[This is a guest post by Rudraksh Lakra.]
Introduction
On March 30, 2026, the Ministry of Electronics and Information Technology (“MeitY”) released the Draft Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Second Amendment Rules, 2026 (“Draft Amendment”), inviting public comments within a fifteen-day window. The government described the Draft Amendment as “clarificatory and procedural in nature.” They are neither. They are a structural expansion of executive power over online speech, dressed in the language of administrative tidying-up.
This piece proceeds in three parts. The first analyses the Draft Amendment on its own legal terms, examining four distinct expansions: the new ministerial direction power, the data retention carve-out, the extension of content regulation to ordinary users, and the transformation of the Inter-Departmental Committee from a grievance body into an instrument of censorship. The second connects the Draft Amendment to the documented recent wave of content and account takedowns in India, and the third situates the Draft Amendment within the framework of structural digital authoritarianism.
Part I: What the Draft Amendment actually do
The government’s characterisation of this Draft Amendment as clarificatory deserves to be confronted directly. A clarificatory rule resolves ambiguity within an existing framework. It does not create new obligations, confer new powers, or expand jurisdictional reach. Each of the four substantive changes introduced by this Draft Amendment does at least one of those things. The “clarificatory” framing is a strategic misrepresentation designed to minimise parliamentary scrutiny and public resistance.
Rule 3(4): Governing by advisory
The insertion of Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 (“IT Rules 2021”) Rule 3(4) grants MeitY the power to issue binding clarifications, advisories, orders, directions, standard operating procedures, codes of practice, and guidelines to intermediaries. Compliance with these instruments is made a condition of due diligence under Section 79 of the Information Technology Act, 2000 (“IT Act”). Failure to comply means loss of safe harbour, which in turn means direct legal liability for all third-party content hosted on a platform. The practical effect is that MeitY can issue a paragraph-length advisory, and every major social media platform in India must comply or face existential legal exposure.
The rule does include procedural guardrails. Directions must be in writing, specify their legal basis, define their scope, and be consistent with the Act. But these safeguards are self-referential. MeitY issues the direction, and MeitY determines whether it is consistent with the Act. There is no independent pre-compliance review, no judicial authorisation requirement, and no public notification mandate. Moreover, MeitY directions are often not made public, like the one mandating the installation of the Sanchar Saathi app on all devices. These procedural requirements are akin to window dressing.
This rule is also ultra vires the IT Act. It is a well-established principle of Indian administrative law that delegated legislation must operate within the limits of the parent statute. Section 87 of the IT Act authorises the making of rules, but it does not contemplate the creation of binding obligations through instruments such as “clarifications, advisories, directions, or standard operating procedures”. The statutory scheme requires that rules be laid before Parliament for legislative oversight and accountability. Rule 3(4) departs from this framework by conferring binding force on executive instruments that bypass both parliamentary scrutiny and the prescribed rulemaking process. In doing so, it effectively enables the creation of legal obligations outside the structure contemplated by the Act.
The Supreme Court’s 2015 judgment in Shreya Singhal v. Union of India (2015) remains the foundational constitutional constraint here. The Court struck down Section 66A of the IT Act precisely because it allowed executive restriction of speech without adequate judicial safeguards. IT Rules 2021 Rule 3(4) revives the same constitutional problem: executive-issued instruments, potentially vague, potentially secret, carrying the sanction of safe harbour loss for non-compliance.
Data Retention: Privacy as collateral damage
The Draft Amendment to IT Rules 2021 Rules 3(1)(g) and 3(1)(h) insert savings clauses making existing deletion obligations subordinate to retention requirements under “any other law for the time being in force.” On its face, this resolves a genuine legal tension: an intermediary ordered to take down content should not simultaneously be required to delete data that another law requires it to preserve for investigation. The logical case for the carve-out is real.
