On 19 June 2026, the Delhi High Court did something no Indian court had done quite so explicitly before: it upheld an order blocking an entire messaging app, Telegram, across the country, rather than a specific channel, bot, or URL on it. The order came out of the NEET-UG 2026 paper leak investigation, after the government found that exam material was circulating through rotating Telegram channels faster than individual takedown notices could keep up. The justification offered to the court was not that any one piece of content was unlawful, but that the platform's own architecture made content-specific enforcement pointless.
That single sentence is the entire tension this piece is about. Section 69A of the Information Technology Act, 2000 was written to let the government block information on a computer resource. It says nothing, in its text, about blocking an app, a website, or a service wholesale. Yet courts have now blocked whole platforms more than once, and the legal reasoning behind each of those orders looks different depending on which judgment you read. If you run a platform that depends on Indian users, defend someone whose business just went dark, or are simply trying to understand why your government can switch off an app overnight, the gap between blocking specific content and blocking an entire service is where your case will actually be won or lost.
What Section 69A Actually Authorises
Section 69A empowers the Central Government, or an officer it authorises, to direct any intermediary to block public access to information hosted on a computer resource, but only on five specified grounds: sovereignty and integrity of India, defence of India, security of the State, friendly relations with foreign States, or public order, plus a residual category for preventing incitement to a cognizable offence connected to these. There is no sixth, open-ended ground. The procedure for exercising this power is laid down separately, in the Information Technology (Procedure and Safeguards for Blocking for Access of Information by Public) Rules, 2009.
Under Rule 5 of the 2009 Rules, it is a Designated Officer, acting on a request from a Nodal Officer or a competent court, who issues the actual blocking direction[3], and that direction must rest on one of the section 69A grounds. The Rules also build in an emergency track: where delay is unacceptable, the Designated Officer can recommend an interim block, which goes to the Secretary of the (then) Department of Information Technology for a same-day order, subject to review by a committee within 48 hours[4].The structure assumes, throughout, that someone is examining a specific piece of information and asking whether it falls within one of those five grounds. It was not written with an entire app in mind.
In practice, the first question is not simply whether access to a platform has been restricted, but under what precise statutory authority the restriction has been imposed. A direction under Section 69A, an intermediary compliance notice, a court-ordered injunction, and a temporary suspension of telecom services operate under different statutory frameworks and attract different procedural safeguards. Conflating these mechanisms can obscure the real grounds on which a blocking action may be challenged.
The Default Position: Block the Page, Not the Internet
The clearest statement of how narrow this power is meant to be comes not from a case about Section 69A in the abstract, but from the Supreme Court's 2020 ruling on the Kashmir internet shutdowns, Anuradha Bhasin. The Court was examining suspension orders under different provisions, but it went out of its way to locate Section 69A within the broader scheme of internet restrictions, and its language on this point is unambiguous[1].
The aim and purpose of Section 69A is not to restrict or to block the internet as a whole, but to block access to particular websites on the internet.
That sentence has done a lot of work in every Section 69A case decided since. It tells a Designated Officer, and later a reviewing court, that the statute's default unit of analysis is the specific resource, the specific URL, the specific item of content, not the service that happens to carry it. The Tripura High Court applied exactly this logic in 2021 in State of Tripura v. Soumen Sarkar, where it walked through the entire 2009 Rules machinery in detail and held that the grounds in Section 69A(1) are exhaustive: a blocking direction cannot rest on defamation, for instance, however serious the allegation, because defamation does not appear anywhere on that list[2]. The Court also held that a Magistrate has no independent power to order blocking outside this statutory chain; the power sits with the Designated Officer and ultimately the Central Government, and nowhere else[2]. Read together, these two judgments describe a power that is deliberately narrow on three axes: narrow grounds, a narrow chain of authority, and a narrow target, namely the offending content itself.
