AI and India
India's relationship with AI is different from the relationship that Silicon Valley assumes. The constraints are different, the opportunities are different, and the version of AI that matters most for India is not the version that dominates the global conversation.
Founder, Majhi Group & Majhi OS
In 2023, I started building Majhi OS from Odisha. The ambient assumption in the international AI conversation was that meaningful AI development happened in San Francisco or London, and that everywhere else was a consumer or laggard. Building from Odisha put me on the outside of that assumption, which turned out to be a useful place to be.
The version of AI that matters most for India's 1.4 billion people is not the version that gets covered in Western technology media.
Where India starts from
India has a large, young, and growing population with a median age of around 28. It has one of the world's largest concentrations of engineering talent, produced by IITs, NITs, and hundreds of other technical institutions, employed in the global technology industry, and increasingly building companies. It has a rapidly expanding digital infrastructure, including UPI (one of the world's most successful digital payment systems), Aadhaar (a biometric identity system at national scale), and a government that has been more actively interventionist in digital infrastructure than most comparable economies.
It also has significant structural challenges: a public education system that is large and uneven in quality, a healthcare system that is under-resourced relative to the population it serves, a large informal economy where economic activity is not well captured by formal systems, and deep regional disparities in economic development.
The AI applications that are most relevant to India are determined by these starting conditions, not by the starting conditions of Silicon Valley.
Where AI actually matters in India
Healthcare at scale. India's doctor-to-population ratio is significantly below the WHO recommended level. The gap between the healthcare needs of 1.4 billion people and the supply of trained healthcare providers is a structural problem that cannot be solved at the pace required by training more doctors alone. AI tools that support diagnosis, that help healthcare workers at lower training levels extend their effective scope, that make specialist knowledge accessible at the primary care level, and that help manage chronic disease in populations that have limited access to regular specialist care are directly relevant to this gap.
Agriculture and rural livelihoods. India has roughly 600 million people whose livelihoods are connected to agriculture. AI tools that provide better crop disease identification, more accurate yield prediction, more useful market price information, and more accessible advisory services for farmers can improve the economic outcomes of a population that is otherwise poorly served by information-based services.
Education access and quality. India's education system produces millions of graduates annually, but the quality is highly variable. AI tutoring tools that provide expert-level explanation and feedback to students in schools that cannot currently provide it are particularly relevant to a country where the gap between the best and worst educational environments is large and consequential.
Language and inclusion. India has 22 officially recognized languages and hundreds of others with significant speaker populations. Most of the best AI tools are built primarily for English. The extension of capable AI tools to Hindi, Telugu, Tamil, Marathi, Bengali, and the other major languages of India is an access question of significant consequence, the version of AI that reaches the majority of India's population will need to work well in languages that currently have far less AI capability than English.
Infrastructure and urban management. India is in the middle of one of the largest urban expansions in human history. AI tools that support better urban planning, better traffic management, better infrastructure maintenance, and better delivery of government services at city scale have direct applications in a country building new cities and managing the rapid growth of existing ones.
India's AI production, not just consumption
India is not only a consumer of AI, it is a producer. Indian engineers are significant contributors to the global AI research and engineering effort, employed at the major AI labs and building AI companies. The Indian AI startup ecosystem is growing, with companies building applications across healthcare, agriculture, fintech, and enterprise software.
The most interesting question for India is not whether it can produce world-class AI researchers and engineers, it demonstrably can. The question is whether India builds the frontier models and the infrastructure layer, or primarily builds applications on top of models built elsewhere. This is a question about capital, about data access, and about the strategic choices that Indian institutions and companies make over the next decade.
The distinctive India AI opportunity
The opportunity that is most distinctive to India is the opportunity to build AI applications at scale for populations that are currently underserved by the global technology industry. The majority of the world's population does not speak English as a first language, does not have reliable internet access, does not have smartphones with the compute to run demanding applications, and does not have the economic profile that most Silicon Valley product decisions are optimised for.
Building AI that works for these populations, that is multilingual, that runs on lower-end hardware, that is designed for intermittent connectivity, that addresses problems relevant to agriculture and healthcare and education in lower-income contexts, is not a charity project. It is a large market opportunity that the global AI conversation consistently underestimates.
India is one of the few places where the talent to build this AI, the populations who would use it, and the institutional will to invest in it are all present at the same time. What gets built with that combination, over the next decade, is genuinely open.
India is one of the few places where the talent to build transformative AI, the populations who need it most, and the institutional will to invest are all present at the same time. What gets built with that is genuinely open.
See also: AI and Human Potential, The Global Hiring Floor, AI and Education
Sources
NASSCOM: India IT industry FY25 performance overview
IBEF: India information technology industry overview
WEF Future of Jobs Report 2025: AI displacement and creation projections
Frequently Asked Questions
Why is India's relationship with AI different from Silicon Valley's?
Because the constraints and the opportunities are different. Silicon Valley AI development is optimised for English-speaking users with high-end hardware, reliable broadband, and problems that matter to the top decile of the global income distribution. India's most important AI opportunity is building for 1.4 billion people, the majority of whom use languages other than English, have intermittent connectivity, use lower-end hardware, and face problems in healthcare, agriculture, and education that the global AI conversation largely ignores. The AI that matters most for India is not the AI that dominates the global product conversation. It is the AI built for the structural gaps that the global market does not serve.
What are the most important AI applications for India's actual development challenges?
Healthcare at scale (India's doctor-to-population ratio is significantly below WHO recommendations, AI that extends the effective scope of healthcare workers is directly relevant), agriculture (roughly 600 million livelihoods depend on it, and AI that provides better crop disease identification, yield prediction, and market price information can improve outcomes for this population), and language (most capable AI tools are built primarily for English, extending them to Hindi, Tamil, Telugu, Bengali, and the other major Indian languages is an access question of significant consequence). These three are the places where AI can produce the highest development return, relative to the scale of the need.
Is India positioned to build AI or primarily to consume it?
India has the talent base, Indian engineers are significant contributors to the global AI research and engineering effort, employed at the major AI labs and building AI companies. The question is whether India builds at the frontier or primarily at the application layer. Frontier model development requires capital at a scale that the US and China have deployed and India currently has not matched. The application layer opportunity is real and large: building AI that works for India's specific constraints, in India's languages, for India's economic and social profile, is a market the global players are not optimised to serve. India's distinctive AI opportunity is in applications, not in frontier model competition, and that is a large enough opportunity on its own terms.
Did this land? Push back? Add something I missed?
Reply to Manas →Continue Reading
Related writing
Future of WorkThe Global Hiring Floor
AI is not just changing who does the work. It is changing what the work is worth. As AI tools raise the output floor for every knowledge worker, the question of what human talent commands, and why, is being rewritten in real time.
Future of WorkAI and Human Potential
AI will either be the most powerful opportunity-expanding technology in history or the most efficient mechanism for concentrating advantage ever built. Which one it becomes depends on decisions being made right now by people who are not asking that question.
Future of WorkLiving Alongside AI
Most productivity advice about AI tells you what to outsource to it. That question matters, but it misses the harder one: what do you keep? The answer to that question will determine what kind of person you become over the next ten years.