‘First published on India Business Law Journal’
By: Pravin Anand, Dr. Ajai Garg and Alvin Antony
The ascendancy of large language models (LLMs) has irrevocably altered the technological and geopolitical contours of the global order. For India, this shift constitutes not merely a matter of digital competitiveness but a sovereign imperative, which has become a serious concern in the era of digital revolution.
To remain subject to foreign AI infrastructure and datasets is to cede ground in domains of national interest, economic independence and cultural continuity. In this context, India’s pursuit of sovereign LLMs emerges as an extension of its entrenched policy frameworks such as “Make in India” and “Atmanirbhar Bharat (Self-reliant India)”, signalling an ambition not just to participate in the AI race, but to shape its future trajectory.
India’s linguistic diversity renders Western-developed LLMs structurally discordant with domestic realities. The imperative is the development of sovereign, Indic-trained models grounded in vernacular corpora to enable genuine inclusivity and mitigate algorithmic bias.
The indiscriminate deployment of foreign LLMs simultaneously raises concerns under India’s Digital Personal Data Protection Act, 2023 (DPDP Act), exposing sensitive personal data to extraterritorial jurisdictions and undermining informational sovereignty. A culturally attuned AI paradigm is essential not merely for narrative preservation or equitable access but as a bulwark against epistemic erasure, systemic distortion and strategic vulnerabilities, particularly in defence and finance.
India’s transformation from a passive digital consumer to an autonomous architect of foundational AI infrastructure marks a pivotal moment in its techno-regulatory evolution. The IndiaAI Mission, with a proposed principal investment of USD1.3 billion, aims to catalyse this shift by building national compute capacity, especially GPU-rich infrastructure.
However, relative to global benchmarks, this quantum remains insufficient. Mobilising domestic capital and crafting targeted incentive regimes are critical. India’s digital public infrastructure (DPI) offers a unique substrate for indigenous LLM training, but must be leveraged within the strict guardrails of consent, purpose limitation and data minimisation, as enshrined in the DPDP Act. The syntactic and semantic complexity of Indic languages necessitates bespoke algorithmic models tailored to India’s linguistic plurality. As India envisions the development of an LLM, it is imperative to address multifaceted legal and regulatory challenges encompassing security, privacy, legitimacy, inclusiveness, transparency, fairness, accountability, reliability and explainability.
The current landscape is fraught with complexities, as evidenced by ongoing global litigations. In Mobley v Workday, Inc, the court acknowledged the potential liability of AI vendors under the “agent” theory, highlighting the risks of algorithmic bias in employment decisions. Similarly, Cousart v OpenAI LLP and Garcia v Character Technologies underscore concerns regarding data privacy and unauthorised use of personal information in AI training processes. These cases illustrate the necessity for India to establish robust legal frameworks that pre-emptively address such issues.
Regulatory clarity remains a cornerstone for fostering innovation and attracting capital. India’s current “pro-innovation” stance, grounded in legacy frameworks such as the IT Act, 2000, and the DPDP Act, and informed by the guidance of the National Institution for Transforming India (NITI Aayog), is yet to crystallise into a cogent and predictable legal framework. Regulatory opacity, especially in an emerging technology landscape, acts as a formidable deterrent to investment. India’s sovereign LLM project is not an isolated technological endeavour but a confluence of economic strategy, cultural stewardship, data governance and national security. The path is fraught with challenges, be it compute scarcity, capital inadequacy or legal indeterminacy, but India is uniquely positioned to meet them. With its robust DPI, vast linguistic corpus, growing pool of AI talent and unambiguous governmental resolve, India can architect a hybrid model that melds sovereign innovation with global collaboration.
India must choose between using foreign AI and leading responsible AI development for the Global South. If executed with vision and regulatory foresight, India’s sovereign LLM initiative may well become the cornerstone of its digital constitutionalism.
etc. whilst wrongfully claiming to be part of our firm and making false claims and allegations.