The AI Data Centres Infrastructure in India

Source: The post “The AI Data Centres Infrastructure in India” has been created, based on “The AI Data Centres Infrastructure in India” published in “The Hindu” on 31 October 2025. The AI Data Centres Infrastructure in India.

UPSC Syllabus: GS Paper -3- Technology

Context: Over the past two decades, India’s electricity demand has grown steadily at about 5 per cent annually. Traditionally, energy demand was managed through coal-based generation and grid planning, but the rapid expansion of Artificial Intelligence (AI), 5G, and digital infrastructure has changed the landscape. The growth of AI data centres, which support AI-driven computing, cloud services, and massive data storage, is expected to cause a steep rise in electricity consumption. Consequently, ensuring a sustainable, low-carbon, and continuous power supply for these data centres has emerged as a strategic priority for India’s energy and technology policy.

Growing Energy Demand of AI Data Centres

1. AI data centres require far greater power than traditional enterprise servers because of their heavy computational workloads.
2. Training Generative AI models and Large Language Models (LLMs) such as ChatGPT demands enormous processing power using Graphics Processing Units (GPUs) and Tensor Processing Units (TPUs).
3. A typical AI data centre can consume 5 to 20 times more power than a traditional one, depending on workload and efficiency.
4. Globally, the power requirement for AI data centres is projected to grow from around 460 terawatt-hours (TWh) in 2024 to 1,000 TWh by 2030.
5. China’s AI and Large Model data centres alone are expected to consume more than 400 billion kWh by 2025.
6. India’s current capacity of 0.4 gigawatts (GW) is projected to exceed 10 GW by 2030, driven by the expansion of Digital India, data localisation, and AI adoption.

Drivers of Data Centre Growth in India

1. The growing need for cloud storage, AI training models, and data-intensive operations has increased the number of data centres across cities like Chennai, Bengaluru, Hyderabad, and Mumbai.
2. Government programmes such as Digital India, AI Mission, and data localisation mandates have encouraged both domestic and foreign investment.
3. Several private and global tech firms are setting up large-scale AI data campuses to support India’s digital and industrial transformation.
4. This expansion, while boosting the economy, is also exerting pressure on the national electricity grid and demanding new, sustainable energy sources.

Current and Emerging Power Sources

To meet the rising energy demand, data centres are using a diverse energy mix:

1. Renewable energy sources like solar and wind are central to decarbonisation targets.
2. Hydropower and natural gas are used to balance the intermittency of renewables.
3. Hybrid systems, integrating renewables with battery storage and green hydrogen, are emerging as dependable clean energy options.
4. However, due to the constant 24×7 power requirements of AI operations, there is a growing need for baseload, non-intermittent, and low-carbon energy, which renewables alone cannot fully guarantee.
5. Hence, attention is turning towards Small Modular Reactors (SMRs) as a long-term, sustainable energy solution for powering AI data hubs.

Role and Advantages of Small Modular Reactors (SMRs)

1. Small Modular Reactors (SMRs) are compact nuclear reactors that can be manufactured, transported, and installed in modular units.
2. They typically have a capacity of 50–300 megawatts (MW) and use slightly enriched uranium as fuel.
3. They offer several key advantages:
   1. Scalability and Modularity: SMRs can be installed in clusters to match energy demand growth.
   2. Passive Safety Features: They use natural convection and advanced cooling systems, reducing the risk of accidents.
   3. Lower Construction Time: Modular design enables faster deployment compared to traditional nuclear plants.
   4. Reduced Land and Water Use: They require smaller sites, making them suitable for integration with industrial parks and data campuses.
   5. Clean, Carbon-Free Energy: SMRs provide constant, zero-carbon baseload power essential for uninterrupted AI computing.
4. India’s nuclear research institutions like BARC and the Department of Atomic Energy (DAE) are working on 300 MW indigenous SMR designs.
5. Globally, SMR projects are being pursued by countries such as the U.S.A, Canada, and the U.K., with support from the International Atomic Energy Agency (IAEA) to establish safety and licensing frameworks.

Challenges and Regulatory Concerns

1. Despite their promise, several challenges slow down SMR adoption:
   1. High upfront capital cost and complex technology integration.
   2. Lengthy regulatory clearances and evolving nuclear safety standards.
   3. Public concerns regarding nuclear waste management and perceived safety risks.
   4. Limited supply chain capacity for reactor fabrication and skilled personnel.
   5. Lack of clear financial and policy frameworks for private participation in nuclear energy.
2. These issues must be addressed before SMRs can play a central role in powering AI data centres at scale.

Way Forward

1. Develop a National SMR Strategy: India should frame a clear roadmap for SMR development with defined milestones, safety standards, and financing models.
2. Promote Public–Private Partnerships: Collaboration between government agencies, private tech firms, and international players can accelerate SMR deployment for data centres.
3. Strengthen Renewable–Nuclear Integration: Combining SMRs with renewable sources can create a resilient, hybrid energy system to power future AI clusters.
4. Regulatory Modernisation: Simplified and transparent licensing procedures should be introduced to ensure faster, safer deployment.
5. Public Awareness and Global Cooperation: Building public trust and aligning with IAEA safety frameworks will be crucial for long-term acceptance and global competitiveness.

Conclusion: The rapid proliferation of AI-based data centres marks a new frontier in India’s digital economy but also presents unprecedented energy challenges. Traditional power sources and standalone renewables will not be sufficient to meet the massive, uninterrupted, and low-carbon energy demand. Small Modular Reactors, with their scalability, safety, and clean energy potential, can serve as a reliable backbone for future AI-driven infrastructure. Adopting an integrated approach combining SMRs, renewables, and green hydrogen will help India ensure energy security, support digital innovation, and advance towards a sustainable, net-zero future.

Question: What will power AI data centres in the future? Discuss the energy demand, power sources, and the role of Small Modular Reactors (SMRs) in ensuring sustainable and reliable electricity supply.