The problem is the absence of any limiting principle. The phrase “any other law” is unrestricted. It encompasses not just the Digital Personal Data Protection Act, 2023 (“DPDPA”) and financial regulations, but any future law, executive order, or rule that imposes a retention requirement. Under Section 8(7) of the DPDPA, a Data Fiduciary must erase personal data when the purpose for which it was collected is no longer served, subject to any other legal retention requirement. The Draft Amendment effectively subordinates this right to erasure to an open-ended governmental mandate to preserve data, with no time limit, no proportionality standard, and no requirement of judicial authorisation. The mandatory 180-day retention period can be extended further by government mandate, raising serious risks of surveillance and data breaches. The state is acquiring a record of users’ digital expression; held on terms they cannot challenge and for purposes they cannot know.
Extending content regulation to users: Overbreadth problem
The substituted proviso to IT Rules 2021 Rule 8(1) marks a significant expansion in the scope of content regulation. The original proviso limited the application of Part III to intermediaries only for Rules 15 and 16, namely, content blocking and emergency blocking. The amended proviso now extends this to Rule 14, thereby bringing both intermediaries and user-generated “news and current affairs content” within the jurisdiction of the Inter-Departmental Committee.
This expansion must be read alongside ongoing constitutional challenges to the IT Rules 2021. Under Rules 9(1) and 9(3), the Code of Ethics and the three-tier grievance redressal mechanism were stayed by the Bombay High Court on 14 August 2021 as prima facie violative of Article 19(1)(a) and ultra vires the IT Act. The Court also permitted challenges to the oversight mechanism under Rules 14, 15, and 16 upon the constitution of the Inter-Departmental Committee. This position was affirmed with a stay order with a pan-India effect by the Madras High Court decision in T.M. Krishna v. Union of India (2021), which cautioned that a government-controlled oversight mechanism risks undermining media independence. These challenges are presently pending before the Delhi High Court.
Against this backdrop, the Draft Amendment appears to reintroduce, in an expanded form, a regulatory architecture circumventing the stay order. By extending the scope of Rule 14 to include “matters” and by bringing user-generated content within its ambit, the executive effectively broadens its oversight powers while the constitutionality of the underlying framework remains unresolved.
The overbreadth of this provision operates on two levels. First, the category of “news and current affairs content” remains undefined. News today is not limited to political reporting; it includes technology coverage, financial commentary, travel, and lifestyle content. A travel vlogger documenting public infrastructure, a technology reviewer critiquing a government-backed platform, or a financial creator analysing policy decisions could all fall within its scope. There is no limiting principle, no threshold of influence, and no clear boundary that excludes ordinary content creation from regulatory control.
Second, the scale of what is potentially regulated is staggering. The Reuters Institute’s 2024 Digital News Report found that 71% of Indians rely on online media for news, with 54% consuming news through YouTube, 48% through WhatsApp, and 35% through Facebook. The 2025 Digital News Report reinforces this trend, documenting a rising preference for influencer-led, video-driven, personalised news formats, particularly among younger audiences. Independent YouTube creators now function as primary news sources for tens of millions of Indians. Dhruv Rathee commands 31 million subscribers. Ravish Kumar, a former NDTV anchor turned independent commentator, reaches 14.5 million, and Akash Banerjee’s satirical The Deshbhakt has 6.59 million.
At the level of individual autonomy, this expansion affects free speech as self-fulfilment. Expression is not merely instrumental but constitutive of identity, creativity, and personal development. When individuals operate under conditions where an undefined and expansive category of their speech may attract regulatory scrutiny, expression becomes cautious and restrained. Over time, this produces a form of internal discipline, where individuals narrow their speech in anticipation of potential consequences. The harm lies not only in the content that is removed, but in the speech that is never articulated.
At a structural level, these changes also undermine free speech as a means of democratic self-governance. As Robert Post argues, democratic legitimacy depends on the ability of citizens to participate in the formation of public opinion through what he terms “public discourse”, which includes all forms of communication socially recognised as necessary for such participation. Similarly, Jürgen Habermas grounds legitimacy in a communicative process governed by the principles of inclusion and equality, where all affected individuals can participate freely and on equal terms, and where outcomes are shaped by the force of the better argument rather than institutional power (see, here, here, and here).
