Betting big on AI, Ambanis and Tatas are pushing India towards new phase of development

By Kunal Roy Chowdhury

New Delhi. After two chaotic days at the AI Summit in New Delhi—marked by breathless panel discussions, competing visions of regulation, and the familiar spectacle of startup evangelism—India appeared to steady itself. The noise gave way to something more consequential. Two of the country’s most powerful conglomerates signaled that India’s artificial intelligence ambitions are entering a new phase, one defined less by rhetoric and more by capital expenditure.

For years, India’s AI narrative revolved around talent density, coding prowess, and the proliferation of startups chasing venture funding. The emphasis was on applications and services: who could build the next model, the next platform, the next productivity tool. But as global competition hardens and compute power becomes the real currency of influence, the axis is shifting. What was once framed as a race for engineers and entrepreneurs is now becoming a contest over infrastructure—over data centers, energy supply, semiconductor access, and the physical backbone that makes advanced AI possible.

The pivot suggests a growing recognition in New Delhi and Mumbai alike that without sovereign-scale compute capacity, India risks remaining a user of global AI systems rather than a shaper of them. The new announcements indicate that corporate India is prepared to move beyond incremental investments and toward the kind of heavy industrial commitments that historically underpin technological power.

The Tata Group—the salt-to-software behemoth that has long mirrored India’s own industrial evolution—announced a partnership with OpenAI to build artificial intelligence infrastructure in India that could scale to as much as one gigawatt over the coming years. In the global AI economy, where compute capacity is increasingly equated with strategic power, a gigawatt is not a metaphor. It is a declaration of industrial intent.

Hours later, at Reliance Industries, Mukesh Ambani unveiled an even more sweeping commitment: an investment of 10 lakh crore rupees aimed at building vast new AI data centers and associated digital ecosystems. For a conglomerate better known for petrochemicals and refineries, the pivot underscored a larger transformation underway in India Inc.—from hydrocarbons to hyperscale computing. The proposed data centers are designed not merely as commercial assets but as anchor points for what Reliance envisions as India’s next industrial leap.

Individually, each initiative would have marked a milestone. Together, they suggest a structural shift. India’s AI strategy is moving beyond rhetoric about innovation hubs and unicorn valuations and toward the harder terrain of power grids, cooling systems, semiconductor supply chains, and green energy integration. The conversation is no longer just about building models; it is about building the physical substrate that makes models possible.

This pivot comes as India already occupies a prominent, if still secondary, position in the global AI hierarchy. According to Stanford University’s Global AI Vibrancy Tool, the country ranks third worldwide, behind only the United States and China. It stands sixth in AI computing power, with roughly 1.2 million compute units, and ranks third in the deployment of AI chips. Eight AI clusters operate across the country, while the AIRAWAT supercomputer at Pune’s Centre for Development of Advanced Computing—Centre for Development of Advanced Computing—is among the world’s top 75.

Under the government’s India AI Mission, overseen by the Ministry of Electronics and Information Technology, New Delhi has pushed to expand datasets, foster indigenous foundation models, and establish an AI Safety Institute. The policy architecture has been steadily assembled. What was missing, critics argued, was scale.

The announcements by Tata and Reliance attempt to fill that gap. Yet the harder questions now begin. Building hyperscale data centers is capital-intensive; running them efficiently requires not only steady electricity but also advanced cooling, cybersecurity resilience, and a pipeline of high-end chips in a world of tightening export controls. As one former semiconductor R&D executive noted, ambition is not the same as execution. The test will lie in utilization rates, energy integration, and whether domestic AI firms can meaningfully tap this infrastructure rather than remain dependent on overseas clouds.

There is also a geopolitical dimension. In an era when compute is a strategic asset and supply chains are weaponized, India’s effort to expand domestic AI infrastructure is as much about autonomy as it is about innovation. The country’s leadership has long spoken of technological sovereignty; the new investments suggest that corporate India is aligning itself with that vision.

For decades, India’s global economic story was one of services layered atop limited industrial depth. AI infrastructure, by contrast, demands a return to heavy capital expenditure, grid planning, and long-horizon bets. If these projects materialize at scale, they may signal not only India’s arrival as a serious AI contender but also a broader reindustrialization driven by code, chips, and kilowatts rather than steel and steam.

Whether India can translate announcements into sustained capacity—and capacity into global influence—will determine whether this winter morning is remembered as a turning point or merely another flourish in the country’s long narrative of technological aspiration.

(Author is a former consultant with Protiviti India. Views are personal)