Testing the Constitutional Limits of India’s Accelerated Content Takedown Regime under the IT Rules, 2026
Author: Siddharth Singh Chaudhary
College: Babu Banarasi Das University
To the Point
On 10 February 2026, the Ministry of Electronics and Information Technology notified the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026, which came into force on 20 February 2026. The most consequential change introduced by this amendment is the compression of the content-removal timeline available to online intermediaries from thirty-six hours to a mere three hours, applicable to specified categories of unlawful material and to a newly defined class of content termed “Synthetically Generated Information,” which covers deepfakes and other AI-manipulated media.
A platform that fails to act within this shortened window, or that fails to label synthetic content as required, forfeits the safe-harbour immunity ordinarily available to it under Section 79 of the Information Technology Act, 2000. Once that immunity lapses, the intermediary can be proceeded against as though it were the original publisher of the unlawful material. The rule is a direct legislative response to the rapid weaponisation of generative artificial intelligence against women, senior citizens, and the integrity of democratic discourse. Yet the very features that make the rule effective as an enforcement tool — its speed, its reliance on automated compliance, and its dependence on delegated rule-making rather than primary legislation — are precisely what expose it to challenge under Articles 14, 19(1)(a), and 21 of the Constitution of India.
Use of Legal Jargon
A meaningful assessment of the three-hour rule requires familiarity with the following terms of art, each of which recurs through the analysis below:
Safe Harbour (Section 79, IT Act, 2000)
A statutory immunity that shields an intermediary from liability for third-party content hosted on its platform, provided it observes the due-diligence obligations prescribed by law and acts as a neutral passive channel for third-party data, so long as it does not itself originate, edit, or select the unlawful material passing through it
Intermediary and Significant Social Media Intermediary (SSMI)
An “intermediary” is any entity that receives, stores, or transmits electronic records on behalf of another person. An SSMI is a subset crossing a prescribed user threshold, attracting heightened obligations such as grievance redressal officers, traceability, and now, the three-hour compliance window.
Synthetically Generated Information (SGI)
A newly codified category referring to any audio, visual, or audio-visual content that has been created or algorithmically altered to appear authentic or indistinguishable from a real person or event — the statutory vocabulary for what is colloquially termed a “deepfake.”
Doctrine of Proportionality
A constitutional test requiring that any restriction on a fundamental right pursue a legitimate aim through a suitable, necessary, and least-restrictive means, while maintaining a rational balance between the right infringed and the objective sought.
Chilling Effect
The self-censorship that speakers or platforms adopt out of fear of penal or civil consequence, even where their expression would ultimately be found lawful — a harm the Supreme Court has recognised as cognisable under Article 19(1)(a) even absent an actual prosecution.
Overbreadth and Vagueness
Grounds on which a statute or subordinate legislation may be struck down where its language sweeps within its ambit both proscribable and constitutionally protected speech, or fails to give an ordinary citizen or intermediary fair notice of what is prohibited.
Delegated / Subordinate Legislation and Ultra Vires
Rules framed by the executive under powers conferred by a parent statute. The validity of delegated legislation depends upon its conformity with the enabling statute and the Constitution. Any rule that exceeds the authority conferred by the parent legislation or infringes fundamental rights may be struck down as legally invalid.
Audi Alteram Partem
A principle of natural justice requiring that no person be condemned unheard; relevant where a takedown order results in the removal of content, or the compromise of a platform’s safe. immunity, even though the affected user or platform was given no advance notice or real-time chance to present their side
The Proof
The factual and regulatory record establishes the following:
• Notification and commencement: The Amendment Rules were notified on 10 February 2026 and took effect ten days later, on 20 February 2026, leaving intermediaries a narrow compliance runway.
• Timeline compression: The takedown window for specified unlawful content and SGI was reduced from thirty-six hours — the standard set under the parent IT Rules, 2021 — to three hours, an approximately twelve-fold acceleration.
• Definitional expansion: The Rules introduce, for the first time in Indian subordinate legislation, a working definition of Synthetically Generated Information, bringing AI-manipulated audio, video, and images within a distinct compliance category.
• Consequence of non-compliance: Failure to meet the three-hour deadline, or to affix the mandated label to synthetic content, results in forfeiture of Section 79 safe harbour, exposing the platform to liability as though it were the content’s originator.
