Introduction
In a world where AI is being increasingly embraced in almost every critical sector-from commercial industries and strategic domains to agriculture, climate change, education, healthcare and governance- it categorically becomes prudent to argue that this ‘new fuel of the digital world’ is predominantly shaping the geopolitics of the twenty-first century. With its advanced algorithms, machine learning, and high-performance computing capacity, AI is seen as thriving as a defining trademark of the data-driven, digitalised world.
As a leading power and the fastest-growing economy of the world, India is making significant strides towards blooming and boosting its AI ecosystem through trailblazing initiatives like- ‘National AI Strategy’ and the application-driven ‘India AI Mission’. With a robust seven-pillar framework, the ‘India AI Mission’ is designed to offer culturally relevant and globally competitive AI solutions that are meaningfully aligned with the long-term goal of inclusive growth. As per the Stanford AI Index 2024[1], India leads globally in AI skill penetration rate with a score of 2.8, outshining the United States and Germany with scores of 2.2 and 1.9, respectively. The index also reports that India secures the highest AI skill penetration rate for women with a score of 1.7, relatively surpassing the United States and Israel with respective scores of 1.2 and 0.9. This impressive achievement indicates India’s concrete yet continuous efforts to address the gender-based discrepancies in AI skill development.
In the wake of these inclusive and strategic cost-effective measures, the country’s vision is primarily centered towards democratising and indigenising AI through open-source development and innovation, along with promoting constructive collaboration within the global AI community for the efficient realisation of the sustainable development goals.
Advancements and Key Priorities
In a major leap towards the ongoing momentum, the ‘Bhashini initiative’ launched by MeitY impregnates tremendous potential for enhancing the “open-source frameworks to build natural language processing models that support 22 official Indian languages and numerous dialects”.[2] This demonstrates India’s sheer commitment to bridging the language and literacy-based digital divides by disseminating AI-powered innovative multilingual solutions on a large scale. With a greater conviction, India believes that by democratising access to digital services and information in the mother tongue or local dialects of the natives, it is considerably scaling up the productivity of the hundreds of millions of people who are lagging at low incomes. Thus, it is perceptible that a rational and equitable use of AI can remarkably upgrade the country’s economy and eventually strengthen its defining position in the global AI landscape.
Steering up further, the recent launch of ‘AI Kosha’ by the Union government (in March 2025) as a dynamic component of the ‘India AI Datasets Platform’- one of the prominent pillars of the ‘India AI Mission’, is noteworthy. The ‘AI Kosha’ platform reportedly contains more than three hundred datasets with over fourteen thousand GPUs[3] (Graphics Processing Units), most of which are being deliberately programmed to create language translation tools and models for a host of Indian languages. With an unrelenting quest to integrate Indian ethos to the emerging ‘tech’ sphere and prioritise the home-grown intelligence, the government has chosen a Bengaluru-based tech startup, ‘Sarvam AI’ has been entrusted with the development of India’s first sovereign Large Language Model (LLM) under the ₹10,372 crore India AI Mission. Highlighting its significance, the Union IT Minister stated that this is India’s opportunity to lead rather than follow. He emphasised that the initiative goes beyond technology—it is a nation-building effort that must embody Indian languages, values, and address the country’s unique challenges.
India’s Global Engagement and Multi-stakeholder Model
While certain challenges like an incoherent ecosystem, infrastructural divides and skill-based inadequacies are yet to be seen, India is reasonably crafting a ‘middle path’ or a ‘third way’ to deal with the expanding ‘tech duopoly’ of China and the United States. The ‘Delhi AI Development Framework’ is inexorably driven by the principles of affordability, scalability, ethicality, sustainability and ubiquity, as against propelling it mainly for the economic predominance or using it largely as a security apparatus.
Being a founding member of the Global Partnership on Artificial Intelligence (GPAI), India proficiently used its presidency (in July 2024) to underline key priorities such as- democratising access to a broad range of AI expertise and skills; addressing the widening digital divides and societal inequities between and within the countries; advancing the application of AI in critical spheres viz. agriculture, education and healthcare; fostering a safe, secure, responsible and human-centric AI development based on coordinated international efforts and inclusive multi-stakeholder approach.
The Global India AI Summit hosted in New Delhi from 3-4 July 2024, was notably attended by around two thousand people, including global AI experts, policy makers, AI practitioners, and people from industries, startups and academia. Over ten thousand people clubbed as ‘AI enthusiasts’ had reportedly joined the sessions of the summit virtually. One of the salient highlights of the summit was the convergence of GPAI and the India AI Mission, which eventually led to the deliberation of some insightful observations over the implementation aspects of the ambitious mission. The major points of discussion decisively reaffirmed the pertinence of multi-large language models for meeting the diverse requirements of India; “platformisation and standardisation of AI-ready data”[4]; and a collaborative multistakeholder ecosystem altogether blending the perspectives from wide-ranging segments such as technology, policy, research, industry, startup, business and academia.
