Deepfakes and harmful synthetic content have been increasing as the AI tools advance. However, the central government has now approved 13 ‘Responsible AI’ projects which are focused on developing the deepfake detection technologies. Along with funding research, the Ministry of Electronics and Information Technology (MeitY) has also introduced stricter rules for social media platforms, including Meta and X, to ensure faster action against fake content. The government said the combined focus on regulation and technology will help improve online safety and strengthen efforts to identify manipulated content before it reaches a wider audience.
Speaking at Lok Sabha, the Union Minister of State for Electronics and Information Technology Jitin Prasada announced the measures. He said that AI is developing so fast that it’s now much easier to make fake content look real. Moreover, he also pointed out that this kind of content can be misused for things like financial scams, stealing people’s identities, and spreading false information.
Also read: Sarvam is building 1T AI model, hires Elon Musk’s xAI expert to take on OpenAI and Anthropic
To deal with these issues, the government has tightened the IT (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021. Under the new rules which were introduced back on February 10, 2026, the social media platforms must now take down the AI-generated content within just 3 hours of getting a valid order from the government or a court. Prior to the change the platforms had up to 36 hours to respond or remove the content.
The updated rules also reduce the response time for user complaints involving sensitive matters such as impersonation, nudity and other harmful content. Officials said the changes are aimed at ensuring quicker action and better protection for users facing online threats.
Social media companies have also been instructed to clearly label AI-generated content wherever it is permitted. Moreover, they need to have traceable metadata, which will aid in identifying the synthetic media and technical measures, including the use of automated detection systems, to block the transmission of illegal material.
The government has also sanctioned 13 Responsible AI research projects in collaboration with educational institutes to enhance the deepfake detection ability in India. The most important one is Saakshya, which is a multi-agent system for detecting deepfakes and images, currently being developed at IIT Jodhpur and IIT Madras.
Aside from that, the government is also developing AI Vishleshak that can help in identifying the manipulated audio and video files using artificial intelligence. IIT Kharagpur is developing an in real-time voice deepfake detection system, which will detect AI-generated voices in real-time conversations.