The extension of blocking powers to a vast and undefined category of user-generated content directly alters these conditions. When large segments of communication fall within a regulatory framework that permits executive intervention, both inclusion and equality are affected. Participation becomes contingent, and expression is shaped by the anticipation of sanction. This does not require constant or visible intervention. The mere existence of such authority reshapes the communicative environment, introducing asymmetries that privilege certain narratives while discouraging others.
In such a setting, public discourse is no longer the product of open and equal participation, but of filtered visibility mediated by state pressure. Citizens engage with an informational environment that reflects compliance decisions rather than unconstrained deliberation. Democratic processes may continue in form, but the conditions necessary for the formation of independent public opinion are progressively weakened.
IDC transformation: From grievance body to censorship apparatus
The Draft Amendment to IT Rules 2021 Rule 14 completes the picture. The original Rule 14(2) confined the Inter-Departmental Committee (“IDC”) to hearing complaints about Code of Ethics violations. The amended version expands its mandate to hear “matters” referred to it by the Ministry of Information and Broadcasting (“MIB”), without any requirement that the matter arise from a complaint, relate to a Code of Ethics violation, or involve an affected party who has been heard. The MIB can, on its own motion, refer any content-related matter to the IDC. The linguistic shift from “complaints or grievances” to IT Rules 2021 the matter” in Rule 14(5) removes even the formal requirement of an aggrieved complainant. This could convert IDC from a reactive grievance mechanism into a proactive censorship committee, operable at the MIB’s initiative and unconstrained by any external trigger.
Connecting the dots
These developments are reinforced by parallel regulatory changes that operate along the same logic. The Synthetically Generated Information Rules Amendment, 2026, shortens takedown timelines to as little as three hours. Such compressed timelines leave little scope for platforms to assess legality or context, pushing them towards immediate removal as a risk-avoidance strategy. The result is predictable: over-removal of lawful but sensitive content, driven by the fear of liability.
At the same time, the Sahyog regime introduces a centralised mechanism through which the government can issue takedown requests across platforms. This bypasses the procedural safeguards under Section 69A of the IT Act, which require reasoned orders and review. By enabling executive action without these constraints, Sahyog risks creating a parallel system of content control outside the statutory framework. The regime is currently under challenge before the division bench of the Karnataka High Court. Read together, these measures point to a clear pattern: speed, centralisation, and executive discretion are being prioritised over process and accountability.
Part II: IT Rules 2021 Draft Amendment in Context
This Draft Amendment does not arrive in a vacuum. It arrived in the middle of a documented and accelerating wave of content suppression. As Prateek Waghre documented in March 2026, India was experiencing a “fresh mini-wave of restrictions/suspensions,” with 42 instances of account-level restrictions on X recorded between March 11 and March 19 alone. Multiple prominent accounts, including parody handles, journalists, and political activists, were withheld under the notice. The accounts targeted were those critical of the government’s treatment of minorities, India’s response to the US-Israeli conflict, and the LPG crisis, with many expressing dissent through satire and political parody.
Anmol Jain, reflecting on the recent event, argues that what is happening in India is not reducible to any single law or institution. It operates through a hydra, a multi-headed organism in which every attempt to cut one instrument of suppression leaves the others fully intact and functioning. The Draft Amendment fits precisely into this model. They do not replace existing blocking mechanisms. They add new ones, create new enforcement routes, and extend the reach of the existing architecture to users who were previously outside it. As Jain documents, the government has already been using the IT Rules 2021 to create an alternative mechanism for itself to unilaterally control the online social media space by threatening the intermediaries with possible legal consequences for hosting inconvenient third-party content. The Draft Amendment accelerates and formalises this process.
