India Orders Social Media Platforms To Detect And Label AI Content Under New Rules

The Indian government has issued fresh directions to social media companies, tightening oversight around AI-created content. Under the new order, social media platforms must clearly mark all material generated using artificial intelligence and ensure it carries built-in identifiers that cannot be altered or erased later. The rules also prevent platforms from allowing users or creators to remove or hide these AI labels or the related metadata once applied. To reduce misuse, companies have been instructed to deploy automated systems capable of detecting and stopping the spread of unlawful, misleading, or sexually exploitative AI-generated material.In addition, social media firms must actively educate users about the risks and penalties linked

to AI misuse. The Modi government has mandated that such cautionary messages be shown at least once every three months, reinforcing accountability and awareness across digital platforms.These fresh guidelines arrive as stress grows over the massive rise of AI-driven deepfakes across popular social media platforms such as X, Instagram, Facebook and YouTube. These directions expand on draft changes floated last month by the Ministry of Electronics and Information Technology to the IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021.The proposed amendments aim to make users explicitly declare when they upload content created or altered using artificial intelligence, while also pushing tech giants to adopt systems that can verify these disclosures. In response, several social media companies have already introduced options that let users tag posts as AI-generated or AI-modified.At present, YouTube asks creators to flag videos that are “meaningfully altered” or synthetically produced in certain scenarios. This includes content that shows a real person saying or doing things they never did, modified visuals of real locations or events, or highly realistic scenes that never actually happened. Similarly, Meta has instructed users on Facebook and Instagram to label posts featuring digitally created or altered lifelike audio and visuals, such as AI-generated conversations, songs produced using AI tools, or reels narrated through artificial voiceovers.