As an expert in artificial intelligence and the intricacies of blockchain technology, Dominic Jainy has watched the global digital landscape evolve from simple data storage to complex, high-velocity neural networks. With a career dedicated to understanding how hardware and software converge, he offers a unique perspective on India’s rapid ascent as a global data powerhouse. In this discussion, we explore the monumental shift occurring within the subcontinent, where data centers are no longer viewed as mere back-office utilities but as strategic national assets equivalent to the ports and power grids of the industrial age.
The conversation covers the extraordinary $126 billion investment pipeline currently flowing into the country, the shift from enterprise colocation to AI-ready hyperscale facilities, and the significant regulatory catalysts like the Digital Personal Data Protection Act. We also delve into the sobering physical realities of this growth—specifically the immense pressure on water and power resources—and the critical execution gap between announced projects and operational capacity. The dialogue provides a deep look at how India plans to leverage its massive user base and engineering talent to claim its share of the global AI infrastructure buildout.
The shift from traditional infrastructure like roads and railways toward data centers seems sudden and profound. How has the perception of these facilities evolved within India’s strategic planning?
Just five years ago, if you sat in a policy meeting in New Delhi, the conversation was almost entirely dominated by physical connectivity—highways, ports, and the expansion of the railway network. Data centers were seen as niche supporting infrastructure, the “plumbing” of the internet that mostly stayed out of sight and out of mind. Today, we are witnessing an incredible inflection point where data infrastructure has moved directly into the boardroom and the highest levels of government strategy. We are looking at a market where the total capacity crossed 1,700 MW in 2025, a figure supported by a record 440 MW of new supply that represents a 160% jump from the previous year. This isn’t just incremental growth; it’s a wholesale reimagining of what constitutes a national asset. When you see capacity additions doubling since FY23, when we only had 778 MW, you realize the scale of the transition from simple enterprise colocation to these massive, AI-first hyperscale buildouts that require a completely different level of power density and physical scale.
With over $126 billion in planned investments, global hyperscalers are making unprecedented commitments. What specifically is drawing players like Google, Microsoft, and Amazon to the Indian market at such a scale?
The gravity of the Indian market is becoming impossible for global tech giants to ignore, largely because of the sheer volume of data being generated by over 900 million internet users. When you consider that the average monthly wireless data consumption has crossed 25 GB per user, you begin to understand why companies like Amazon Web Services are pledging $35 billion by 2030 to expand their digital footprint here. Google has stepped up with a $15 billion investment in Andhra Pradesh that integrates compute power with renewable energy, while Microsoft has committed $17.5 billion through 2030. These companies aren’t just looking for cheap real estate; they are responding to India’s Digital Personal Data Protection Act, which essentially mandates that Indian user data stay within our borders. Furthermore, the country accounts for nearly 20% of all global AI-related development activities, and with AI hiring growing by 33% year-on-year, the talent pool is a massive magnet. Being able to hire software engineers at roughly $20,000 per year—a fraction of the cost in Silicon Valley or Singapore—creates a structural cost advantage that makes India a strategic necessity rather than an option.
The regulatory environment has moved from being passive to actively enabling this boom. How are the specific incentives in the recent Union Budget changing the financial math for these massive projects?
The shift toward active facilitation is perhaps the most significant tailwind we’ve seen in decades. The Union Budget 2026-27 introduced a 20-year tax holiday for foreign cloud providers that extends all the way to 2047, which provides a level of long-term fiscal certainty that was previously unthinkable. Additionally, the government is offering 25-35% capital incentives for developers who adopt green technologies in their construction processes. From a financial perspective, these moves are estimated to lower the weighted average cost of capital for these projects by 200 to 400 basis points. That might sound like a dry technical detail, but it directly improves project returns and accelerates the “go” decisions for operators who might have been hesitant. When you add a 15% safe harbor margin to reduce transfer pricing disputes, you create an environment where capital can flow without the friction of constant regulatory ambiguity.
We often hear about the “boom,” but there seems to be a significant gap between project announcements and what is actually operational. What is the reality on the ground regarding project execution?
