The Energy Behind Intelligence: India’s Data Centre Boom Meets the Coming Power Challenge

In 1973, the oil crisis reshaped the global economy and transformed energy security into a central pillar of national strategy. More than five decades later, the world may be approaching another strategic inflection point not driven by oil shortages, but by the unprecedented energy demands of artificial intelligence.

Governments, corporations, and investors across the world are racing to build AI capabilities. Trillions of dollars are being committed to semiconductors, hyperscale data centres, cloud infrastructure, advanced computing, and sovereign AI programs. Yet beneath the excitement surrounding algorithms, large language models, autonomous systems, and digital transformation lies a reality that is often overlooked: artificial intelligence is fundamentally an energy-intensive enterprise.

The critical question confronting India is therefore not whether it can build data centres, develop AI models, or create digital platforms. It is whether the country can supply abundant, affordable, reliable, and sustainable electricity to power them over the next three decades.

Former Chief Economic Adviser Arvind Subramanian recently raised concerns about the possibility of persistently elevated energy prices and the difficulties governments face in indefinitely shielding consumers from market realities. While his remarks were directed toward broader economic policy, they have profound implications for India’s digital future. If energy prices remain structurally high, the economics of AI, cloud computing, semiconductor manufacturing, digital infrastructure, and advanced technology industries could be fundamentally altered.

As energy historian Daniel Yergin observed, “Energy is the capacity to do work. It is the basis of industrial society.” In the AI era, energy is rapidly becoming the basis of digital society as well.

The Global AI Race Is Quietly Becoming an Energy Race

Much of the public discussion surrounding artificial intelligence focuses on software breakthroughs, talent acquisition, semiconductor manufacturing, and data governance. However, a profound shift is underway in boardrooms and government offices worldwide. Increasingly, the most important question is not where AI talent resides but where electricity can be secured.

According to the International Energy Agency, electricity consumption by data centres, AI systems, and cryptocurrency operations is expected to grow dramatically during this decade. Recent IEA analyses indicate that global data centre electricity consumption could more than double by 2030, driven overwhelmingly by AI workloads. Some forecasts suggest that data centres may consume more electricity than entire industrialized nations currently consume today.

The reason is simple. AI systems require enormous computational power. Training a frontier AI model involves processing trillions of parameters across thousands of advanced graphics processing units (GPUs). Inference the continuous operation of AI systems after deployment creates an even larger long-term electricity demand.

Traditional enterprise data centres historically operated at rack densities of 5–15 kilowatts. Modern AI facilities are increasingly designed around rack densities exceeding 100 kilowatts, with some next-generation deployments targeting 150–250 kilowatts per rack. What was once primarily a digital infrastructure business is rapidly becoming an energy business.

Vaclav Smil, one of the world’s foremost energy scholars, has repeatedly emphasized that every major technological transformation in history has ultimately been constrained by energy availability. The AI revolution is unlikely to be an exception.

India’s Data Centre Boom: Impressive Growth, Hidden Risks

India has emerged as one of the fastest-growing data centre markets in the world.

The country’s digital economy already contributes substantially to national output and continues to expand through fintech, e-commerce, cloud adoption, digital public infrastructure, defence digitisation, healthcare technology, and artificial intelligence applications. Industry estimates suggest that India’s data centre capacity could increase several-fold by the early 2030s, supported by investments potentially exceeding ₹2–3 lakh crore.

Major global operators and domestic players have announced large-scale projects across Mumbai, Chennai, Hyderabad, Bengaluru, Noida, and emerging technology corridors. The combination of data localisation requirements, digital adoption, favourable demographics, and government initiatives has created a powerful growth engine.

Yet there is an uncomfortable reality beneath these optimistic projections.

Electricity represents the single most important operational input for data centres. In many facilities, power and cooling expenses account for a significant share of operating costs. Unlike traditional industries, where energy is one of several important inputs, data centres effectively convert electricity into digital services.

Every search query, AI inference request, cloud transaction, digital payment, streaming session, and cybersecurity operation ultimately consumes electricity.

Consequently, the economics of India’s digital future are becoming increasingly dependent on the economics of its power sector.

The Scale of the Challenge Is Greater Than Most Policymakers Realize

India’s total electricity demand is already growing rapidly due to urbanisation, industrialisation, rising living standards, electric mobility, and manufacturing expansion.

Now artificial intelligence is emerging as a major new source of demand.

The convergence of sovereign AI programs, defence applications, smart manufacturing, digital twins, autonomous systems, robotics, geospatial intelligence, financial analytics, and advanced simulation environments will require unprecedented computational infrastructure.

Globally, leading technology companies are making investments that would have seemed unimaginable only a few years ago. Estimates suggest that major hyperscalers could collectively invest hundreds of billions of dollars annually in AI-related infrastructure during the coming decade.

The power requirements are equally extraordinary.

Several AI campuses currently under development in the United States are planned at capacities approaching or exceeding one gigawatt. To place this in perspective, one gigawatt of continuous power can support a city with a population of several hundred thousand people.

