India’s Data Centre Boom: Costs, Beneficiaries & Sustainability
Engineers working in a data center in India.
Key Highlights
Explore the rapid expansion of India’s data centres, the hidden costs, resource implications, and the true economic beneficiaries of this tech boom.
India’s Ambitious Data Centre Expansion Faces Scrutiny
India is currently experiencing a significant surge in data centre construction, attracting substantial investments from global technology leaders and benefiting from government policies related to infrastructure and land allocation. This burgeoning sector, encompassing server farms, cloud regions, and AI-ready capacity, is projected to more than double its market valuation from INR 9.33 Lakh Crore in 2025 to INR 20.53 Lakh Crore by 2030. Major players like Google, with a $15 billion AI data centre hub planned in Visakhapatnam, and Microsoft, investing $3 billion to expand its cloud and AI infrastructure, are at the forefront of this expansion. AWS, AdaniConneX, STT GDC, and CtrlS are also scaling their operations across key cities like Mumbai, Chennai, and Hyderabad. However, this rapid growth prompts crucial inquiries into India’s natural resource sustainability, the actual beneficiaries of this infrastructure boom, and whether it fosters domestic capabilities or facilitates global resource extraction.
The Economic Advantage Masking Resource Strain
India’s appeal as a data centre hub is significantly driven by its lower land and labour costs, coupled with improving infrastructure and a rapidly digitizing economy. This cost advantage, as noted by Amit Chaurasia, founder and CEO of Dataneers, makes building and operating data centres in India substantially more cost-effective than in Western markets. While this translates to millions in capital savings for companies investing in AI infrastructure, it also places a considerable burden on India’s resources and workforce. A study by the Council on Energy, Environment and Water (CEEW) indicates that data centres consumed approximately 0.5% of India’s electricity and 150 billion litres of water in 2025, with these figures expected to more than double by 2030. A single 100 MW hyperscale facility can consume up to 2 million litres of water daily, presenting a difficult choice between water-intensive cooling and energy-intensive air cooling. Operational expenses, with energy costs accounting for 60-70%, are often managed through preferential power arrangements and state incentives, highlighting the complex interplay between cost efficiency and resource management.
Global Hyperscalers Dominate Demand, Indian Startups Lag
The demand landscape for India’s data centre capacity is largely shaped by global hyperscalers, who account for an estimated 65-70% of current capacity. According to Manoj Dhanda, founder and CEO of Utho Cloud, these major cloud providers pre-book significant capacity, making large-scale infrastructure financing viable. In contrast, India’s over 3,000 AI startups and enterprises currently consume a small fraction of this infrastructure, often relying on hyperscaler credits rather than paying for dedicated compute. The cost of training an AI model on high-end GPUs can range from $80,000 to $120,000 per month, a prohibitive expense for most Indian startups. This creates a tiered market where global enterprises and hyperscalers are the primary consumers, while domestic startups operate with subsidised access, lacking full ownership or control. While rising GPU costs and data localisation mandates are expected to boost domestic demand to 15-20% in the next four to five years, global workloads still significantly influence current GPU consumption, according to Karan Kirpalani, Chief Product Officer at Neysa.
Value Capture and Future Directions for India’s Data Ecosystem
The ongoing data centre boom in India, characterised by large campuses and frequent cloud region announcements, is undeniably building critical digital infrastructure and attracting foreign capital. However, a significant portion of this infrastructure is driven by global demand, financed by foreign capital, and anchored by hyperscalers whose primary value chains are located outside India. Consequently, the revenue generated from AI workloads running on this infrastructure does not always directly benefit India. The rising consumption of power and water necessitates evolving policy frameworks to address long-term resource trade-offs. As India continues to build its AI infrastructure, it faces a strategic choice: remain a low-cost hub for global data centres or strive to capture more value by fostering domestic demand, investing in its own compute capacity, and implementing policies that ensure greater benefits remain within the country. A.S. Rajgopal, MD and CEO of NxtGen Cloud Technologies, points out that while India gains jobs, tax revenues, and infrastructure credibility, it has yet to secure ownership of the value chain.
Emerging Trends and Startup Innovations in AI and Sustainability
Beyond the core data centre infrastructure, the broader AI and technology landscape in India is also dynamic. Palo Alto Networks is set to acquire Portkey to enhance its AI security capabilities. Fractal is restructuring its operations around enterprise AI, appointing new leadership. Maharashtra has launched a new AI policy targeting significant investments and job creation, while Goa is developing its own draft policy. SoftBank is investing heavily in AI infrastructure through its new venture, Roze AI, focused on robotic data centre construction. Fintech companies like Stripe are expanding agentic payment capabilities, and Google is enhancing its Gemini AI with direct file creation features. In the sustainability sector, startups like EarthSync are developing AI-powered platforms to optimize renewable energy investments, addressing the complex decision-making involved in decarbonisation strategies for enterprises.