Where Courts Have Let an Entire Site Go Dark Anyway
That said, Anuradha Bhasin was decided in the context of internet shutdowns, not piracy, and Delhi High Court orders dealing with rogue websites tell a more complicated story. In 2016, the Department of Electronics and Information Technology, acting on a broadcaster's complaint, first restricted a blocking order to specific URLs carrying pirated cricket coverage. On review, the same matter came back, and this time the Court reversed course and directed that the entire website be blocked[5]. The reason given was practical rather than doctrinal: URL-level blocking on a rogue piracy site is trivial to defeat, because the operator simply moves the infringing stream to a new URL within minutes, and the rights holder is left filing fresh complaints in an endless loop while the broadcast it was trying to protect has already finished airing.
A practical distinction is important here. Challenges to blocking orders generally operate at two levels. The first is jurisdictional: whether the authority acted under the correct statutory provision, relied upon a permissible ground, and followed the prescribed procedure. The second is constitutional: even where the government possesses the power to act, whether blocking an entire platform was necessary and proportionate to the identified harm. A platform may therefore succeed without establishing that the underlying content was lawful; procedural illegality or the availability of a less restrictive and equally effective measure may independently undermine the blocking direction.
The Delhi High Court formalised this reasoning three years later in UTV Software Communication Ltd. v. 1337X.To, where it introduced what has since become known as the rogue website test[6].The test asks whether a website is qualitatively rogue, meaning its predominant purpose is infringement and it shows little regard for legal consequences, rather than merely quantitatively rogue, meaning it happens to host a lot of infringing material alongside legitimate content. Once a site clears that qualitative bar, the Court held, blocking the site as a whole, rather than chasing individual URLs, is a proportionate response, because the site's entire existence is bound up with the unlawful activity. This is the closest the Indian judiciary has come, outside Section 69A itself, to a coherent legal test for when whole-platform blocking is not overreach but the only enforcement method that actually works.
The analogy, however, has limits. A rogue website whose predominant purpose is infringement is materially different from a general-purpose intermediary carrying substantial lawful speech and commercial activity. The more legitimate uses a platform supports, the greater the collateral impact of a platform-wide restriction and the stronger the case for requiring the government to demonstrate why account-level, channel-level, feature-level, or URL-level measures would be ineffective. The rogue website doctrine should therefore not become an automatic doctrinal shortcut for blocking legitimate multi-use platforms.
The 2021 Rules, and Where a Court Pushed Back
Parliament and the executive have since layered an additional, faster blocking track on top of the 2009 Rules. Rule 16 of the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 lets an Authorised Officer, in a genuine emergency, recommend blocking directly to the Secretary, Ministry of Information and Broadcasting, who can issue an interim direction without first hearing the affected party, subject to committee review within 48 hours[7]. Structurally this mirrors the 2009 emergency Rule, but it sits inside a much more contested set of Rules overall, and at least one High Court has already found parts of that framework wanting.
In 2021, the Bombay High Court stayed Rule 9 of the 2021 Rules, the provision that required digital news and OTT publishers to follow a government-administered Code of Ethics and oversight mechanism, holding it prima facie ultra vires Article 19(1)(a) of the Constitution[9]. The Court was not dealing with platform blocking directly, but the ruling matters here because it shows that the 2021 Rules, taken as a package, have not had an easy run in the courts, and that judges are willing to strike down or suspend parts of the executive's expanded toolkit when they look like they regulate speech rather than specific unlawful content. Any blocking order traceable to this Rules framework, rather than to Section 69A's own narrower procedure, inherits some of that constitutional exposure.
National Security Cases Run on a Different Calculus
Courts apply noticeably less scrutiny when the platform being restricted is not blocked for the general public but for a specific, security-sensitive group. In LT COL P K Choudhary v. Union of India, the Delhi High Court was asked to interfere with an order banning serving Army personnel from using a list of social media and messaging platforms, including some that had nothing objectionable on them individually[8]. The Court declined, holding that questions of military discipline and national security fall substantially within the executive's domain, and that the judiciary will not second-guess a blanket platform restriction imposed for force protection reasons in the same way it would scrutinise a public-facing Section 69A order[8].The lesson is not that Section 69A's proportionality requirement disappears for security cases. It is that the audience of the restriction changes the analysis: a platform-wide ban on a defined, consenting group of government employees is judged far more leniently than the same ban applied to the entire Indian public.