There is a very real tension between the headlines and the physical reality of these massive construction sites. While the pipeline is overflowing with 3.5 GW of planned capacity from projects launched between 2025 and 2026, less than 20% of that announced capacity is actually live and operational. You can feel the frustration in the industry as developers navigate a landscape where frontier talent is growing at only 15% CAGR, while the AI workloads they are trying to support are scaling at 25-35%. Even though there are technically 1.25 lakh vacant industrial plots across the country, finding land that is “infrastructure-ready”—meaning it has confirmed, high-capacity power access—is becoming increasingly difficult. We are seeing a heavy concentration in Mumbai, which accounts for over 50% of our current inventory, and while cities like Hyderabad and Visakhapatnam have over 2 GW of planned capacity, the journey from a press release to a humming server rack is fraught with delays.
Environmental constraints, particularly regarding water and power, are often the silent killers of large-scale infrastructure. How is the industry addressing the massive resource requirements of AI-ready facilities?
The physical demands of AI are staggering and, frankly, quite visceral when you stand on-site at a major facility. A single 100 MW hyperscale facility requires approximately 2 million liters of water every single day for cooling, a requirement that clashes harshly with the fact that 60-80% of Indian data centers are located in high water-stress areas. Take Visakhapatnam, for example, where a 1 GW AI-native campus is planned; that district has some of the lowest groundwater availability for industrial use in the region, leading to significant legal and environmental challenges. On the power side, the transition is equally intense, with AI racks demanding over 50 kW of power density, which puts an incredible strain on the local grid. While it’s encouraging that operators like Nxtra and STT GDC India are reaching 60-63% renewable energy integration, we cannot ignore that coal still supplies about 70% of the nation’s actual generation. The smell of diesel from backup generators and the heat radiating from these high-density racks are constant reminders that our digital ambitions must eventually reconcile with our physical resources.
The workforce is often described as India’s greatest strength, but you’ve mentioned a mismatch in the specific skills needed for data center operations. Where is the human capital gap most visible?
India certainly has a world-leading IT workforce in terms of software and services, but operating a modern hyperscale campus requires a very different, highly specialized set of hands. We need mechanical and electrical engineers who understand the nuances of liquid cooling, HVAC specialists who can manage massive thermal loads, and power systems experts who can navigate complex grid integrations. These aren’t the roles typically produced by the standard engineering pipeline that has served the software export industry for twenty years. Currently, we have a talent pool where basic engineering is abundant, but the specialized operational workforce required for high-uptime data centers is still in its infancy. Building this human capital layer is just as important as pouring concrete or laying fiber; without it, we risk building high-tech shells that we don’t have the local expertise to maintain at a world-class level.
With Mumbai and Chennai dominating the current landscape, how is the geography of India’s digital infrastructure beginning to shift toward Tier-II cities?
The concentration in Tier-I cities is undeniable, with Mumbai, Chennai, Delhi-NCR, and Bengaluru hosting nearly 90% of our Tier-I capacity. However, the sheer cost of land and the rollout of 5G are forcing a diversification that is quite exciting to watch. Cities like Ahmedabad, Patna, and Bhopal are starting to see interest from developers because they offer lower land costs and meet the low-latency requirements needed for edge AI applications. As AI moves from centralized training to localized inference, we need the compute power to be closer to the user. This geographical shift isn’t just about saving money; it’s about the operational resilience of the entire national network. When you see a 1 GW project planned in a Tier-II city, you’re looking at a future where the digital economy is no longer confined to a few coastal hubs but is distributed across the heartland.
What is your forecast for India’s AI infrastructure over the next decade?
I believe we are entering a decade where India will transform from a consumer of global digital services into the very backbone that supports them. My forecast is that if we can bridge the current execution gap, India will successfully capture at least 4-5% of all global data center capacity additions by 2032. We will see the sector reach a value of $22 billion by 2030, but the real story will be the $46 billion opportunity in AI-optimized facilities specifically. However, this success is contingent on a fundamental shift in how we manage resources; we will likely see a massive move toward liquid-cooling technologies and a much more aggressive integration of renewable energy to bypass grid constraints. The “Indian way” has always been about leveraging scale and cost, and I expect us to do exactly that, turning our 900 million users and our $20,000-a-year engineering talent into a competitive advantage that makes the subcontinent the indispensable engine of the global AI era. The pipeline is there, the capital is there, and now the next seven years will be a test of our pure engineering and execution grit.