• Economic impact: Industry estimates cited in professional commentary suggest content-moderation costs for major platforms could rise by roughly a quarter within the 2026 fiscal year, with additional capital expenditure anticipated on regional data infrastructure to meet the compressed response times.
• Industry and civil-society reaction: Trade associations representing global technology companies have sought a phased rollout or suspension of the three-hour regime, while digital-rights groups have cautioned that automated moderation systems triggered by such tight deadlines risk misclassifying legitimate speech — including satire, journalism, and commentary — as unlawful synthetic content.
• Anticipated litigation: Legal commentary through mid-2026 anticipates that the constitutional validity of the amendment will be tested before multiple High Courts, with the Supreme Court’s existing jurisprudence on intermediary liability, privacy, and proportionality expected to furnish the operative constitutional standard.
Abstract
This article examines the constitutional sustainability of the three-hour content-takedown mandate introduced by the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026. It situates the rule within India’s broader intermediary-liability architecture under Section 79 of the Information Technology Act, 2000, and measures it against the constitutional benchmarks laid down by the Supreme Court in the Shreya Singhal, Puttaswamy, and AnuradhaBhasin rulings . The article argues that while the State’s objective — curbing the circulation of deepfakes and other synthetically generated harms — is legitimate and pressing, the compressed compliance window, the absence of a pre-decisional hearing mechanism, and the delegation of speech-restrictive power to executive rule-making without adequate procedural safeguards create a substantial risk of disproportionate and chilling regulation. The article concludes by proposing a calibrated framework that preserves rapid response capability for genuinely dangerous synthetic content while restoring proportionality safeguards for borderline cases.
Case Laws
1. Shreya Singhal v. Union of India, (2015) 5 SCC 1
The judgment in Shreya Singhal occupies an important position in Indian jurisprudence concerning speech on the internet. The Supreme Court invalidated Section 66A of the Information Technology Act, 2000 because the provision used uncertain and expansive expressions that could permit unreasonable interference with freedom of speech. The Court also examined intermediary liability under Section 79 and explained the circumstances in which online platforms could be expected to remove unlawful material.
Its relevance to the present issue lies in the possibility of excessive content removal. When intermediaries are given an exceptionally short period to respond, they may prefer deleting disputed material rather than undertaking a detailed assessment of its legal status. This could particularly affect satire, criticism, political communication and other expression whose legality may depend upon context. The case therefore supports the need for safeguards that prevent intermediary regulation from becoming a source of arbitrary censorship.
2. Justice K.S. Puttaswamy (Retd.) v. Union of India, (2017) 10 SCC 1
In Puttaswamy, a nine-judge Constitution Bench recognised privacy as a fundamental right under Article 21 and affirmed the importance of proportionality when State action affects fundamental rights. In essence, constitutional restrictions must be directed towards a legitimate purpose and must not impose a burden that is disproportionate to that purpose.
This reasoning is relevant to the regulation of deepfakes and synthetic media. Preventing deception, impersonation, fraud, privacy violations and other serious harms is a valid regulatory objective. The difficulty lies in determining whether the same highly compressed response period is justified across every situation covered by the regulatory framework. Some synthetic material may be obviously unlawful, whereas other content may require factual investigation or contextual interpretation.
Accordingly, the constitutional assessment should consider not merely whether the government has a legitimate reason for regulating synthetic media, but also whether the chosen enforcement mechanism goes further than reasonably necessary. Adequate safeguards become particularly important where a rushed decision may result in lawful expression being removed.
3. Anuradha Bhasin v. Union of India, (2020) 3 SCC 637
The Supreme Court in Anuradha Bhasin considered restrictions on internet access in Jammu and Kashmir and examined the relationship between executive action and fundamental rights. The judgment reinforced the requirement that measures affecting constitutional freedoms must satisfy appropriate standards of proportionality and remain open to judicial scrutiny.
Although the case was not concerned with deepfakes or intermediary takedowns, its principles have significance for digital regulation. Restrictions affecting online material can influence how individuals communicate, obtain information and participate in public discourse. Consequently, the existence of an executive regulatory framework does not remove the need for constitutional safeguards.
