India's content moderation architecture is at a constitutional inflection point. X Corp's continuing challenge in the Karnataka High Court to the Union government's Sahyog portal — a centralised takedown interface operated by the Indian Cyber Crime Coordination Centre (I4C) under the Ministry of Home Affairs — is forcing a long-overdue judicial conversation about a quiet but consequential shift in how the State asks platforms to remove speech in India.
At stake is not whether the government can require intermediaries to act on unlawful content. It plainly can. The question is whether the procedural floor laid down by Parliament and read into the IT Act by the Supreme Court in Shreya Singhal v. Union of India (2015) can be routed around by treating Section 79(3)(b) — the intermediary safe-harbour provision — as an independent takedown power.
What the Sahyog portal actually does
Sahyog, launched by the MHA in late 2024, is described officially as a coordination tool to streamline notices issued to intermediaries. In practice, it allows an expanding list of "authorised officers" across central ministries, state police, and law-enforcement agencies to direct platforms to remove URLs, posts, and accounts. The government's position is that these are merely "notices of unlawful content" under Section 79(3)(b) of the Information Technology Act, 2000 — communications that, if ignored, cost a platform its safe harbour.
X's contention, articulated through its Karnataka High Court petition, is that this framing has quietly built a parallel censorship regime. Section 69A, the express blocking power, comes with a written reasoned order, a review committee of senior bureaucrats, and confidentiality obligations that the Supreme Court accepted as procedurally sufficient in Shreya Singhal. Section 79(3)(b), by contrast, was never designed to be a blocking power at all. It is a safe-harbour conditionality clause. Repurposing it as a primary takedown route — at scale, through a portal, with thousands of officials empowered to push the button — is constitutionally different in kind, not degree.
The Bombay High Court already saw the pattern
This is not the first time the executive has tried to widen its takedown perimeter beyond Section 69A's guardrails. In Kunal Kamra v. Union of India (September 2024), the Bombay High Court struck down the amended Rule 3(1)(b)(v) of the IT Rules, 2021 — the so-called "Fact Check Unit" amendment — that would have empowered a government-designated FCU to flag content about the Union government's own business as "fake" or "misleading," with intermediaries losing safe harbour if they did not comply.
Justice A. S. Chandurkar, writing the tie-breaking opinion, found the provision violated Articles 14, 19(1)(a) and 19(1)(g) of the Constitution. The reasoning matters: a takedown mechanism cannot be sustained simply because it routes through intermediaries rather than directly censoring the speaker. If the practical effect is the suppression of constitutionally protected speech without the safeguards demanded for direct restriction, the courts will look through the form.
Why proportionate regulation, not parallel regimes, is the answer
India is right to want effective tools against genuinely harmful content — CSAM, incitement, coordinated fraud, doxxing. Platforms, including X, have every reason to cooperate with such requests, and most do, swiftly. The problem with Sahyog as currently designed is not its existence but its architecture:
- Scale without scrutiny. A portal that allows hundreds of officials across multiple states to issue takedown demands at click-speed, without a centralised reasoned order, inverts the Supreme Court's Shreya Singhal bargain.
- Opaque categories. Section 79(3)(b) covers "unlawful" content broadly. Without a transparent taxonomy and published guidelines, intermediaries are pushed to over-remove to preserve safe harbour.
- No public accountability. Unlike Section 69A orders, Sahyog notices are not routinely disclosed even to affected users, foreclosing the possibility of judicial review of specific takedowns.
A proportionate fix is available and well-trodden internationally. The EU's Digital Services Act, for all its complications, requires statements of reasons for takedowns, out-of-court redress, and transparency reports. Brazil's Marco Civil insists on judicial orders for most content removals outside narrow categories. India's own Section 69A regime, properly resourced, already provides a workable template — what is missing is the political will to channel takedowns through it rather than around it.
What the Karnataka High Court should do
The court need not strike down Sahyog wholesale to vindicate the constitutional concern. A narrower remedy would suffice: read Section 79(3)(b) as not authorising executive takedown orders, but only the communication of court orders or Section 69A directions; require reasoned, written notices for any Sahyog-routed demand; and direct that the categories of officials empowered to issue notices be published and rationalised.
This is the kind of judicial calibration India's free-speech jurisprudence has long favoured — preserving the State's legitimate regulatory interest while refusing to let safe-harbour mechanics become an unaccountable censorship pipeline. An open, innovation-friendly internet in India depends on it. So, ultimately, does the trust that allows global platforms to invest deeply in the Indian market rather than retreat from it.