IT Ministry advises social media platforms to take down ‘manipulated’ content

What has MeitY asked social media platforms to do about manipulated, misleading, or out-of-context content?

MeitY has asked platforms to scrutinize more than AI-generated deepfakes. They should act against false, misleading, manipulated, or out-of-context posts presented as current or authentic. Such content can confuse the public, create apprehension, disturb public order, and trigger law-and-order problems.

One example is a deepfake video that went viral this week, prompting outrage among Telugu film stars. The advisory also warns against old or archived photos and videos presented as current events. Platforms should prioritize complaints from authorized government agencies.

Platforms already must offer users a way to label AI-generated images under the 2021 IT Rules. Amendments earlier this year made deepfakes liable for takedown notices. A February change reduced response timelines to one to three hours. Meta has automatically removed some government-referred posts under Section 79(3)(b), though removal is not legally required. The advisory means faster, careful moderation, especially before the October 10 CJP protest.

What is the difference between an AI-generated deepfake, manipulated content, and an old image or video falsely presented as current?

An AI-generated deepfake fabricates or alters media using artificial intelligence, often making someone appear to say or do something. Manipulated content is broader: it can be misleadingly edited or presented, even without AI. An old image or video may be genuine, but becomes deceptive when falsely described as a current event.

For example, a deepfake video that went viral this week triggered outrage among Telugu film industry stars. By contrast, the advisory separately warns about archived photographs and videos presented as current. An old recording could mislead viewers as today’s footage, while AI could invent the scene entirely.

MeitY’s October 8 advisory warns about content affecting public order. The advisory supplements existing rules offering AI-image labelling and tells platforms to prioritise complaints from authorised government agencies. The warning arrives before the October 10 CJP protest and stresses risks from confusion and false context.

Why does the advisory focus especially on content that could create confusion, disturb public order, or affect social tranquillity?

The advisory focuses on content that could affect public order because people may mistake misleading posts for real, current events. That can create confusion, fear, unrest, and law-and-order concerns, especially when old material appears new.

The timing is significant: it came before the Cockroach Janta Party’s October 10 protest seeking the Chief Election Commissioner’s resignation. It also followed outrage over a viral deepfake involving Telugu film industry stars. An altered clip or recycled image can make events seem immediate, prompting reactions before facts are checked.

The advisory expands attention beyond AI deepfakes to manipulated, false, or deliberately out-of-context posts. Existing rules require platforms to offer AI-content labelling, while the advisory urges stronger checks and priority for government complaints. Since February, platforms have faced one-to-three-hour compliance timelines, leading Meta to automatically remove many referred posts and drawing criticism from digital rights groups.

How many government ministries and departments can forward online-content takedown notices to the IT Ministry, and how quickly may platforms be expected to respond?

India’s rules allow 65 ministries and departments to forward online-content takedown notices to the Information Technology Ministry. Platforms may face response timelines of just one to three hours, making rapid review or removal especially important during sensitive events.

The advisory followed outrage over a viral deepfake video involving Telugu film industry stars. It also warned against old photographs or videos presented as current events. Platforms were urged to scrutinize false, misleading, manipulated, or out-of-context posts.

Under Section 69(A) rules, the 65 government bodies can send notices through the IT Ministry. Separately, agencies can flag allegedly illegal content under Section 79(3)(b), though that route does not legally require removal. Meta has often removed such posts automatically after compliance windows fell to one-to-three hours, drawing digital-rights criticism.

What can happen to a post when an authorised government agency flags it under Section 79(3)(b), and why has Meta often removed such posts automatically?

When an authorised government agency flags a post under Section 79(3)(b), officials say they believe it is illegal. The notice does not legally force removal. Platforms can therefore assess the post instead of automatically deleting it.

The article gives Meta’s Instagram and Facebook as the clearest example. Although Section 79(3)(b) does not require removal, a February amendment reduced response deadlines to one-to-three hours. Meta has often chosen automatic takedowns, leaving little time for review.

This creates a fast but blunt system. The advisory also tells platforms to prioritise complaints from authorised government agencies. That may speed action against misleading or harmful posts, but it can also increase mistaken removals. Digital rights groups have criticised Meta’s approach because automatic removal can restrict lawful speech without a full platform review.

How do takedown notices under Section 69(A) of the IT Act differ from government requests made under Section 79(3)(b)?

The key difference is force: Section 69(A) notices use a formal government blocking process, while Section 79(3)(b) requests flag allegedly illegal content. Under the 2009 rules, 65 ministries and departments can send blocking requests to the IT Ministry under Section 69(A).

For example, a ministry may refer a post through the Section 69(A) route for official blocking. An authorised agency can instead tell a platform under Section 79(3)(b) that it believes the post is illegal. That route does not legally require removal, unlike a blocking direction under the government process.

In practice, the distinction has blurred. A February amendment cut response times to one-to-three hours, prompting Meta to automatically remove many Section 79(3)(b) referrals. Digital-rights groups criticised this, because nonbinding requests can still produce immediate takedowns. The October 8 advisory also urged platforms to prioritise authorised agencies’ complaints.

What is an online intermediary, and why do laws give social media platforms certain protections while also requiring them to respond to illegal-content notices?

An online intermediary is a service that connects users and hosts or distributes their content, such as a social media platform. Laws often give intermediaries conditional protection from liability because they did not create every post. In return, they must follow safety duties and respond to valid legal processes.

In India, authorised agencies can flag suspected illegal content under Section 79(3)(b) of the IT Act. That provision does not automatically require removal, but platforms may act quickly because amended rules reduced response timelines to one-to-three hours. Meta has reportedly chosen automatic removal for many such referrals, prompting criticism from digital rights groups.

The October 8, 2026 advisory urged platforms to prioritise complaints from authorised government agencies. It also warned about deepfakes, manipulated material, and old images presented as current events. These rules try to balance open communication with public safety, but fast removals can also risk taking down lawful speech before proper review.

Key Facts:

📌 Advisory covers manipulated, misleading, and out-of-context content.

📌 Response timelines fell to one-to-three hours in February.

📌 October 10 CJP protest precedes the public-order warning.

📌 MeitY issued its advisory on October 8, 2026.

📌 Deepfakes face takedown notices under amended 2021 IT Rules.

📌 Archived media falsely presented as current is also targeted.

📌 The advisory came before the October 10 CJP protest.