Block the App, Not Just the Post: How a Leaked Exam Got Telegram Switched Off Across India
The Delhi High Court upheld a nationwide, time-limited block of Telegram over NEET-UG 2026 exam-leak misuse, ruling that Section 69A can switch off an entire app, not just a post. What the judgment said, and why it matters.
For about a week in June 2026, one of the world's biggest messaging apps simply went dark in India. Open Telegram, and nothing loaded. Behind that blank screen sat a leaked medical-entrance exam, roughly 2.2 million anxious students, and a legal question the country had never squarely answered: can the government switch off an entire app, not just a single post?
On 19 June 2026 the Delhi High Court said yes. In Telegram FZ LLC v. Union of India, Justice Tejas Karia dismissed Telegram's challenge and upheld the block, and in doing so set out reasoning that will be quoted in every future fight over India blocking a platform. Here is what happened, what the court actually decided, and why it matters far beyond one exam.
- The Delhi High Court upheld a nationwide, time-limited block of Telegram imposed under Section 69A of the IT Act over NEET-UG 2026 exam-leak misuse. Telegram's petition was dismissed.
- Biggest ruling: the word "information" in Section 69A is broad enough to cover software and an app itself, so the government can block an entire platform, not only specific posts or channels.
- The court held the block passed the proportionality test from Anuradha Bhasin because it was time-bound and narrower takedowns had repeatedly failed.
- The court leaned heavily on Telegram's architecture, mass-broadcast channels, bots, username concealment, message-editing, and mirror channels that revive after takedowns.
- It is a major precedent for platform liability and free expression in India, affecting how the state can act against any app.
What the court decided
The government, through the Ministry of Electronics and Information Technology (MeitY), ordered Telegram blocked across India until 22 June 2026 and its message-editing feature disabled until 30 June 2026, days before a NEET-UG re-examination set for 21 June. Telegram argued the block was disproportionate, beyond the government's powers, and procedurally flawed. The court disagreed on every count and dismissed the petition. The block stood.
How we got here
- 21 May 2026: the National Testing Agency (NTA) tells MeitY that Telegram bots and channels are being used to leak and sell NEET-UG 2026 material.
- 1 to 9 June: MeitY convenes meetings with Telegram, shares a list of around 1,300 offending URLs; Telegram says it disabled about 900 of them.
- 16 June: MeitY issues the interim order under Section 69A, blocking Telegram nationwide until 22 June and disabling message editing until 30 June.
- 17 to 18 June: a committee under the 2009 Blocking Rules hears Telegram, then passes a final order confirming the block, citing reports from the NTA and the Indian Cybercrime Coordination Centre (I4C).
- 19 June: the Delhi High Court delivers judgment, dismissing Telegram's petition.
- 21 June: the NEET-UG 2026 re-examination is held.
The two sides of the case
Strip away the legalese and this was a clash between two reasonable positions. Here is how each side framed it.
| Telegram's case (strike the block down) | The Government's case (uphold the block) |
|---|---|
| Too blunt. Blocking the whole app punishes roughly 150 million lawful users to stop a handful of channels. | Nothing narrower worked. Specific channels were taken down repeatedly but kept reappearing as mirrors, with bots re-seeding the files. |
| Beyond the law. Section 69A lets the state block specific "information", not an entire platform. | Within the law. "Information" is defined to include software, so an app falls inside Section 69A. |
| We cooperated. Telegram disabled about 900 of roughly 1,300 flagged URLs and offered to do more. | Cooperation was not enough. Fresh leaks kept surfacing despite takedowns, with a re-exam only days away. |
| Disproportionate. Under Anuradha Bhasin the state must use the least restrictive measure; this was the most restrictive. | Proportionate and calibrated. The block was time-bound, days not forever, and tied to the exam window. |
| Flawed process. The first order lacked reasons and the satisfaction was mechanical. | Process followed. An emergency interim block, then a post-decisional hearing, then a reasoned final order. |
What the misuse looked like
The court's concern was not abstract. Public Telegram channels, some with tens of thousands of subscribers, advertised "leaked" papers and answer keys before and during the exam. Many of these are themselves scams that take students' money and deliver nothing, but their mere circulation can trigger panic, copycat leaks, and demands for a re-test. The reconstruction below shows the type of message at issue. It is illustrative, not a real post.
Question 1: did the government apply its mind?
Telegram argued the block was a knee-jerk order with no real reasoning, and that the later "final order" could not be used to backfill reasons the first order lacked. The court rejected this. Under Section 69A read with the 2009 Rules, in an emergency an interim block can be issued on the Secretary's satisfaction, followed by a post-decisional hearing before a final order confirms or revokes it. That is exactly what happened. The orders, the court found, rested on real material, the NTA and I4C reports, and were adequately reasoned. Telegram's point that it had complied with many takedowns did not, by itself, make the block invalid.
Question 2a: can Section 69A block a whole app?
This is the heart of the ruling. Telegram argued Section 69A lets the government block specific "information", a post, a channel, a file, but not an entire platform. The court read the statute the other way. "Information" is defined in Section 2(1)(v) of the IT Act to include "codes, computer programmes, software". An app or platform, the court reasoned, is software. So blocking public access to Telegram as a whole falls squarely within the power Section 69A grants. A narrow reading, confining it to individual posts, would, in the court's words, risk rendering the provision ineffective against a platform built to evade post-by-post takedowns.