Ethical Frameworks and Safe AI
While recognising the multifaceted transformative potential of AI, the summit also drew attention towards the requirement of “robust governance frameworks” to deal with the emerging safety as well as transparency-based challenges and ensure an adequate balance between the innovation and regulation mechanisms. Concerning this, the ‘Safe and Trusted AI’ pillar under the India AI Mission is already leveraging efforts to develop trusted guardrails, innovators’ innovator-centric self-assessment checklists, and the comprehensive governance frameworks for promoting ethical, responsible and transparent AI practices. That is why Mr. Timothy Curtis, Director of the UNESCO Office in New Delhi, emphasized during the summit that India’s developments in AI hold global significance. He noted that India’s population size, economic momentum, and innovative capacity position it to shape international trends in artificial intelligence.
With an objective to comprehensively analyse the prevalent gaps in the process of AI governance and to offer substantial recommendations, a sub-committee on ‘AI Governance and Guidelines Development’ was established by the MeitY in November 2023. In its detailed report, the sub-committee has recommended adopting a “digital by design” and “whole-of-government” approach by enhancing a “systems-level understanding of India’s AI ecosystem” and creating an empowered “Inter-Ministerial AI Coordination Committee or Governance Group”[5] To dynamically bring together several authorities and institutions dealimng with AI governance at the national level.
Concerns and Imperatives
Although the sub-committee on AI governance envisions fostering a systemic risk-based approach driven by a mix of centralised and decentralised frameworks, there is a considerable lack of quantified empirical data delineating aptly classified AI-related risks in the Indian context. Hence, it is imperative to develop a more nuanced and deeper understanding of the technical functionalities, limitations, risks and potentialities of the AI systems. In reference to this, some key adaptations can be construed from the EU AI Act- the first ever comprehensive legislation to address various risk aspects conjoined with the development and deployment of AI.
It is also to be noted that a well-balanced techno-regulatory governance approach should essentially be guided by an amalgamation of self-regulation, binding normative commitments, transparency obligations and targeted interventions. Liability mechanisms pertaining to the safeguards and quality controls must be clearly and broadly defined, especially for the cognisance and compliance of the foundational actors, viz. AI developers and deployers. Alongside this, an exuberant discussion about the ‘AI liability and sustainability framework’ needs to be promptly expanded within the global AI governance community for the sake of creating a level-playing field and protecting vulnerable communities (mostly belonging to the Global South) from being disproportionately affected.
Challenges and the Way Forward
As a leading proponent of ‘tech democracy’ through its globally recognised and widely shared DPI (Digital Public Infrastructure) architecture, India firmly calls for enhancing the equitable access of AI, especially for the Global South countries and concomitantly promotes it for the cause of ‘global good’. In the light of its ethical, interoperable, innovative and risk-managing tech drive, India is unremittingly positioning itself as a key navigator of the global AI era. Undoubtedly, the holistic AI development model of Bharat under the highly acclaimed ‘India AI Mission’ brand will be one of the crucial indicators for attaining the ambitious vision of ‘Viksit Bharat 2047’.
References
- Artificial Intelligence Index Report 2024, Stanford University Human-Centered Artificial Intelligence, https://hai.stanford.edu/assets/files/hai_ai-index-report-2024-smaller2.pdf
- Elbashir M. and K.B. Desikachari (2025), “India’s path to AI autonomy”, Atlantic Council, https://www.atlanticcouncil.org/in-depth-research-reports/issue-brief/indias-path-to-ai-autonomy/#:~:text=India%20is%20making%20significant%20progress,a%20diverse%20country%20like%20India.
- The Hindu (2025), “Government launches data platform for Indian AI models, beefs up shared GPU capacity”, https://www.thehindu.com/news/national/government-launches-data-platform-for-indian-ai-models-beefs-up-shared-gpu-capacity/article69298801.ece
- Ministry of Electronics and Information Technology (MeitY), Government of India, “Global India AI Summit Report”, July 3-4, 2024, Bharat Mandapam, New Delhi, https://gpai.ai/gpai-india-ai-summit-report-2024.pdf
- Ministry of Electronics and Information Technology (MeitY), Government of India, “Report on AI Governance Guidelines Development for Public Consultation”, January 2025, https://indiaai.s3.ap-south-1.amazonaws.com/docs/subcommittee-report-dec26.pdf