No Safe Harbour: What the Telegram judgment means for every platform in India

For the 150 million Indians who found Telegram unavailable on June 16, 2026, the six-day ban has come and gone. For every platform operating in India, however, it will continue to cast long shadows. The Delhi High Court's judgment upholding the block is an inflection point for the power to block content in India. Long after the NEET examination leaks of 2026 are forgotten, platforms in India will have to contend with the consequences of this judgment.

The judgment is unprecedented because it is a judicial endorsement of the blocking of an entire platform under Section 69A of the IT Act. The power under Section 69A extends to blocking "information" — defined in the statute to include data, messages, text, images, codes, computer programmes, and software. The Court held that since a platform involves software and code, it is itself "information," and can therefore be blocked in its entirety.

This is a somewhat strained reading of the statutory language. The components of the definition — data, text, images, code — are all formats or modes in which information can exist. A platform is not a format. Rather, it is a vessel on which information is “generated, transmitted, received, stored or hosted” in terms of 69A. Thus, if a platform is to be located anywhere in section 69A, it belongs in the category of "computer resource," not "information." That reading would imply that the statute permits the blocking of information on a platform — particular URLs, specific channels, or identified content — but not the blocking of the platform itself. If one were determined to read platforms somewhere within the architecture of Section 69A and the blocking Rules, this reading is more plausible than the one chosen by the Court.

But this actually begs the question - should one try to read such a substantial power into the statute in the absence of clear language anyway? Justice Scalia once memorably observed that legislatures do not hide elephants in mouseholes. A power to block a platform serving 150 million people is serious enough, and consequential enough, to have merited its own section - even a Chapter in the IT Act, its own procedure, or at the very least an explicit reference. That such a consequential power lay dormant and undiscovered for the eighteen years since section 69A was enacted strains credulity. Had Parliament intended to confer the power to block a platform, nothing stopped it from saying so directly.

The Blocking Rules of 2009 operate on the premise that it is individual pieces of information — particular URLs, identified posts — that are blocked. Can a platform used by millions to share vast and varied quantities of information be treated as a single unit of "information" susceptible to a single blocking order? The judgment does not engage with this question. If section 69A and the Blocking Rules had always permitted the blocking of everything from a single URL to a whole platform, one would have expected a more rigorous procedure for the more drastic remedy. No such procedure exists and the same rules govern all blocking, whether a URL of a platform.

Telegram admittedly complied with the government's directions, blocking and disabling hundreds of URLs that allegedly shared the examination materials. Due diligence and proper compliance with government directions, which are the requirements laid down under the Intermediary Guidelines, do not need anything further. Pavel Durov’s tweet, where he castigated the blocking as mistaken due to this compliance, is straightforward evidence of this. The judgment, however, treats the tweet as an admission that the efforts of Telegram were inadequate and ineffective, as contended by the State without additional scrutiny of the claim. It goes on to hold that the inadequate and ineffective measures taken by Telegram made the wholesale blocking of the platform to be proportionate. Thus, the judgment, indirectly, imposes on platforms a vague and hitherto unknown obligation, should they wish to avoid being blocked: the measures taken by them pursuant to government orders must be “adequate” and “effective” in the eyes of the government.

The troubling result is that a platform which demonstrably complied with every direction it received can still be found to have fallen short — and can be banned as a consequence.

Is there an objective standard for determining when compliance becomes effective enough to stave off a ban? The judgment does not supply one, nor does common sense. No platform has ever succeeded in removing every fraudster, every scammer, every bad actor from its services. If residual unlawful activity — despite active and documented takedown efforts — is sufficient to characterise a platform's compliance as inadequate, then every platform operating in India is, at any given moment, a single ministerial decision away from a blocking order.

There is a particular irony inbuilt in the structure of this problem. The Intermediary Guidelines extend a safe harbour to platforms that comply with their obligations. The Telegram judgment suggests that compliance with those obligations is no defence once the government has formed the view that the compliance has been ineffective or inadequate. The safe harbour, it turns out, offers no shelter from the most serious storm of them all: blocking.

The judgment also implicitly adopts a procedural model that is concerning. Telegram and the government had been in active dialogue for weeks before the blocking order was issued. What, then, was the “emergency” that necessitated an immediate blocking order instead of the normal procedure that called for a hearing before any such order was passed?  Was there, as contended by the State, an “escalation” of the situation at the time? Was such escalation sufficient to constitute an emergency? These value judgments by the State were, at the very least, contestable. However, the Court did not enter into this inquiry at all, instead accepting the framing of the situation as an emergency for the purposes of the Blocking Rules. Accordingly, the judgment allowed the State to block Telegram and offer only a post-decisional hearing, denying it valuable rights.

The Court has held that the ban was narrowly tailored because of its mercifully short tenure. However, narrow tailoring is intended to measure the breadth of the measure, not its duration. Even brief measures can be disproportionate because of what they entail. Putting 150 million innocent people in jail to catch 15 culprits is not proportionate, even if the jail time is only 2 hours. The point here was the wholesale blocking, not its duration.

The judgment carefully tries to anchor the ban to the particular features of telegram: large scale addition of users permissible, an architecture conducive to large scale amplification, and enabling storage of massive quantities of data. However, the ability to add users, store data and amplify communication is the whole point of all such platforms. Telegram may or may not be better than other platforms in delivering such features, but if the existence of such features is enough to permit wholesale blocking, WhatsApp, with many millions more customers, can be blocked by the same logic.

Why should this judgment give all Indian platforms sleepless nights? It is because it upholds a strained reading of the IT Act to empower the State to ban platforms. It is because substantial compliance and sustained engagement are not enough to stave off “emergency” blocking when the State feels the compliance is “inadequate” or “ineffective." It is because the Court found the blocking of a platform with 150 million users to be “proportionate” due to the actions of a few users in an analysis that applies the Anuradha Bhasin proportionality framework but in a way that is much more deferential than the Supreme Court intended. Considering that all blocking orders are confidential, challenges of this nature by platforms are the only means to obtain some sort of public accountability and this judgment leaves much to be desired on this front. This precedent will not confine itself to Telegram, or to examination fraud. It will be available to every government, facing every inconvenient platform, whenever it can characterise the situation as an emergency and the platform's efforts as insufficient.

For Platforms, policy experts and lawyers are undoubtedly working on small-bore changes in processes and compliance that could reduce risks of blocking going forward. Also, the Supreme Court precedent in Anuradha Bhasin continues to apply, subjecting blocking orders to vigorous proportionality tests that will not always be applied deferentially.

However, this judgment ought to be reversed. Until it is, platforms sink or swim in a sea of subjectivity guided by a paraphrase of Peruvian President Oscar Benavides’ memorable quip - "For my friends anything, for my enemies the (blocking) law."

About the author: Rahul Narayan is a Partner and the Head of Dispute Resolution at Chandhiok & Mahajan.

Disclaimer: The opinions expressed in this article are those of the author(s). The opinions presented do not necessarily reflect the views of Bar & Bench.

If you would like your Deals, Columns, Press Releases to be published on Bar & Bench, please fill in the form available here.