The government’s approach to discourse management, he argues, operates through three overlapping strategies: amplifying ideologically convenient voices, silencing inconvenient ones, and, where those tools prove insufficient, deploying coercive power directly. The Draft Amendment corresponds to the second and third of these strategies. IT Rules 2021 Rule 3(4) and the IDC expansion give the government new instruments for silencing inconvenient voices through platform coercion. The extension of content regulation to users broadens the silencing mechanism to the entire ecosystem of independent digital commentary that has emerged, precisely because institutional media has been progressively captured.
Part III: Digital structural authoritarianism and the architecture of silence
Tony Roberts and Marjoke Oosterom, based on a systematic literature review, define digital authoritarianism as involving a specific actor or actors, the powerholder, using digital technology or a combination of analogue and digital tools to carry out a repressive act such as surveillance or disinformation, producing a first order effect such as the repression of public opposition to secure a second order effect, namely the retention or consolidation of political power.
This definition captures much of what is unfolding in India. Yet James S. Pearson introduces a crucial extension. His work highlights the limits of intention-based frameworks, which assume a deliberately repressive agent consciously deploying digital systems for authoritarian ends. He argues that authoritarian outcomes can also emerge without deliberate design, for instance, through attention-harvesting algorithms that privilege amplification, polarisation, or suppression as by-products of optimisation logic.
Building on this, we can refine the account further to develop a concept of digital structural authoritarianism. The key shift here is analytical. The concern is not only what the state does with an architecture of control, but what that architecture does by virtue of its existence. Once established, it begins to shape behaviour, incentives, and outcomes on its own. Seen through this lens, the Draft Amendment to the IT Rules, 2021, reflects digital structural authoritarianism. The architecture of regulation and platform governance produces repressive effects irrespective of whether each action is guided by a consciously articulated authoritarian objective. This does not exclude intent. Rather, it shows that intent and structure operate together, while remaining analytically distinct. Digital authoritarianism in India thus operates as both intentional and structural, anchored in law, infrastructure, and platform design that reinforce each other, making the architecture of suppression self-sustaining in ways that outlast any individual act of censorship.
This Draft Amendment forms part of a broader regulatory continuum that includes recent additions such as the Income Tax Amendment, 2025, the Telecommunications Act, 2023, the Broadcasting Services (Regulation) Bill, 2023, and the Post Office Act, 2023. Several of these measures have been framed as efforts to modernise or decolonise existing legal regimes. In practice, they expand the state’s capacity to shape digital communication across multiple layers. Their combined effect extends regulatory reach across content creators, users, and the technological stack itself, including the application, network, and infrastructure layers. The outcome is a consolidation of authority over what circulates within the digital public sphere.
The cumulative trajectory resembles what Jürgen Habermas describes as the erosion of the conditions necessary for legitimate democratic authority. His communicative theory of legitimacy rests on the idea that outcomes should be determined by the force of the better argument rather than the power of the speaker. When the state occupies overlapping roles as narrator, regulator, and adjudicator of permissible discourse, the communicative process is constrained at its source. Citizens under such conditions cannot form accurate preferences if the informational environment is shaped from above.
Part IV: End Notes
The Draft Amendment expands executive authority, recalibrates the balance between privacy and retention, extends content regulation across the full breadth of India’s online information ecosystem, and reshapes an oversight mechanism into a more interventionist body. Placed in context, these changes align with recent patterns of content restriction, including a shift towards account-level actions and accelerated takedown practices. They reinforce a broader system of discourse management that operates through a combination of legal rules, administrative discretion, and platform compliance incentives.
The significance of this shift lies in its structural character. Regulation operates at the level of individual decisions and also shapes the environment within which those decisions are made. When that environment encourages caution, limits visibility, and concentrates authority over information flows, the effects are felt across the public sphere. Democratic institutions may continue to function in form. Yet the quality of democratic life depends on more than institutional continuity. It depends on the conditions under which citizens speak, listen, and deliberate. The Draft Amendment, taken together with the wider regulatory landscape, altered those conditions in ways that merit popular pushback rooted in the language of constitutionalism.