A Delhi court's May 2026 order directing OpIndia to temporarily pull down two articles about journalist Swati Chaturvedi — reportedly on the ground that continued circulation could cause "irreparable harm" — is, on its face, a narrow civil defamation matter. In reality, it is the latest data point in a much larger trend: Indian trial courts are increasingly using interim injunctions to extinguish online content before any full hearing on the merits. The cumulative effect is a quiet but consequential shift in how speech is regulated in India — one that bypasses statute, sidesteps appellate scrutiny, and pushes platforms toward over-removal.
The injunction reflex
India's framework for online speech is, on paper, fairly speech-protective. Section 66A of the IT Act, which criminalised vaguely "offensive" online messages, was struck down by the Supreme Court in Shreya Singhal v. Union of India (2015) as unconstitutionally vague. Section 79 of the IT Act, read with the 2021 Intermediary Guidelines, gives platforms safe harbour so long as they act on valid legal orders. And in March 2024, in Bloomberg Television Production Services India v. Zee Entertainment, the Supreme Court explicitly reaffirmed the "Bonnard rule" — that pre-trial injunctions restraining publication should be rare, reserved for cases where the defence of truth or fair comment is plainly unsustainable.
Trial courts have not always read the memo. The Bloomberg court complained that lower courts were granting ex parte gag orders "in a mechanical manner," treating injunctions as routine relief in defamation suits. The OpIndia order is best understood against that backdrop. Whatever one thinks of the underlying articles, an interim takedown — issued before the defendant has fully tested the plaintiff's case, and without a trial finding that the speech is defamatory — sets a precedent that publishers and platforms are now navigating dozens of times a month.
From publishers to platforms
The OpIndia case targets a publisher. But the same procedural shortcut is now being used to conscript platforms into upstream policing. In March 2026, Motorola's India arm sued X, YouTube, Instagram, Facebook, Threads, Google and Meta over more than 360 user posts alleging that its devices were unsafe, asking the court not only to take down existing posts but to prevent similar posts from appearing in future. As Rest of World reported, what makes the suit unusual is the decision to name platforms as co-defendants rather than file standard takedown requests — a litigation strategy that effectively asks a civil court to impose proactive monitoring obligations.
That ask collides directly with the Supreme Court's own jurisprudence. Shreya Singhal held that intermediaries are only obliged to remove content on receipt of a court order or government notice under Section 79(3)(b); general monitoring is not part of the bargain. Yet "John Doe" injunctions and "dynamic" takedown orders — originally crafted for piracy of films and live cricket — are increasingly being extended to political and reputational speech. Each individual order looks modest. The aggregate is a de facto notice-and-staydown regime built by civil courts, with none of the procedural safeguards a legislature would have to debate.
The platform incentive problem
For platforms, the rational response to legal uncertainty is to over-remove. The Motorola suit illustrates the dynamic neatly: faced with a defamation claim naming them as defendants rather than third parties, platforms can either litigate each post on the merits — expensive, slow, and reputationally fraught — or pull the content and move on. Digital rights groups quoted in the Rest of World piece warned that this is precisely the chilling effect such suits invite.
The cost is not borne by the platforms alone. It is borne by every reviewer, journalist, whistleblower or ordinary user whose lawful criticism is collateral damage. India does not need an EU-style Digital Services Act to import a German NetzDG-grade incentive to remove first and adjudicate never; its own courts are building one a docket at a time.
What proportionate regulation looks like
None of this is an argument against defamation law, or against meaningful remedies for harmful content. A democracy can — and India does — protect reputation, dignity and physical safety. But three corrections would bring practice back in line with both Indian constitutional doctrine and global best practice:
- Reserve pre-trial injunctions for clear cases. Trial courts should treat the Bonnard rule, as reaffirmed in Bloomberg v. Zee, as a binding standard rather than a soft preference. "Irreparable harm" cannot mean "the article is still online."
- Stop naming platforms as co-defendants in user-speech disputes. Section 79 already provides a clean takedown route. Civil suits that drag intermediaries in to manufacture proactive monitoring obligations should attract costs, not orders.
- Build appellate plumbing for interim orders. Today, an ex parte takedown can sit in force for months while appeals crawl. A fast-track review mechanism — modelled on the urgency the orders themselves invoke — would restore symmetry between speech and silence.
India has the world's largest internet-using population and one of its most vibrant digital public spheres. Preserving that depends on resisting a quiet drift toward a regime in which the easiest way to lose an argument online is to be the one who answered second. The OpIndia order is a single ruling. The pattern it fits is the policy problem.