India’s digital future just got a powerful signal the recent reports that Google plans to invest $10 billion to build a 1-gigawatt data-centre cluster in Visakhapatnam mark more than a single corporate announcement. These reports mark India’s arrival as a serious contender in the global AI infrastructure race, if realized this will be one of the largest foreign investments in India’s compute backbone and it could reshape how the country powers connect and sustains the data economy of the AI era.
At the same time this investment also forces a deeper question which is Can India convert capital into reliable sustainable and scalable capacity? Because while demand is exploding the supply side still faces hard limits.
According to Deloitte’s May 2025 report “Attracting AI Data Centre Infrastructure Investment in India” the country generates nearly 20% of the world’s data but hosts only around 3% of global data-centre capacity. The report calls this “a once-in-a-generation inflection point where digital sovereignty meets infrastructure readiness”. That imbalance is both a challenge and an opportunity, one that India must be addressed quickly if it wants to host the next wave of AI-driven workloads rather than export them.
At the same time Knight Frank’s Global Data Centres Report 2025 finds that “data centres have moved from being alternative assets to essential infrastructure”. Global institutional capital is flooding into the sector compressing yields and transforming campuses into long-term energy-backed investments. BCG’s report titled “Breaking Barriers to Data Center Growth” projects that global data-centre power demand will reach approx. 30 GW by 2028 driven by GenAI and a $18 trillion capex wave. “GenAI will be the fastest-growing segment of computing demand”. BCG also noted that the traditional enterprise workloads will still account for the majority of demand meaning sustained not cyclical infrastructure growth.
These insights paint a clear picture which states that demand isn’t the problem rather its infrastructure execution. India’s success depends on how it tackles three structural gaps power land and compute supply chains.
The first and hardest challenge is power, A gigawatt-class campus requires stable redundant power that most Indian grids aren’t yet equipped to deliver. The BCG report notes that the growing number and size of data centres will have a transformative impact on electrical grids with global power demand from data centres expected to grow 16% annually. The report warns that mismatched planning horizons between data-centre developers (2–3 years) and utilities (4–8 years) are now a critical bottleneck. India can address this through early collaboration between operators and DISCOMs long-term power purchase agreements and shared-risk grid investment models.
The second challenge is land and approvals, Hyperscale’s need large contiguous parcels near fibre and ports and as Deloitte points out “land aggregation environmental approvals and clearances remain the longest poles in the tent”. The solution lies in state-level Data Centre Economic Zones (DCEZs) and Data Centre Facilitation Units (DCFUs) that offer powered land single-window approvals and pre-certified “plug-and-play” campuses.
The third challenge is the compute and cooling supply chain the access to GPUs high-density racks and efficient thermal systems has become the defining constraint. BCG’s research finds that lead times for critical equipment like backup generators have gone from months to years. India’s response must include pooled GPU procurement (under the India-AI mission) incentives for local assembly and capacity-building for mechanical-electrical-plumbing (MEP) contractors.
Despite these constraints the opportunity is immense as highlighted in Knight Frank report stating that data centres are increasingly being recognized not only as an economic driver but as a social infrastructure multiplier creating thousands of skilled jobs and anchoring new renewable-energy demand. A $10 billion hyperscale project could trigger ripple effects across renewables fibre connectivity and advanced manufacturing while unlocking an entirely new class of energy-backed investment vehicles.
In order to make this happen India should act decisively publish an inventory of powered land parcels, standardize take-or-pay PPAs for large data loads, pilot AI-ready showcase campuses that integrate renewable power, waste-heat recovery and community benefit programs. Each step signals predictability the single biggest attractor of global capital. In BCG’s words “creative collaboration throughout the ecosystem among companies’ utilities and regulators is what will separate the leaders from the laggards”. India now has a chance to prove it can be the former.
The big picture is simple hyperscale capital is coming and AI compute demand is compounding, but to truly lead India must fix power reliability land efficiency and compute supply now and If it does India won’t just host the world’s data but it will power the world’s intelligence. Because this decade’s infrastructure race isn’t about cables or clouds anymore it’s about who powers intelligence itself.
*****
Disclaimer: The entire contents of this article have been prepared based on relevant provisions and as per the information existing at the time of the preparation. Maximum care has been taken to ensure the accuracy, completeness, and reliability of the information provided. Users of this information are expected to refer to the relevant existing provisions of applicable Laws. The user of the information agrees that the information is not a piece of professional advice and is subject to change without notice. I assume no responsibility for the consequences of the use of such information.
Author Bio
