework, weighs competing perspectives, and outlines practical next steps. Readers will gain a precise roadmap for navigating India’s accelerated deepfake crackdown. Therefore, executives must reassess resource allocation immediately. Timelines remain brutally tight.

Rules Enter Into Force

MeitY's amendment was published on 10 February and activates on 20 February 2026. Consequently, the compressed timeline leaves intermediaries almost no buffer for phased rollouts.

The Gazette introduces “synthetically generated information,” covering audio, visual, or audio-visual content created through algorithms. Moreover, routine edits and hypothetical educational clips remain carved out, limiting overreach on benign Synthetic Media. Under the revised IT Rules, lawful takedown orders now demand action within three hours, down from thirty-six. Additionally, certain urgent categories, including non-consensual imagery, require removal inside two hours. These statutory changes mark India’s fastest digital enforcement shift yet. However, later sections explore how companies will cope. That signal anchors India’s assertive AI Policy among the world’s strictest governance regimes.

India's Compressed Takedown Timelines

Platforms now confront a granular matrix of deadlines. In contrast, previous frameworks granted longer buffers for investigation.

• Lawful orders: 36 hours → 3 hours
• Urgent SGI removals: 24 hours → 2 hours
• User grievances: 15 days → 7 days
• Urgent grievances: 72 hours → 36 hours
• Quarterly user warnings mandated

Consequently, moderation teams must operate 24/7 with India-based escalation officers and automated evidence capture. Experts argue the accelerated clocks reflect the government’s evolving AI Policy priorities instead of mere procedure. Moreover, missing a window risks losing safe-harbour protections, intensifying Compliance pressure. Analysts remind companies that notice periods start when the governmental email lands, not when legal teams acknowledge it. Therefore, latency in internal routing can inadvertently consume half the allotted window. These deadlines redefine operational baselines for every significant intermediary. Consequently, labeling duties become the next critical hurdle.

New Labeling Obligations Emerge

The amendment insists that all permitted Synthetic Media display visible notices and permanent provenance metadata. Therefore, platforms must integrate watermarking, C2PA signatures, or equivalent hashing solutions across upload pipelines. Rule 4(1A) also forces significant social platforms to collect user declarations at upload and verify them using technical measures. Consequently, any failure may constitute lack of due diligence under the IT Rules. This requirement aligns with the broader AI Policy vision for accountable innovation. Professionals tasked with implementation can deepen governance skills through the AI Executive Essentials™ certification, which covers provenance architectures and Compliance playbooks. Integrating provenance also touches mobile app latency budgets. In contrast, users expect instantaneous uploads, forcing clever client-side optimizations. Labeling rules create technical heavy lifting beyond simple content flags. Nevertheless, operational feasibility remains the leading industry concern discussed next.

Industry Raises Feasibility Concerns

Meta’s policy chief Rob Sherman welcomed the intent yet warned the three-hour clock is “really challenging.” Furthermore, platform lawyers highlight verification, evidence preservation, and local language moderation hurdles. Smaller Indian startups express deeper worries. In contrast, global giants can fund tooling and night-shift reviewers; emerging firms cannot match Compliance budgets. Sherman noted internal dashboards must map obligations across the IT Rules matrix to avoid gaps. Technical vendors add that reliable provenance infrastructure, while advancing, still lacks uniform standards across devices. Consequently, early deployments may suffer false negatives or brittle watermark removal. Industry associations urge MeitY to open a sandbox under the AI Policy roadmap, enabling iterative testing before penalties escalate. Operational doubts will influence rollout speed and accuracy. Therefore, civil-society critiques move the debate from logistics to rights.

Civil Society Rights Debate

The Internet Freedom Foundation argues compressed windows eliminate meaningful human review and foster automated over-removal. Moreover, critics fear prior restraint incompatible with constitutional speech protections. Civil groups also flag limited transparency around draft consultations. Consequently, they filed Right to Information requests to trace deliberations. Activists stress that Regulation must balance safety with due process, warning that unchecked filtering chills satire and political criticism. Legal scholars predict petitions will leverage proportionality doctrine established in recent Supreme Court privacy rulings. Consequently, the judiciary’s stance could recalibrate enforcement intensity. Rights defenders will likely pursue judicial review if over-zealous takedowns emerge. Subsequently, global observers watch for precedent that shapes wider jurisdictions.

Global Ripple Effects Expected

India hosts more than 500 million social users, giving its standards international weight. Therefore, product teams may adopt provenance defaults worldwide to avoid market-specific versions. Major studios worry about cross-border leakage of unlabeled Synthetic Media. Analysts compare the mandate to Europe’s GDPR moment for deepfakes. In contrast, some governments may emulate the model to accelerate their own Regulation. Meanwhile, investors predict rising demand for detection software, compliance consultancies, and certified workforce pipelines. Consequently, upskilling becomes a board-level priority. Forward-looking enterprises treat India’s stance as a bellwether for harmonized AI Policy frameworks across the Global South. International echoes could reshape platform architecture far beyond Delhi’s borders. Nevertheless, immediate focus remains on meeting next week’s deadline.

Conclusion And Next Steps

India’s deepfake amendment compresses takedowns, mandates provenance, and redefines intermediary duties under the IT Rules. Consequently, platforms must fuse detection, labeling, and record-keeping into live workflows or risk losing legal shields. Meanwhile, unresolved feasibility questions invite litigation and global copycats. Yet, the direction of the AI Policy appears irreversible as Delhi pursues harm reduction at speed. Leaders should audit resource allocation, engage with MeitY, and invest in specialist talent. Finally, professionals can future-proof careers by earning the AI Executive Essentials™ credential and turning Compliance burdens into strategic advantage. Moreover, clear internal playbooks, updated every quarter, will sustain momentum amid rapid rule evolution. Consequently, early movers may convert regulatory rigor into consumer trust and market differentiation.