The Telegram Order and Why It Goes Further Than Anything Before It
Which brings the discussion back to June 2026. Two separate MeitY directions are involved in the Telegram matter: one blocking the app nationwide until 22 June, and a second, narrower order disabling Telegram's message-editing feature until 30 June, aimed specifically at the mechanism investigators said was being used to alter and redistribute leaked exam content after the fact. According to reporting on the hearing, the government's core argument before Justice Tejas Karia was structural rather than content-specific: Telegram's channel and bot architecture let leaked material be re-hosted across new channels within minutes of any single channel being taken down, which made the kind of URL-by-URL or channel-by-channel enforcement that Anuradha Bhasin assumes practically meaningless, per Tech Times' coverage of the hearing.
To get there under Section 69A's text, the government leaned on Section 2(1)(v) of the IT Act, which defines information to include code, computer programmes, and software, not just discrete pieces of content. If the application itself, as a piece of software facilitating the unlawful distribution, qualifies as information within that definition, then blocking the application is, on this reading, simply blocking information at a different layer of abstraction, not stepping outside Section 69A's scope at all. Commentators have described this as the first time an Indian court has validated a full-platform block under Section 69A on that statutory footing, as opposed to the rogue-website doctrine the Delhi High Court used against piracy sites, see Bar and Bench's coverage.
The Internet Freedom Foundation pushed back hard on this reasoning in the days after the order, arguing that extending information to cover an entire platform collapses the distinction Anuradha Bhasin insisted on, and that millions of Telegram users with no connection to the exam leak lost access to a communication tool because a small number of channels misused it, a position set out in IFF's public statement. Whether that critique succeeds will likely depend on how the order is treated on appeal, and at the time of writing it has not been tested by a higher court. It is worth being precise about what is settled and what is not here: the order itself, and the specific judicial reasoning behind it, has not yet been entered as a judgment in Cassie's database, so it is discussed in this piece on the strength of contemporaneous reporting rather than as a numbered citation. Treat it as a live, unresolved data point, not settled precedent.
It is also worth keeping the Telegram order separate from a different and, in some ways, more far-reaching dispute decided by the Karnataka High Court earlier in 2025. In X Corp v. Union of India, X Corp challenged the government's Sahyog Portal and argued that Section 79(3)(b) of the IT Act, the intermediary safe-harbour provision, does not itself empower the government to issue blocking orders, and that this power belongs exclusively to Section 69A and its Rules[10]. The petition framed the two provisions as mutually exclusive pathways, with Section 69A as the only lawful route to a blocking direction. The distinction matters for platform operators specifically because it determines which procedural protections, including the hearing rights and review timelines built into the 2009 Rules, actually apply to a given takedown or blocking demand, and whether a government can route around Section 69A's safeguards by characterising an order as something else entirely.
The Opposite Instinct: Target the Page, Not the Platform
Not every recent case has pushed toward broader blocking. In 2025, the Supreme Court dealt with a Wikipedia-specific dispute arising out of a defamation suit, where a High Court Division Bench had directed the removal of specific Wikipedia pages discussing ongoing court proceedings[11]. Nobody asked the Court to block Wikipedia itself, and nobody seriously suggested it. The remedy under discussion stayed at the level of individual pages, which is precisely the unit Anuradha Bhasin said Section 69A was built around. Cases like this are a useful corrective to any impression that whole-platform blocking is becoming the default tool. It remains the exception, reserved for situations where content-level enforcement has been tried, or is demonstrably futile, not a first resort.
So When Does Whole-Platform Blocking Actually Hold Up?