The three-hour mechanism raises a related concern. Rapid intervention may be justified where unlawful content creates an immediate and serious danger, but urgency should not eliminate procedural fairness. Clear criteria, reasoned decisions and an effective mechanism for reconsidering mistaken removals can help maintain the necessary balance between prompt enforcement and protection of lawful speech.
4. Kunal Kamra v. Union of India, 2024:BHC-OS:14750-DB
The Bombay High Court’s decision in Kunal Kamra v. Union of India dealt with the 2023 amendment to the IT Rules concerning information related to the business of the Central Government and the proposed Fact Check Unit. The proceedings produced conflicting judicial opinions, after which Justice A.S. Chandurkar, acting as the third judge, considered the matter. The majority view ultimately resulted in the challenged provision being struck down.
The decision is significant for the present discussion because it demonstrates judicial scrutiny of regulatory provisions that may confer broad authority over online expression. It highlights the constitutional importance of precise standards and limits on executive discretion. A governmental objective, even when legitimate, cannot by itself justify a mechanism that creates an excessive possibility of interference with protected speech.
Although Kunal Kamra does not specifically address the legality of the three-hour requirement introduced in 2026, the principles emerging from the judgment could still be relevant in any future constitutional challenge to the new regulatory framework.
Conclusion
The 2026 IT Rules seek to address the growing risks associated with deepfakes and other forms of synthetic media. Such technologies can be exploited for fraud, impersonation, privacy violations, reputational attacks and other forms of harm. There is therefore a legitimate public interest in ensuring that unlawful synthetic content can be dealt with promptly.
The constitutional issue is primarily concerned with the method of enforcement rather than the objective itself. An extremely short period for taking action may encourage intermediaries to adopt a precautionary approach and remove material even where its legal status is uncertain. Such a tendency could adversely affect expression that is lawful but controversial, satirical or dependent upon context.
A more balanced regulatory model could therefore differentiate between various levels of harm. Content involving an immediate and clearly identifiable threat could be subject to the fastest intervention, while material requiring contextual or factual assessment could receive additional time for review. A reliable mechanism for challenging incorrect removals would further strengthen procedural protection.
The principles reflected in Shreya Singhal, Puttaswamy, Anuradha Bhasinand Kunal Kamra provide useful constitutional benchmarks for this assessment. The objective should be to regulate genuinely harmful synthetic content without allowing the urgency of technological regulation to weaken the fundamental protection accorded to lawful expression.
Frequently Asked Questions
Q1. Key Changes Brought by the 2026 Amendments ?
The amendments introduce additional obligations concerning synthetically generated information, including requirements relating to its identification and labelling. The amended provisions additionally impose more stringent timelines for responding to particular categories of content governed by the framework.
Q2. What are the consequences of non-compliance by an intermediary?
Failure to satisfy applicable due-diligence obligations may affect an intermediary’s entitlement to rely upon the safe-harbour protection under Section 79. However, it would be inaccurate to treat every failure to meet a particular deadline as automatically making the platform the publisher of the disputed material.
Q3. Which constitutional rights may be affected?
Article 19(1)(a) is central because removal of online material can interfere with freedom of speech and expression. Article 14 may become relevant where regulatory decisions are arbitrary or unreasonable, while Article 21 may arise in matters involving privacy and related personal interests.
Q4. Has a court already ruled on the constitutional validity of the three-hour requirement?
The authorities discussed in this article deal with broader questions concerning online speech, proportionality, executive action and intermediary regulation. They do not amount to a final judicial determination specifically approving or invalidating the three-hour requirement introduced by the 2026 Rules. Its constitutional status must therefore be assessed through existing precedent unless and until a court directly considers the provision.
Q5. Does the three-hour period cover every form of AI-generated content?
No. The 2026 framework creates different obligations for different categories of synthetically generated information. The applicable requirement depends upon the nature of the material and the particular provision under which it falls.
Q6. What reform is proposed?
The article favours a graded regulatory structure. The most stringent deadlines should apply to serious and clearly unlawful material, while content involving genuine factual or legal uncertainty should receive greater opportunity for review. Accessible challenge procedures and safeguards against erroneous removal should accompany the system.
