“Precisely because there is a misuse potential, because it’s being used by so many people, we always had been emphasising that this power be restricted only to certain limited people,” said Rakesh Maheshwari, Former Senior Director and Group Coordinator of MeitY’s Cyber Laws Division, referring to Section 79(3)(b) of the IT Act, which allows government officers to send content takedown notices to platforms.
Maheshwari, the longest-serving designated officer for Section 69A and the official who helped build India’s content takedown architecture, made this observation at the second session of MediaNama’s discussion on IT Rules and the Future of Online Speech in India on April 23 in Delhi, where he explained how the framework was designed, where it has gone wrong, and why the proposed Rule 3(4) crosses a line. Aditi Agarwal of Tech Trace moderated the session.
How 69A and 79(3)(b) were designed to work: India’s content takedown framework rests on two provisions of the IT Act, 2000. Section 69A governs blocking orders. Section 79(3)(b) governs takedown notices directed at platforms.
“At the time the act was written, it was maybe assumed that it is under the control of something which Internet Service Provider (ISP) will be able to control, and hence the word blocking. Under 79, the word is takedown because the content is in the control of that particular platform,” Maheshwari said.
With advances in technology, both blocking and takedown became possible globally, not just in India.
On how Section 69A works, Maheshwari walked through the process.
“For each state, there is a nodal officer, there is a touch point. For ministries which are likely to be impacted because of such content and which have a national security aspect, those are also the ministries which have got their own nodal officers,” Maheshwari added.
Requests are received and processed by a committee of five joint secretaries from relevant ministries, including the Department of Law and Justice and the Department of Legal Affairs. The Secretary gives final approval, with a provision for review every two months. “So per se, checks and balances are there,” he said.
Section 79(3)(b) is what Maheshwari described as a “distributed function, focusing on the unlawfulness.” Any authorised officer of a certain rank or higher can issue a takedown notice to a platform. The grounds are broader, covering any unlawful content, and the procedural safeguards are fewer.
“The power should be available only with certain limited number of people, and they should be accountable in the system if they believe that something which is unlawful and for which notice is required to be issued,” Maheshwari said.
The Sahyog portal, which centralises these notices, currently has at least 94 intermediaries onboarded. X Corp is challenging it before the Karnataka High Court.
The discrepancy in who can send notices: Maheshwari added that Delhi notified 23 officers at the Deputy Commissioner of Police (DCP) level and above to send Section 79(3)(b) notices. Karnataka was demoted to the rank of Assistant Commissioner of Police (ACP). Some states went to the Deputy Superintendent of Police (DSP) level. Bihar chose just one officer.
“Because of such discrepancy, it is good that at least in 2025, the number of touchpoints have been reduced,” Maheshwari said.
On historical notifications, he was direct. “After 2025, I believe those notifications also need to be reviewed. Once again, we made it available. That has not been done,” he said.
“There should be a consolidated list of such notifications which should be made available so that at least people know as to whether you are authorised or not, by which notification, and whether it is within the law or not,” Maheshwari said.
On the absence of signatures on notices, he flagged a concern. “It’s a serious communication of maybe unlawful thing which has been taken into cognisance. Should it not be signed by the person who has been so empowered?” he said.
What platforms are actually expected to do: The February 2026 amendments compressed the takedown window for unlawful content to three hours, down from 36 hours. The three-hour provision did not appear in the October 2025 draft rules and went through no public consultation before MeitY added it to the final notified version.
“Three hours basically mean that you have to see the completeness of the notice, the competency of the person who is actually issuing it, authorisation of this person, and the sections of the law which is being quoted,” Maheshwari said.
“Maybe the government’s expectation is, further, do not look into content because that is where a responsible officer of the government has already done that due diligence,” he added.
On what makes a notice complete, he was clear. “Just simply sharing a URL saying that it is unlawful is not good enough. What is really required is some SOP which really talks of the elements of what makes it a complete and comprehensive notice to be acted upon,” he said.
When Nikhil Pahwa, Founder of MediaNama, pointed out that three hours work for a single order but not for hundreds of orders at a time, Maheshwari agreed. “I agree. I’m only trying to now justify what is there in the law,” he said.
“I completely agree that if something is already lying for five years and all of a sudden, government or somebody finds it out and you ask it to be removed in three hours, it’s quite illogical,” Maheshwari added.
Snehashish Ghosh, Founder of TechNiti and former Meta policy executive, argued that the three-hour timeline originated from a voluntary code of ethics for a very specific use case: Section 126 violations during election campaign silence periods, where the decision tree was binary, and the enforcement window was fixed. “There might be situations where three hours might be warranted,” he said, but extending that standard to all unlawful content was not proportionate.
“In three hours, what you are asking an intermediary to do is not apply their mind,” said Rahil Chatterjee, Partner at Ikigai Law. “If there is something to claim that this is a defamatory post, should the intermediary be sitting in that position to decide, yes, this is defamation or not defamation?” he added.
Advisories should not have the force of law: On the proposed Rule 3(4), which would make MeitY-issued advisories, SOPs, and guidelines binding on all intermediaries as a condition of safe harbour, Maheshwari drew a clear distinction.
“As long as there are conditions which have been stated, that they have to be issued in writing, they must clearly provide the statutory provision in the rule, maybe they should be considered. But in respect of advisory, I have a different opinion because advisory is per se advisory in nature. It cannot be considered part of the rules,” Maheshwari said.
He pointed to the recently published Digital Personal Data Protection (DPDP) rules as evidence. “In the DPDP rules, at many places, exactly the same thing has been conveyed. There is nothing of such nature in the IT Act. To that extent, maybe from a law perspective, it is an invalid proposition,” he said. “Guidance, directions, SOPs are all valid. Advisories continue to be advisory,” he added.
Giving advisories the force of law impacts every digital business, not just social media platforms, argued Rahil Chatterjee, Partner at Ikigai Law. “Tomorrow, you’re going to have an advisory saying, hypothetically, a cloud service provider must do XYZ. An e-commerce platform must do XYZ. It is linked to the safe harbour. That is the real fear in three, four,” he said.
No SOP for when to take down an account vs a post: Section 79(3)(b) has no criteria for when a platform should take down an individual post versus an entire account or channel, Maheshwari said. Under Section 69A, MeitY developed SOPs on this distinction. No equivalent exists for Section 79(3)(b).
“Something similar should also be there in respect of section 79(3)(b) as well, as to how do you decide an individual post, individual video versus a complete channel or a complete account. Criteria need to be clear,” he said.
Why de-platforming is done? Because the impact it creates is much, much higher. Removing a video, removing one post actually creates more problem than it solves,” Maheshwari added.
“It is a problem of politics”: The problem stated in Session 1 was not a legal one but a political one, Maheshwari said. The law cannot solve it.
Vasudev Devadasan, lawyer and Master of Laws candidate at the University of Melbourne, argued that the legal framework addresses the wrong crisis entirely. “Today, the crisis we’re facing is government censorship. It’s government notices. Your right to free speech is directly engaged,” he said.
MediaNama hosted this discussion with support from Meta, Amazon and Google. Our community partners for this event are the Internet Freedom Foundation (IFF), the Centre for Communication Governance (CCG), and Digipub.
Also read:
For You
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。