Government Strengthens Framework to Counter AI-Generated Deepfakes Through Legal Safeguards, Platform Accountability and Citizen Protection

Key Highlights:

• The Government of India is strengthening its framework to counter AI-generated deepfakes through legal safeguards, platform accountability, and citizen protection.
• Key legislation includes the Information Technology Act, 2000, Bharatiya Nyaya Sanhita, 2023, and Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021.
• Recent amendments to the IT Rules, 2021 (February 2026) mandate clear labeling for AI-generated content, strengthen user accountability, and reduce content removal timelines.
• Significant Social Media Intermediaries (SSMIs) face stricter obligations, including deploying automated tools to detect unlawful content and tracing originators.
• Indigenous AI initiatives like Gyan Bharatam, Gyan Mitram, and BharatiyaGPT are preserving cultural heritage, while projects under the IndiaAI Mission focus on deepfake detection.
• Awareness campaigns by Information Security Education and Awareness (ISEA) and Indian Computer Emergency Response Team (CERT-In) educate citizens on cyber hygiene and deepfake threats.

Detailed Insights:

• The IT Act, 2000 addresses identity theft, impersonation, privacy violations, and allows blocking or removal of unlawful content.
• Bharatiya Nyaya Sanhita, 2023 penalizes misinformation causing public mischief (Section 353) and organized cybercrimes (Section 111).
• IT Rules, 2021 obligate intermediaries to prevent hosting content that is obscene, misleading, impersonating, or threatens national security.
• SSMIs must publish compliance reports, appoint local officers, and offer voluntary user verification and internal appeals.
• A Grievance Redressal Mechanism requires intermediaries to appoint Grievance Officers, with users able to appeal to Grievance Appellate Committees (GACs).
• The February 2026 amendments explicitly cover child sexual exploitation material (CSEM) and non-consensual intimate imagery (NCII).
• Timelines for removing unlawful information upon government or court intimation are reduced from 36 hours to 3 hours.
• Grievance redressal timelines are shortened to 36 hours (general) and 2 hours (sensitive matters like nudity/impersonation).
• Intermediaries lose their Section 79 exemption from third-party liability if they fail to observe legal obligations.
• Advisories have been issued to intermediaries on due diligence, handling religious content, and synthetically generated information.
• A Standard Operating Procedure (SoP) was formulated in November 2025 to curtail the dissemination of Non-Consensual Intimate Imagery (NCII).
• Deepfake detection projects include Saakshya (IIT Jodhpur & IIT Madras) and AI Vishleshak.

Key Concepts Involved:

• Deepfakes: AI-generated synthetic media (audio, video, text) that appears authentic, often used for malicious purposes.
• Intermediaries: Entities that store or transmit data on behalf of others, such as social media platforms, internet service providers, and search engines.
• Significant Social Media Intermediaries (SSMIs): Social media intermediaries with a large user base (50 lakhs or more) in India, subject to stricter regulatory compliance.
• Synthetically Generated Information (SGI): Any information, including deepfakes, created or significantly altered using artificial intelligence technologies.