India's Data Centre Capacity Surges Over Fourfold to 1,575 MW as AI Demand Accelerates

India's data centre capacity rises over four-fold to 1,575 MW as AI demand accelerates: Govt

Since 2020, India's data centre capacity has surged fourfold to around 1,575 MW. This impressive growth stems largely from rising needs in artificial intelligence and high-performance computing sectors. The government is also prioritising renewable energy initiatives and efficient resource management, implementing new regulations and codes that emphasize energy and water conservation within data centre infrastructure. India's data centre capacity has increased more than four-fold from 375 MW in 2020 to around 1,575 MW currently, driven by the rapid growth of artificial intelligence (AI) and high-performance computing, the government informed the Lok Sabha on Wednesday. In a written reply in the Lok Sabha, Minister of State for Electronics and Information Technology Jitin Prasada said the IndiaAI Mission is building a sovereign AI ecosystem while the government is putting in place measures to ensure the environmentally sustainable expansion of AI infrastructure and data centres. "Data Centre industry in India has been growing at a steady pace and has grown from 375 MW in 2020 to current capacity of about 1575 MW. The rapid growth of AI and high-performance computing is driving increasing demand for data centre capacity in the country," the reply said. The minister said the IndiaAI Mission aims to build a sovereign AI stack comprising compute capacity, foundational models and indigenous applications, with a focus on democratising access to AI technologies. To support sustainable expansion, the government has introduced multiple measures to promote renewable energy and improve resource efficiency. These include the Green Energy Open Access Rules, Green Energy Corridor Scheme, Green Term Ahead Market (GTAM), National Green Hydrogen Mission, National Programme on High Efficiency Solar PV Modules, Offshore Wind VGF Scheme and Solar Park Scheme. The reply also stated that the recently enacted Sustainable Harnessing and Advancement of Nuclear Energy for Transforming India (SHANTI) Act is intended to strengthen reliable clean power for emerging sectors such as AI and data centres, including the future deployment of Small Modular and Micro Nuclear Reactors. The government informed Parliament that the Bureau of Indian Standards (BIS) has published Indian standards covering Power Usage Effectiveness, Carbon Usage Effectiveness, Cooling Efficiency Ratio and Water Usage Effectiveness for data centres. The Bureau of Energy Efficiency has also notified the Energy Conservation Building Code (ECBC 2017) and the Energy Conservation & Sustainable Building Code (ECSBC 2024), which prescribe energy efficiency and water conservation norms relevant to data centres. According to the reply, the industry is increasingly adopting advanced cooling technologies, including direct-to-chip liquid cooling, adiabatic cooling, immersion cooling and closed-loop liquid cooling systems, while deploying high-density racks to support AI workloads with lower power and water consumption. The government said these measures together provide a framework for improving electricity efficiency, renewable energy use, cooling efficiency, carbon emissions and water usage as AI infrastructure continues to expand.

