Centre directs Meta to curb deepfakes

The Ministry of Electronics and Information Technology (MeitY) has asked Meta to share a detailed plan to update its algorithms and improve content moderation. According to Government sources, the goal is to stop the spread of propaganda, deepfakes, and illegal material on Meta’s platforms.

Senior Government officials gave these instructions to Meta representatives during meetings this week. Meta’s technical team described how they currently handle harmful online content. Officials asked Meta to make specific changes to its algorithms to slow the spread of propaganda, deepfakes, altered images, and other illegal or manipulated material.

Sources said the Government also told Meta to fully follow India’s data localisation rules and remove flagged content more quickly. The Centre asked Meta to work more closely with Quick Response Teams (QRTs) in different Government departments. This should help find and remove deepfake videos, altered images, and other misleading content faster, so online misinformation can be addressed in real time.

Officials also said that as concerns grow about the misuse of artificial intelligence to create fake content, platforms MUST be more accountable. They stressed that social media companies need to use technology to stop the spread of manipulated material and must follow the law and regulations under the Information Technology Act.

These meetings come as the Government pays more attention to Meta after recent controversies. These include the temporary restriction of a video post by Prime Minister Narendra Modi and reports of child sexual abuse material (CSAM) and deepfake content on Instagram and Facebook. Meta’s top executives were asked about its content recommendation algorithms, whether it follows Indian laws instead of foreign ones, and what steps it takes to stop bots and illegal material. The company admitted there are gaps in its systems and told officials it is “working hard” to fix problems with deepfakes, CSAM, and related issues.

Government sources said Meta has been told to follow the Indian Constitution and local laws. One official said, “You cannot run your platform in India and follow the US Constitution,” pointing out that Indian law allows certain limits on speech. The Centre said it will keep working for improvements, and more technical discussions are expected. Earlier this year, regulations became stricter. In February 2026, MeitY announced changes to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021.

These changes officially brought synthetically generated information, like deepfakes, under a clear due diligence process. Platforms now have to label AI-generated or altered audio-visual content with clear markers and add traceable metadata. Major social media companies must also ask users to say if their uploads are synthetic and use technical tools to check those claims.

The amendments also reduced the time platforms have to comply. Now, unlawful content must be taken down within three hours of a valid government or court order, instead of the earlier 36 hours. For sensitive cases like nudity or impersonation, the response time is two hours. If platforms do not meet these rules, they could lose safe-harbour protections under Section 79 of the IT Act.

With over a billion internet users, India needs strong content moderation. Deepfakes and AI-made propaganda have raised concerns about threats to public order, personal dignity, elections, and financial safety. The Government has supported local detection technologies through the IndiaAI Mission, approving several projects to spot synthetic media in real time.

Officials said the recent talks with Meta are not meant to punish the company but to improve cooperation and make sure platforms take responsibility. By asking for a formal plan, changes to algorithms, and better coordination with QRTs, the Centre wants to stop harmful content before it spreads widely. Meta is still a major player in India, with Facebook, Instagram, and WhatsApp used by hundreds of millions. The company’s efforts to work with the government and share clear plans will be watched closely in the coming days.

Sources said both sides have agreed to keep talking until all concerns are settled. As artificial intelligence tools become easier to use and more advanced, global platforms will face more pressure to align their systems with national priorities. More meetings are expected, and the government says lasting trust between platforms and Indian users depends on clear, ongoing improvements to protect against propaganda, deepfakes, and illegal content, sources added.