AI Video Advances Amid New Indian Regulations
AI company Runway has launched a new video generation model, Gen 4.5, with HD quality and native audio, signaling rapid advances in synthetic media. A recent podcast noted that this progress coincides with countries like India implementing new rules requiring the labeling and control of such AI-generated content, creating new compliance challenges for digital commerce.
- Runway's Gen 4.5 model introduces significant improvements in motion realism, making objects move with more believable weight and physics compared to previous versions. It also offers stronger adherence to complex user prompts and greater temporal consistency, meaning details like fabric texture or facial features remain stable throughout a video clip. - India's new regulations are amendments to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, which went into effect on February 20, 2026. The rules introduce a legal definition for "synthetically generated information" (SGI) to cover AI-made content. - The core mandate requires digital platforms to ensure any AI-generated content is clearly and prominently labeled. Furthermore, this content must have embedded, non-removable metadata or unique identifiers to help trace its origin. - The amended IT Rules drastically shorten the timeframes for removing illegal or harmful content. Platforms now have as little as two to three hours to take down certain categories of unlawful material, a significant reduction from the previous 36-hour window. - Social media intermediaries must now ask users to declare if their uploaded content is AI-generated and then use automated tools to cross-verify these declarations. Failure to properly disclose and label AI content can result in the platform losing its "safe harbour" legal protections. - These rules are part of a broader "techno-legal" AI governance framework India is developing, which integrates with existing laws like the Digital Personal Data Protection (DPDP) Act, 2023. This approach focuses on embedding legal and technical safeguards directly into the design and deployment of AI systems.