Reading these cases together, rather than in isolation, produces a workable test, even though no single judgment states it in these exact terms. A whole-platform block is far more likely to survive judicial scrutiny when most or all of the following are true:
The grounds invoked map onto one of Section 69A's five named categories, or its incitement clause, not a broader or analogous harm the statute does not list
Content-specific enforcement has genuinely been tried, or there is concrete evidence that the platform's own design defeats it, as the courts found with rogue piracy sites and, on the government's account, with Telegram's rotating channels
The order followed the procedural chain in the 2009 Rules, through a Designated Officer and, where relevant, the committee review built into the emergency track, rather than an ad hoc direction outside that structure
The restricted population is either the general public acting on a narrowly drawn, content-linked justification, or a defined group such as government or military personnel where the institutional justification is independently strong
The block is time-bound and tied to a specific investigation or threat, rather than open-ended
The decision-making record demonstrates consideration of less restrictive alternatives, including blocking identified accounts, channels, URLs, functionalities, or other technical vectors of dissemination, and explains with evidence why those measures would not adequately address the identified harm.
Where these factors are weak or absent, the safer prediction, based on Anuradha Bhasin and State of Tripura, is that a court will treat the order as disproportionate and push the government back toward content-specific remedies.
From a litigation perspective, legality of a platform-wide block should not turn solely on whether the government can establish a legitimate objective. The government should also be able to demonstrate, through the decision-making record, why substantially less restrictive measures were technically inadequate. Assertions that content can be rapidly replicated or redistributed should ordinarily be supported by platform-specific evidence rather than treated as sufficient in themselves.
What This Means If You Are on Either Side of a Blocking Order
When a platform is affected by a blocking direction, the immediate exercise should be evidentiary as much as doctrinal. The first step is to identify the precise statutory basis of the restriction and obtain, to the extent accessible, the blocking direction and the material underlying the decision-making process. The authority issuing the direction, the statutory ground invoked, compliance with prescribed procedure, availability of hearing or review mechanisms, and consideration of less restrictive alternatives should all form part of the initial assessment.
At the same time, the platform should preserve all communications with government authorities, document its compliance history, identify the legitimate uses and user base affected by the restriction, and place on record technically feasible alternatives such as account-level, channel-level, URL-level, feature-level, or functionality-specific restrictions. Evidence demonstrating the economic, operational, and speech-related consequences of the block may also become relevant when proportionality is tested before a constitutional court.
A platform’s ability to challenge a blocking order may depend significantly on the systems and records already in place before the restriction is imposed. Documented responses to takedown requests, coordination between legal and technical teams, and evidence that narrower restrictions can be implemented effectively may help demonstrate that a nationwide block was not the least restrictive measure available. Such evidence can become particularly relevant when the necessity and proportionality of the restriction are tested in judicial review.
If you operate a platform that has just received, or fears, a blocking direction, the first document to demand is not a press statement but the actual order issued under Rule 5 or Rule 9 of the 2009 Rules, naming the Designated Officer and the specific Section 69A ground relied on. An order that cannot point to a named officer, a named ground, and a documented request is vulnerable on procedure alone, independent of any argument about proportionality. If the order instead arrived through Section 79(3)(b) or the Sahyog Portal rather than Section 69A, the X Corp litigation is directly relevant to whether that route was even available to the government for this purpose.
Beyond examining the legality of the order, a platform should build a clear factual record showing whether narrower measures were technically available. Evidence of past compliance with targeted takedowns, the effectiveness of account or channel-level restrictions, and the impact of the block on legitimate users can materially strengthen a proportionality challenge. In practice, the outcome may depend as much on this evidentiary record as on the interpretation of Section 69A itself.
If you are a user affected by a platform-wide block and have no commercial standing to litigate it yourself, your most realistic avenue is usually through a public interest petition or by supporting an organisation already challenging the order, since individual users rarely have the resources to mount a constitutional challenge over an app outage alone. Either way, the central question a court will ask is the one Anuradha Bhasin already answered for Section 69A in the abstract: was blocking the whole service actually necessary, or would blocking the specific channels, URLs, or accounts have achieved the same legitimate aim. The government carries the burden of showing the former. Increasingly, as the Telegram order shows, it is prepared to argue that a platform's own architecture can discharge that burden for it.