The message-editing problem
One feature drew special attention: Telegram lets a sender edit a message, including its attached file, after it was posted. The court accepted the government's worry that this can be abused to make a leak look like it happened before an exam when it was actually planted after, manufacturing the appearance of a genuine paper leak. That is why the order separately disabled message-editing until 30 June. The illustration below shows the concern.
Question 2b: was the block proportionate?
Even with the power to act, the state must use the least restrictive measure, the test laid down by the Supreme Court in Anuradha Bhasin v. Union of India (2020). Telegram said blocking the whole app for 150 million users to stop a handful of channels failed that test. The court walked through the Anuradha Bhasin factors and held the block passed:
- Legitimate aim: protecting the integrity of an exam for 2.2 million candidates and averting a public-order problem.
- Rational nexus: blocking the platform where the leaks spread directly serves that aim.
- Necessity: narrower takedowns had repeatedly failed, because channels reappeared as mirrors and bots kept re-seeding content.
- Least restrictive: crucially, the measures were time-bound, the block ran only until 22 June and editing only until 30 June, tied to the re-exam window. That limited duration, the court held, made them narrowly tailored rather than a blanket ban.
On that reasoning, the action was held not disproportionate, and the petition was dismissed.
Why Telegram specifically
The judgment is unusually detailed about why Telegram, in the court's view, defeats surgical enforcement. The features it singled out:
| Feature | Why the court saw it as a problem |
|---|---|
| Large public channels | Content can reach tens of thousands of people almost instantly, amplifying a leak into a public-order risk. |
| Bot ecosystem | Automated accounts re-seed and redistribute content without human involvement, so takedowns do not stick. |
| Usernames instead of phone numbers | Operators stay anonymous, making attribution and targeted action hard. |
| Mirror channels | When one channel is removed, subscribers are funnelled to a fresh clone, reviving the network within minutes. |
| Message editing | A message and its attachment can be altered later to fake a pre-exam leak. |
Taken together, the court concluded, these made "block the specific channel" an inadequate remedy for this platform, which is what pushed it toward upholding a platform-wide measure.
Was the block necessary? The honest answer
On the facts before the court, the block is defensible, even if you instinctively dislike it. Three things make the case for necessity.
First, the clock. A re-examination for 2.2 million students was days away. A leaked or faked paper reaching tens of thousands of people in minutes is not a slow-burn problem; it can force yet another re-test and throw a national exam into chaos. In an emergency measured in hours, "keep filing takedown requests" is not a real remedy.
Second, the architecture genuinely defeats surgical action. This is the part critics sometimes skip. When a removed channel instantly reappears as a mirror, bots re-seed the file, and message-editing lets an old post be turned into a fake "pre-exam" leak, removing items one by one is like bailing a boat with a hole in it. The court did not assert this in the abstract; it recorded that narrower measures had been tried and had failed.
Third, the limits. The order was not "ban Telegram". It was a block until 22 June and an editing freeze until 30 June, bounded to the exam window. A temporary, dated restriction is a very different thing from an open-ended ban, and that boundedness is what carried it across the proportionality line.
Why it is still debatable. None of that erases the cost. A platform-wide block, however short, silenced roughly 150 million people who did nothing wrong, and the most determined offenders simply switched to VPNs while ordinary users lost access. It also normalises a powerful tool: once "switch off the whole app, briefly" is blessed by a court, the temptation to reach for it in less urgent situations grows. The honest verdict is not that the court was obviously right or obviously wrong, but that this was a hard call resolved in favour of exam integrity and public order, with a time limit as the safeguard, and reasonable people can disagree about where that line should sit.
What it means
For the government: a clear High Court endorsement that Section 69A can reach an entire app, and that a time-limited platform block can survive the proportionality test when narrower steps demonstrably fail. Expect this judgment to anchor future blocking orders.
For platforms: architecture is now a legal exposure. Features that frustrate targeted takedowns, anonymity, easy mirroring, post-hoc editing, can be cited as reasons a whole-platform block is justified. Faster, verifiable cooperation on takedowns becomes the best defence against the nuclear option.
Frequently asked questions
Is Telegram banned in India now? No. The block was temporary, tied to the NEET-UG 2026 re-exam window, and the message-editing restriction was set to lift on 30 June 2026. The judgment upheld that time-limited action, not a permanent ban.
What is Section 69A of the IT Act? It lets the central government direct the blocking of public access to "information" through any computer resource on specified grounds such as public order, following the procedure in the 2009 Blocking Rules. This ruling reads "information" broadly enough to include an app itself.
Did Telegram lose entirely? Yes. On both issues, application of mind and legality/proportionality, the court ruled for the government and dismissed the petition along with the pending application.
Could this happen to other apps? In principle, yes. The reasoning is not Telegram-specific in law; it turns on whether narrower measures fail and whether a block is time-bound and proportionate. The architecture findings, though, were tailored to Telegram.
Sources
- Delhi High Court judgment — Telegram FZ LLC & Anr. v. Union of India & Ors., W.P.(C) 8259/2026, 2026:DHC:5145 (19 June 2026) (full PDF)
- Section 69A, Information Technology Act, 2000 (India Code, Ministry of Law and Justice) · and Section 2(1)(v) definitions
- Information Technology (Procedure and Safeguards for Blocking for Access of Information by Public) Rules, 2009 (MeitY)
- Anuradha Bhasin v. Union of India, (2020) 3 SCC 637 (Supreme Court of India, proportionality test)