Navigating the AI Regulatory Landscape

India is moving toward establishing a dedicated framework for Artificial Intelligence, Union Minister Ashwini Vaishnaw indicated during a recent high-profile policy summit. The announcement comes at a critical juncture as AI rapidly permeates critical sectors including financial services, healthcare, education, commerce, media, and public administration.

Currently, the legal architecture governing artificial intelligence in the country remains dispersed across multiple statutes. India has yet to adopt a formal statutory definition of artificial intelligence. Instead, existing oversight relies on a patchwork of instruments, such as the Digital Personal Data Protection Act, 2023, the Information Technology Rules, 2021, and the Copyright Act, 1957, depending on the nature of the digital harm or activity involved.

Legal experts suggest that this fragmented approach should not be dismissed as a regulatory failure. Rather, it serves as a pragmatic starting point for designing the next phase of national AI governance.

The Case for Sector-Specific Rules

Unlike traditional single-function technologies, artificial intelligence presents diverse legal and social risks heavily dictated by its specific deployment context. An algorithmic system utilized in healthcare raises entirely different safety concerns compared to one deployed in digital advertising or automated recruitment.

Given these profound variations, policymakers are increasingly leaning away from a rigid, omnibus AI statute. Instead, the consensus points toward a framework focused on regulating specific uses, identifiable risks, and tangible harms across distinct sectors, underpinned by a unified set of core national principles.

Placing Liability at the Core

As deployment expands, accountability remains the most consequential legal hurdle. Because an AI system cannot legally be treated as a juridical person capable of bearing liability, responsibility must rest within the human and institutional ecosystem surrounding the technology.

Depending on the circumstances, liability chains will involve developers, deployers, operators, and institutional users. However, legal analysts emphasize that liability allocations cannot be uniform. A medical misdiagnosis carries life-threatening consequences, whereas financial algorithmic errors cause purely economic damage. Regulatory responses must therefore remain strictly proportionate to the risk profile and severity of potential failure within each unique industry.