AI and the Constitution: Building Digital Guardrails for Indian Democracy
As artificial intelligence reshapes the information ecosystem, the debate shifts from mere regulation to a constitutional imperative to protect democratic values. We explain the threats, the legal vacuum, and the proposed frameworks.
The Groundwork: What you need to know first
Before delving into the interplay between artificial intelligence and India’s constitutional fabric, it is essential to understand the foundational concepts, key legal milestones, and the institutions shaping the discourse.
(1) KEY TERMS
- Deepfake: AI-generated synthetic media where a person in an existing image or video is replaced with someone else's likeness, used for creating convincing but false content to mislead or defame.
- Algorithmic Amplification: The process by which platform algorithms prioritise content that generates high user engagement (such as outrage or sensationalism), leading to the rapid dissemination of misinformation and polarising narratives.
- Safe Harbour Immunity: A legal principle, enshrined in Section 79 of the Information Technology Act, 2000, that protects online intermediaries from liability for content posted by third-party users, provided they adhere to due diligence requirements.
(2) BACKGROUND & TIMELINE
The governance of technology in India has evolved significantly, often struggling to keep pace with innovation.
- 2000: The Information Technology (IT) Act is enacted, providing India's foundational legal framework for e-commerce and cybercrime. Its provisions, including Section 79, were designed for a much simpler internet.
- August 2017: In Justice K.S. Puttaswamy (Retd.) vs Union of India, a nine-judge Supreme Court bench unanimously declares the Right to Privacy a fundamental right under Article 21 of the Constitution, altering the landscape for data governance.
- February 2021: The government notifies the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, imposing more stringent due diligence and content moderation obligations on social media companies.
- May 2024: The European Union formally adopts its Artificial Intelligence Act, the world's first comprehensive, risk-based legislation for AI, setting a global precedent.
- Late 2025: The United Kingdom's Online Safety Act, 2023, is expected to become fully operational, introducing a 'duty of care' for platforms to protect users from harmful content.
- 2026-2027: The Government of India is expected to introduce a draft of the new Digital India Act for public consultation, aiming to create a comprehensive legislative framework to replace the IT Act, 2000 and address challenges posed by AI.
(3) INSTITUTIONAL FRAMEWORK
- Ministry of Electronics and Information Technology (MeitY): The nodal central ministry responsible for formulating and implementing policies on IT, electronics, and the internet. MeitY is leading the government's efforts to draft new regulations for the digital ecosystem.
- Parliamentary Standing Committee on Communications and Information Technology: A committee of MPs that scrutinises MeitY's policies and functioning. It plays a key role in examining bills and providing legislative recommendations on technology governance. Its analysis, including under former chairpersons like Shashi Tharoor, often informs the public debate.
The rapid proliferation of Artificial Intelligence (AI) has moved the conversation about its governance from technical labs to the heart of constitutional discourse. The central question is no longer just how to regulate AI, but why it demands a framework rooted in India's fundamental rights.
Why is AI governance a constitutional issue?
Initially viewed as a matter of technical regulation, AI governance is now framed as a constitutional imperative. The argument, articulated in analyses such as a recent piece by Member of Parliament Shashi Tharoor for The Hindu, is that AI's core capabilities directly impact the foundations of a democratic republic. These technologies can distort reality, manipulate public opinion, and erode the shared factual basis upon which democratic debate and informed consent depend.
This elevates the issue beyond ordinary legislation. When AI-driven systems influence voter choice or deepen social cleavages, they touch upon fundamental rights. The Right to Freedom of Speech and Expression under Article 19(1)(a) and the Right to Life and Personal Liberty under Article 21 are directly implicated. The Supreme Court's interpretation of Article 21 includes the right to live with dignity, which legal scholars argue, in the digital age, includes the right to an unmanipulated information environment.
What are the primary threats AI poses to Indian democracy?
Analysts identify three interconnected threats that AI systems, particularly on large digital platforms, pose to the democratic process. First is the proliferation of AI-generated disinformation. Sophisticated synthetic media, or 'deepfakes', can be deployed during electoral cycles to fabricate scandals or spread misinformation, directly undermining electoral integrity. Second is the challenge of algorithmic polarisation. The business models of many technology platforms, as extensive research has shown, rely on maximising user engagement, which often leads to the algorithmic amplification of divisive and sensationalist content, trapping users in echo chambers. This is often described not as an accidental flaw but as a structural feature of the current digital economy. Finally, these vulnerabilities can be weaponised for foreign information warfare. Adversarial actors can conduct AI-driven psychological operations to exploit and deepen pre-existing societal fault lines, posing a threat to national stability and sovereignty.
How do existing laws fall short in addressing these threats?
India's current legal framework is considered by many experts to be inadequate for the systemic challenges of AI. The primary legislation, the Information Technology Act of 2000, predates modern social media and AI. While amendments and new regulations like the IT Rules, 2021, have been introduced, critics argue they remain reactive, focusing on content takedown rather than systemic design.
The core issue, according to these critiques, lies in the 'safe harbour' protection under Section 79 of the IT Act. This provision largely shields platforms from liability for third-party content, which is said to disincentivise platforms from altering the algorithms that profit from amplifying harmful material. In contrast, other democracies are moving towards more structural approaches. The European Union's AI Act (2024) establishes a risk-based framework, while the United Kingdom's Online Safety Act (2023) places a 'duty of care' on platforms to proactively manage risks. India's framework currently lacks a comparable, forward-looking approach to algorithmic accountability.
What are the key pillars of a proposed rights-based framework?
To address these shortcomings, a new governance model anchored in constitutional principles has been proposed. Drawing from analyses published in The Hindu, this framework rests on several concurrent pillars. It begins with a rights-based legal foundation that prioritises individual dignity and digital autonomy, building on the principles of the Puttaswamy (2017) judgment to ensure citizens have rights over their data and are protected from algorithmic discrimination. This is coupled with a push for systemic platform accountability, moving beyond the current safe harbour regime to compel large technology firms to conduct independent audits of their algorithms and accept liability for harms, such as violence, resulting from the amplification of dangerous content. To avoid censorship, such regulation would focus on the mechanisms of disinformation, like bot networks, rather than policing ideological speech. This technical and legal scaffolding must be supported by two broader state-led initiatives: a nationwide digital literacy mission to build cognitive resilience against manipulation, and the development of sovereign defence capabilities to detect and neutralise coordinated foreign information warfare campaigns.
The Way Forward: From Regulation to Constitutional Resilience
Why does this matter right now? The urgency stems from a mismatch in speed: AI technology is evolving exponentially, while democratic legislative processes are deliberative. With India's dynamic electoral cycles and pre-existing societal fault lines, the risk of AI-driven manipulation causing real-world harm is immediate. The window to establish effective, democratically legitimate guardrails is narrowing.
What is the likely trajectory? The Government of India is actively developing the Digital India Act to replace the IT Act, 2000. A draft is anticipated for public consultation by 2026 or 2027. The ensuing debate will pivot on a critical choice: adopt a 'light-touch' regulation to prioritise innovation, or implement a comprehensive, rights-based framework that embeds constitutional values into the digital architecture. The final legislation, likely to be enacted by 2028, will shape India's digital future for a generation.
What are the governance implications? A well-crafted framework could position India as a leader in responsible AI governance. Conversely, a failure to act decisively could deepen social polarisation and leave the world's largest democracy vulnerable to manipulation. The challenge is not merely technical but foundational—it is about ensuring that the architecture of the digital public square is compatible with the constitutional principles of the republic.