Copyright Liability of Intermediaries: Safe Harbour in the Age of Digital Platforms

Introduction:

For more than two decades, the “Safe Harbour” principle under Section-79 of the Information Technology Act, 2000^[1] has played an important role in the growth of the internet in India. This provision protected online platforms such as social media websites, search engines and content-sharing platforms from being held legally responsible for illegal or infringing content uploaded by their users. As long as these platforms acted as intermediaries and removed unlawful content after receiving notice, they were not treated as the publishers of that content. This system, often called the “notice-and-takedown” model^[2], allowed digital platforms to function without the constant fear of legal liability for user-generated content.

However, the digital environment has changed with the rapid development of generative artificial intelligence (AI) and the rise of synthetically generated information (SGI). AI systems can now generate text, images, music, and videos that may closely resemble existing works, raising questions regarding who should bear responsibility for such content. In response, the Indian government introduced the Information Technology Amendment Rules, 2026^[3], which move beyond the traditional notice-and-takedown approach and require intermediaries to exercise “proactive vigilance.” Platforms are now expected to take reasonable steps to detect and prevent unlawful or infringing content. This shift suggests that the protection provided under Section 79 may become narrower as digital platforms are expected to take greater responsibility for the content shared on their services.

The Traditional Shield: Section 79 and Passive Cnduits:

For many years, Section 79 of the Information Technology Act, 2000 acted as a protective shield for online intermediaries. Platforms were not held responsible for illegal or infringing content posted by users as long as they acted as a “passive conduit.” This means the platform only provided the space or technology for sharing information and did not create, start, or modify the content. Because of this limited role, the law treated platforms as neutral facilitators rather than publishers.

Indian courts have supported this view. In Myspace Inc. v. Super Cassettes Industries Ltd.^[4], the Delhi High court held that an intermediary becomes responsible only when it has “actual knowledge” of infringement. In Shreya Singhal v. Union of India^[5]The Supreme Court clarified that intermediaries are required to remove unlawful content only upon receiving actual knowledge through a court order or government notification. This judgement reinforced the limited liability framework for intermediaries under Section 79. This approach is known as the “notice-and-takedown” model.

The 2026 Paradigm Shift: From Passive to Proactive:

With the growth of artificial intelligence, the government introduced the Information Technology Amendment Rules, 2026. The rules recognise synthetically generated information (SGI), which includes AI-generated text, images, videos, or audio. Since AI tools can generate large amounts of content quickly, some of this material may copy or resemble existing copyrighted works.

Under Rule 3(1)(b)^[6]intermediaries must take reasonable technical measures to manage such content. Platforms may use automated tools or detection systems to identify AI-generated material. This shows a shift from reacting to complaints to preventing problems in advance.

Key Changes Affecting Digital Content:

The amended rules introduce several significant changes. One change is the three-hour takedown requirement. Earlier, platforms had up to 36 hours to remove unlawful content after receiving an order. The new rules reduce this time to three hours for certain harmful content, including AI-morphed images, deepfakes, or content affecting the sovereignty or integrity of India.

Another significant change is mandatory labelling and metadata for AI-generated content.^[7] Platforms must clearly label SGI and attach traceable metadata. If platforms fail to follow these requirements, they may lose safe harbour protection under Section 79. Courts are also examining whether platforms remain neutral. In observations related to the Lifestyle Equities CV v Amazon Technologies Inc.^[8] dispute, the Delhi High Court suggested that safe harbour may not apply if a platform actively promotes or controls certain content.

Challenges for Digital Platforms:

These rules create practical challenges for digital platforms. Platforms must carefully examine complaints and legal orders, which requires the application of mind. Although automated moderation tools are allowed, the large volume of online and AI-generated content makes timely review difficult.

Large companies may manage these requirements, but smaller platforms and start-ups may struggle. Another challenge is the changing line between an intermediary and a publisher. When platforms integrate their own AI tools, they may become involved in creating or shaping content and could face direct responsibility for copyright infringement.^[9]

The Way Forward:

Going forward, a balanced approach is necessary to ensure both effective copyright protection and the smooth functioning of digital platforms. Clearer regulatory guidelines should be developed to define the responsibilities of intermediaries when dealing with AI-generated content and automated moderation systems. Digital platforms should also invest in reliable detection technologies, transparent labelling practices, and stronger compliance mechanisms to identify and manage synthetically generated information. At the same time, the government may consider providing practical compliance frameworks and reasonable implementation standards so that smaller platforms and start-ups are not placed at a disadvantage. Such coordinated efforts between regulators, courts, and digital platforms can help create a system that protects copyright while also supporting innovation and responsible use of artificial intelligence.

Conclusion:

The 2026 regulatory framework represents a significant shift in the regulation of digital platforms in India. Earlier, intermediaries relied on the argument that they were neutral platforms hosting user content. The new rules show that this approach is no longer sufficient.

Although safe harbour protection under Section 79 still exists, it has become more conditional. Platforms must detect unlawful content, label AI-generated material, and respond quickly to legal orders. These developments strengthen copyright protection but also increase compliance responsibilities for digital platforms. The future of intermediary liability will depend on how effectively the law balances copyright protection with the need to preserve innovation, free expression, and the open nature of the internet.