New Delhi, India – The Indian government is moving to significantly tighten regulations around artificial intelligence (AI) generated content on social media platforms. The Ministry of Electronics and Information Technology (MeitY) has unveiled draft amendments to India's IT Rules, which include a drastic reduction in the time platforms have to remove unlawful AI-generated material.
Under the proposed changes, platforms such as Facebook, Instagram, X (formerly Twitter), and YouTube would be required to take down AI-generated deepfakes and other illegal content within just three hours of receiving a valid notice. This marks a substantial shift from the current 36-hour deadline, signaling the government's heightened concern over the rapid spread of misinformation and harmful synthetic media.
Key Proposed Changes to AI Content Moderation
The draft rules outline several critical obligations for social media intermediaries:
- Expedited Takedown: The most significant change is the reduction of the content removal deadline from 36 hours to a mere 3 hours for unlawful AI-generated material. This applies to deepfakes, AI-enabled impersonation, child sexual exploitation and abuse material, non-consensual intimate images, and other illegal content.
- Mandatory AI Labelling: Platforms will be required to clearly label all images, videos, audio clips, and other content generated or significantly modified using AI. This includes both visible labels and the attachment of metadata to trace the content's origin, helping users distinguish authentic content from synthetic media.
- Proactive Detection and Prevention: The new regulations advise platforms to take strict action against harmful AI-generated content. This necessitates proactive measures, including the deployment of advanced technical solutions like AI-based moderation systems, automated detection tools, and image and video scanning to prevent the creation and distribution of such content.
- Quicker Response to User Complaints: Beyond AI-specific content, the draft rules also propose shortening the general deadline for addressing user complaints from 72 hours to 36 hours. For sensitive complaints involving impersonation, fake accounts, nudity, or sexually explicit content, the resolution timeline would be cut from 24 hours to just 2 hours.
- User Awareness of Legal Consequences: Social media platforms will also be obligated to inform users that sharing unlawful AI-generated content can have serious legal ramifications. This aims to foster greater responsibility among users regarding the content they disseminate.
These amendments underscore India's commitment to combating the challenges posed by AI-driven misinformation and deepfakes, placing a greater burden on social media companies to ensure a safer digital environment.