India produces approximately 1.5 million engineering graduates a year. The United States — the country leading the global AI race — cannot find enough people who can build AI systems.

These two facts sit in the same world. They have not yet produced a coherent strategy from either government.

The Scale of the Mismatch

Let's be specific about what "the US cannot find enough AI talent" actually means.

It means that the major AI laboratories — OpenAI, Anthropic, Google DeepMind, Meta AI — are competing ferociously for a tiny pool of researchers and engineers who can work at the frontier. It means that the next tier of companies — the enterprises trying to integrate AI into real business workflows — are desperately short of the engineers who can implement and maintain these systems. It means that AI-adjacent fields like data infrastructure, MLOps, and AI safety are chronically understaffed.

The H-1B visa programme — the primary channel through which Indian engineers enter the US technology workforce — is a political football that gets kicked around every election cycle. It is capped at 85,000 visas a year for a country of 330 million people trying to maintain its position at the frontier of the most consequential technology wave in a generation.

India sends more H-1B recipients to the US than any other country. The demand is there. The system is a bottleneck.

Why This Is a Geopolitical Asset India Is Leaving on the Table

The flow of Indian engineering talent to Silicon Valley is often framed in India as a brain drain problem. That framing misses something important.

The Indian-American technology community is one of the most concentrated pools of economic and technological influence in the world. Sundar Pichai runs Google. Satya Nadella runs Microsoft. The CEOs of Adobe, IBM, Palo Alto Networks, and dozens of other major technology companies are of Indian origin. The diaspora's contribution to US venture capital, AI research, and startup creation is disproportionate to its demographic size.

This is not brain drain. This is a network — a web of relationships, capital, and capability that sits at the intersection of the two largest democracies on earth. It is an asset of extraordinary potential strategic value.

And neither the Indian nor the American government has a serious strategy for leveraging it.

What a Real Strategy Would Look Like

From the Indian side: stop treating the diaspora as a source of remittances and NRI deposits and start treating it as a strategic intelligence network. Create formal mechanisms for diaspora expertise to flow back into Indian institution-building — in AI research, in regulatory design, in education system reform.

From the American side: modernise the immigration system for high-skill workers as a matter of national security rather than immigration politics. The US is in a technology race with China that it cannot afford to lose. Restricting access to the world's best engineers because of a political debate about immigration is strategically self-defeating.

From both sides: build the India-US technology corridor as a formal pillar of the bilateral relationship — not just a nice talking point at state visits, but an operational programme with budgets, timelines, and accountability.

The Trillion-Dollar Number

The number is illustrative rather than precise, but it is directionally correct.

The value of deepening the India-US AI collaboration — through freer talent mobility, joint research programmes, co-investment in AI infrastructure, and shared regulatory frameworks — runs into the trillions over a decade. Not because of any single deal, but because of the compounding effects of having the world's largest democracy and the world's oldest democracy building the AI future together, rather than each operating sub-optimally in isolation.

The gap is visible. The opportunity is real. The absence of anyone filling it is, at this point, a strategic choice — even if nobody made it consciously.

Someone should make it consciously. In the other direction.