Govt cites nuclear power law, BIS to address data centre strains

India's data centre capacity has increased more than fourfold in six years, from 375 MW in 2020 to approximately 1,575 MW. This week, the government informed Parliament that it is addressing the environmental impact of expanding AI infrastructure through renewable energy initiatives, a new nuclear power law, and technical standards. Minister of State for the Ministry Electronics and Information Technology (MeitY) Jitin Prasada provided this information in a written reply to the Lok Sabha on August 5, 2026, in response to MP Kanwar Singh Tanwar's questions about the environmental sustainability of AI data centres. Tanwar asked: The core numbers: The minister's reply confirmed that data centre capacity increased from 375 MW in 2020 to approximately 1,575 MW, due to higher demand for compute capacity from AI and high-performance computing. No projections were provided for future capacity or resource demand. Nuclear power enters the mix: One key disclosure was confirmation that the Sustainable Harnessing and Advancement of Nuclear Energy for Transforming India (SHANTI) Act has been enacted to secure reliable clean power for sectors such as AI and data centres. The law supports future deployment of Small Modular Reactors and Micro Nuclear Reactors, marking a shift from the government's previous focus on renewables for data centre energy needs. Renewable energy schemes cited: The reply identified several renewable energy initiatives supporting the sector, such as the Green Energy Open Access Rules, Green Energy Corridor Scheme, Green Term Ahead Market (GTAM), National Green Hydrogen Mission, National Programme on High Efficiency Solar PV Modules, Offshore Wind Viability Gap Funding Scheme, and Solar Park Scheme. However, it did not clarify what portion of current or future data centre power demand these schemes will address, nor did it mention any dedicated funding for AI infrastructure. The standards body has published efficiency metrics: The Bureau of Indian Standards, through its LITD 31 technical committee on Cloud Computing, IT, and Data Centres, has published Indian Standards for Power Usage Effectiveness, Carbon Usage Effectiveness, Cooling Efficiency Ratio, and Water Usage Effectiveness. The reply did not clarify whether compliance with these standards is mandatory or voluntary or specify any related enforcement or auditing mechanisms. Separately, the Bureau of Energy Efficiency has notified two building codes: the Energy Conservation Building Code (ECBC 2017) and the Energy Conservation & Sustainable Building Code (ECSBC 2024). These codes establish energy and water efficiency norms for buildings, including provisions the government identifies as relevant to data centres. Water use left to State and Ministry discretion: The response provided limited information on water consumption. It only noted that water requirements depend on the cooling technology used and that groundwater extraction, including for industrial purposes, is regulated by Ministry of Jal Shakti guidelines. No data centre-specific water usage figures, limits, or monitoring requirements were provided. The response stated that the industry is adopting advanced cooling methods, such as direct-to-chip liquid cooling, adiabatic cooling, immersion cooling, and closed-loop liquid systems. It also mentioned the use of high-density server racks to reduce both power and water consumption per unit of compute. What wasn't answered: MP Tanwar's question included five parts, seeking details on operator incentives, monitoring, inter-ministerial coordination, and expected outcomes of sustainability initiatives. The government responded collectively rather than addressing each part individually, leaving several issues unresolved. Specifically, the response did not provide figures on current AI-related electricity or water consumption, a timeline for adopting BIS standards, or quantified environmental targets and outcome metrics. Context: The reply presents these measures as part of the broader IndiaAI Mission, which aims to build a domestic AI stack covering computing capacity, foundational models, and applications. The Lok Sabha exchange offers one of the most detailed official accounts to date of the regulatory and infrastructure tools the government is using to manage the environmental impact of AI expansion. However, it does not set measurable targets or disclose current consumption data. The government's focus on technical standards and reliable power follows a significant data centre disruption earlier this year. In June, a fire at a third-party data centre in Delhi led to an emergency shutdown of Google Cloud networking equipment, disrupting network traffic for customers in Delhi, Mumbai, Chennai, and nearby areas. Google reported that the incident isolated a local point of presence (PoP), reduced network capacity, and increased latency as traffic was rerouted. A CEEW study estimates that installed capacity, currently about 1.5 GW as cited by the minister, will rise to 6.5 GW by 2030. Both the sector's share of national electricity and its annual water consumption, now near 150 billion litres, are expected to more than double in this period. Location adds to the risk: a WRI India mapping exercise found that over half of the country's data centres are in water-stressed regions. Earlier in April the Human Rights Forum has called for suspension of environmental clearance for a hyperscale Google facility under construction in Tarluvada, Andhra Pradesh, due to undisclosed water-sourcing plans. The reliance on fossil fuels remains significant, with coal and other sources supplying about 71-72% of India's electricity. As a result, continued data centre growth is likely to depend on coal, despite government references to renewable and nuclear options. Data centres are not classified as a separate category under the Environment Impact Assessment notification and are instead approved as building or township projects, which exempts them from public hearings. This gap allowed at least one hyperscale facility to receive environmental approval within 10 days of application.