On 6 November 2026, X told the Delhi High Court that two takedown reports filed in the Delhi Police's name against the Internet Freedom Foundation (IFF), SFLC.in and the account @geoblockedindia appear to be fake. According to IFF's account of the hearing, X said its internal investigation found that the email the reports came from "appears to be fake". It has complained to the Delhi Police and is "not likely to take action on the reports at this stage". The court recorded the statement and disposed of the petitions (W.P.(C) 16671/2026) with liberty to return if any action is taken.
Nothing was removed, so the immediate harm was contained. The episode still shows how a forged document could have reached a platform's compliance queue, and how little stood between it and a takedown.
The strongest case for fast takedown channels
The case for quick, informal channels is real. Platforms host impersonation, threats and incitement that can do irreversible harm within hours. Police often know first. A rule requiring a court order for every removal would be slow and would leave victims exposed. India's 2021 intermediary rules also push platforms to act quickly on government intimations. Most of that urgency is legitimate, and a platform that ignores genuine law-enforcement requests invites worse regulation.
That case depends on one thing: the request has to be genuine and lawful. This episode suggests neither was checked at the point of entry.
What the record shows
IFF says the sequence began with a legal notice on 29 October from a Proton Mail address, purportedly on behalf of the Rashtriya Swayamsevak Sangh (RSS). The notice named no enrolled advocate. Two days later, on 31 October, a report in the name of the Special Commissioner of Police (Crime) reached X through its public reporting form. A second report followed on 2 November, listing 40 URLs and naming the three accounts. IFF says the court-order fields on both forms were left empty, and that the second report ticked "enforcement of court order" anyway. The reports cited Section 356 of the Bharatiya Nyaya Sanhita (BNS), the criminal defamation provision, and Section 79(3)(b) of the IT Act. They said "lawful notices require action within 3 hours".
Two outlets report that the Delhi Police also denied sending the reports, according to Newslaundry's syndicated report from The News Minute. The RSS did not officially confirm sending the notices. These are reported statements. X's conclusion is a preliminary finding, and the identity of the sender is still unknown.
Why the legal theory failed on its own terms
Even if the reports had been genuine, the legal basis was thin. In Shreya Singhal v. Union of India, decided on 24 March 2015, the Supreme Court read down Section 79(3)(b). The judgment held that an intermediary loses safe harbour only on "actual knowledge" through a court order or a notification by the appropriate government or its agency. It added that either must "strictly conform to the subject matters laid down in Article 19(2)", the list of permitted grounds for restricting speech. Defamation of a private organisation is a listed ground, but a one-line reason covering 40 URLs does not show that the grounds are met.
IFF also argues that BNS s.356 is a non-cognizable offence. On its account, police cannot investigate it without a Magistrate's order, and cognizance generally requires a complaint by the person defamed. IFF says the Delhi Police has no public statement on who authorised the reports. SFLC.in points to a Delhi Home Department order of 2 June 2026 listing officers authorised to issue Section 79(3)(b) notices. It notes that the Special Commissioner (Crime) is not among them. These are petitioners' arguments, not findings. The court did not rule on them, because X's statement made the dispute moot for now.
The structural problem
The reports reached X through a public form, with no way to check who sent them. A genuine notice and a forged one looked the same on arrival. This is the design weakness in informal takedown channels. They put the burden of verification on a private platform, which has every incentive to comply quickly and no easy way to authenticate the sender.
The separate challenge to the Delhi police-notice framework makes the point sharper. SFLC.in challenged a December 2024 notification by the Delhi Lieutenant Governor that empowers Delhi Police officials to issue takedown notices. The petition argues that blocking power sits solely with the Central Government under Section 69A. If authority is contested even when real, a forged claim of it is hard to catch.
What proportionate reform looks like
None of this requires slowing genuine emergency takedowns. Three changes would keep speed and add accountability.
- Authenticated channels. Government requests should arrive through verified portals or signed, domain-authenticated email. Public web forms should not accept official-capacity reports.
- Published authorisation lists. Platforms and the public should be able to check which officers may issue Section 79(3)(b) notices, and the list should be kept current.
- Transparency by default. Requests should be published promptly to a database such as Lumen, so targets learn of them and forgeries surface quickly. X itself forwarded the reports to IFF on 3 November, which is how the challenge was filed so fast.
Transparency worked here. X's decision to notify the targets and to investigate is what stopped the reports. That depends on platform goodwill, and it should be a requirement. A takedown system that can only be trusted when a platform happens to act well is fragile.
What happens next
The petitions were disposed of with liberty to return, so the broader questions stay open. These include whether such reports comply with Rule 3(1)(d) of the IT Rules, who may issue them, and whether platforms must verify them. The practical lesson is narrow. Speech is protected best by making real authority easy to verify and by making forged or unlawful requests easy to see.