The Ruling and Its Reach
On June 19, 2026, Justice Tejas Karia of the Delhi High Court handed down a decision that will outlast the exam season that prompted it. Upholding the Union government's June 16 blocking order under Section 69A of the Information Technology Act, 2000, the court declared that a six-day nationwide ban on Telegram — affecting an estimated 150 million Indian users — was "least restrictive and proportionate." A separate order disabled the platform's message-editing feature through June 30. The case was Telegram FZ LLC & Anr. v. Union of India & Ors.
This was not simply a ruling about one app or one examination. It was a formal judicial endorsement of something Indian digital rights advocates have long warned against: using Section 69A to extinguish an entire intermediary, rather than targeting specific unlawful content within it.
The Government's Case Is Genuinely Strong
To understand the ruling fairly, start with the facts on the ground. On May 12, 2026, the National Testing Agency (NTA) cancelled NEET-UG 2026 — India's high-stakes medical entrance examination — after investigators found exam questions circulating through Telegram groups before test day. Over 22.8 lakh students (approximately 2.27 million) were directly affected. Refunds exceeded Rs 300 crore. At least three student suicides were reported in the aftermath. A re-examination was scheduled for June 21.
MeitY did not reach for the block order immediately. After NTA's May 21 alert, the government issued notices, held meetings with Telegram between June 1 and 5, and provided a consolidated list of fraudulent channels on June 9. Telegram complied with 900 of the 1,300 identified URLs. The problem was structural: "mirror channels" and "backup networks" reconstituted blocked groups nearly instantly. The government's argument — that targeted enforcement was functionally impossible within the exam timeline — had real evidence behind it.
The proportionality framework the court applied came from Anuradha Bhasin v. Union of India (2020), the Supreme Court's landmark internet shutdown ruling. Four questions: legitimate objective, rational nexus, necessity, and least restrictive measure. On each, the court found for the government, noting the temporary duration (six days for the platform, two weeks for the editing feature) and the extraordinary public interest in examination integrity.
Where the Legal Reasoning Breaks Down
The government's factual case may be strong. Its legal theory is something else.
Section 69A was drafted to enable blocking of specific "information" — a URL, a post, a piece of content. The IT (Procedure and Safeguards for Blocking for Access of Information by Public) Rules, 2009 contemplates notice, review, and targeted action. What Justice Karia held is materially different: that "information" under the IT Act encompasses "codes, computer programmes and software" — meaning an entire application is itself blockable as a unit of information. By this logic, if a platform's architecture makes targeted compliance structurally difficult, the government may treat the whole platform as the offending content.
This interpretation has no internal limiting principle. Signal, WhatsApp, Briar, and Session share Telegram's core architecture: end-to-end encryption, large group capabilities, and decentralized relaying that makes content persistence possible. These are precisely the design choices that make these apps valuable — and, under this ruling's logic, that make them vulnerable to the same emergency blocking argument the next time a government agency finds one inconvenient.
The court's treatment of proportionality also struggles under scrutiny. Calling a ban on 150 million users "least restrictive" requires a quiet definitional shift: least restrictive among options that could realistically work within a six-day exam window, not least restrictive in absolute terms. But that framing lets the urgency of the timeline do the constitutional work. Users circumvented the block immediately via VPNs and alternative apps, confirming that the measure was, in the analysis of Tech Policy Press, "both overbroad for innocent users and remarkably ineffective at stopping determined criminal networks."
The Normalization Risk
India's content-blocking apparatus was already growing rapidly. The government blocked a record 28,000 URLs in 2024 alone — up sharply from previous years. The historical aggregate shows over 22,000 websites blocked under Section 69A between 2015 and 2022. Until June 2026, platform-wide suspensions had been reserved for geopolitically exceptional circumstances — the 2020 TikTok ban, framed as a national-security action against Chinese applications.
The Telegram ruling creates a new category: exam-emergency blocking. It provides a replicable template. The next time any government agency identifies a platform as the vector for sufficiently serious content spread — election misinformation, communal incitement, financial fraud — the June 19 precedent supplies the legal architecture for a platform-wide shutdown, dressed in the proportionality language the Supreme Court blessed in Anuradha Bhasin.
The secondary effects on intermediaries are equally concerning. A platform that knows it can be switched off entirely will calibrate its compliance posture accordingly. The incentive shifts decisively toward over-compliance: quietly removing content that might attract regulatory attention even without a legal basis, rather than resisting ambiguous demands. As one commentator observed, "the Indian state no longer needs to censor the message if it can simply turn off the medium."
What Reform Requires
The NEET paper leak crisis was real, and the government had legitimate cause to act. But cause to act and cause to block an entire platform serving 150 million users are distinct thresholds that India's legal framework currently fails to separate.
A proportionate approach would require judicial pre-clearance for any platform-wide block (not a post-decisional hearing that ratifies a fait accompli), a statutory ceiling on duration, and a genuine burden on the government to demonstrate — not merely assert — that targeted alternatives were exhausted. The current Section 69A framework provides none of these at the platform scale, because it was never designed for it.
The IT Act is now twenty-six years old. Parliament should define the outer boundary of Section 69A explicitly: what counts as a blockable unit of "information," what procedural safeguards apply when that unit is a platform with hundreds of millions of users, and what independent review must occur before a shutdown order is executed. Until that clarification comes, every large-scale encrypted service operating in India operates under a judicially validated off-switch.