The emerging reality is that future AI facilities will increasingly compete directly with other sectors of the economy for electricity resources.

Henry Kissinger, in his final writings on artificial intelligence, warned that AI would reshape the foundations of economic and geopolitical power. What was less widely appreciated is that this transformation would also reshape the foundations of energy demand.

Why Persistently High Energy Prices Could Become a Strategic Threat

The implications of elevated energy prices extend far beyond higher operating costs.

First, rising electricity tariffs would directly affect the competitiveness of India’s data centre industry. Even relatively modest increases in industrial power tariffs can significantly alter the economics of hyperscale facilities operating twenty-four hours a day throughout the year.

Second, expensive electricity raises the cost of AI development and deployment. Training advanced models, operating inference clusters, supporting cloud infrastructure, and delivering AI services all become more costly. This can reduce innovation, discourage startups, and slow adoption across sectors.

Third, investor confidence may be affected. Data centre investments are long-term infrastructure bets with economic horizons extending twenty to thirty years. Investors evaluating locations increasingly prioritize power availability, tariff predictability, and energy security.

A data centre developer can often overcome challenges related to land acquisition or permitting. Uncertain long-term electricity costs are considerably harder to manage.

Fourth, persistently high energy prices could undermine India’s ambition to become a preferred destination for global AI infrastructure.

Capital is increasingly mobile. Investors can choose between multiple jurisdictions. If power economics become less attractive, future investments may gravitate toward regions offering more predictable and competitive energy environments.

The New Global Competition: Energy Economics as Industrial Policy

Around the world, governments are recognizing that digital infrastructure competitiveness depends on energy strategy.

The United States is witnessing substantial investments in power generation and transmission infrastructure linked directly to AI development. Several technology companies are exploring dedicated nuclear energy arrangements to support future computing requirements.

The United Arab Emirates and Saudi Arabia are leveraging abundant energy resources to position themselves as AI and digital infrastructure hubs.

Malaysia and Indonesia are aggressively competing for hyperscale investments through energy availability, industrial incentives, and infrastructure development.

Even Singapore, despite land constraints, has adopted highly sophisticated approaches to balancing data centre growth with energy sustainability objectives.

The lesson is clear: data centres are no longer viewed merely as real estate projects. They are increasingly regarded as strategic national assets.

The Water-Energy Nexus: India’s Overlooked Vulnerability

Electricity is not the only challenge.

The next generation of AI data centres requires sophisticated cooling systems that consume both energy and water. This creates a dual-resource dependency that is particularly relevant for India.

According to various global assessments, India remains among the most water-stressed countries in the world. Several regions already face growing pressure on freshwater resources due to urbanisation, industrial activity, agricultural demand, and climate variability.

Future hyperscale facilities will therefore require integrated planning for electricity, water, cooling technologies, recycling systems, and environmental sustainability.

A nation cannot become a global AI hub if either electricity or water becomes a binding constraint.

India’s Advantages Are Significant and Strategic

Despite these challenges, India’s position remains fundamentally strong.

The country possesses one of the world’s most ambitious renewable energy programs. Solar tariffs are among the lowest globally. Wind energy capacity continues to expand. Grid modernization efforts are progressing. Interstate transmission networks are becoming increasingly robust.

India’s target of achieving 500 GW of non-fossil fuel electricity capacity represents one of the largest energy transitions underway anywhere in the world.

Equally important, India enjoys advantages that many competitors cannot replicate: a vast domestic market, world-class digital public infrastructure, an expanding technology workforce, strong entrepreneurial capabilities, and growing geopolitical relevance.

As Nandan Nilekani has frequently argued, digital infrastructure has become a foundational layer of economic development. India’s achievements in this domain are globally recognized.

However, digital infrastructure ultimately requires physical infrastructure beneath it—and that infrastructure depends on energy.

The Nuclear Question Is Returning to the Strategic Agenda

One of the most significant developments globally is the renewed interest in nuclear power as a source of reliable electricity for AI infrastructure.

Major technology firms are increasingly exploring partnerships linked to advanced nuclear technologies, including Small Modular Reactors (SMRs). The attraction is obvious: carbon-free baseload electricity available twenty-four hours a day irrespective of weather conditions.

For India, which has long maintained nuclear expertise and ambitions, the emergence of AI-driven electricity demand may provide a compelling rationale for accelerating advanced nuclear deployment.

If AI becomes a strategic national priority, reliable baseload electricity will become equally strategic. In the coming decade, discussions about AI policy may increasingly overlap with discussions about nuclear policy.

A National Imperative: Energy Sovereignty for the Age of Intelligence

The central policy challenge is not simply increasing electricity generation. It is ensuring long-term energy affordability, reliability, and predictability.

India should consider establishing dedicated AI and Data Centre Power Corridors supported by renewable generation, energy storage systems, advanced transmission infrastructure, and eventually SMR-based baseload capacity. Long-term power purchase arrangements should provide tariff certainty for strategic digital infrastructure investments.

Dedicated water-security frameworks, accelerated grid modernization, expanded energy storage deployment, and integrated planning between digital and energy ministries should become national priorities.

Most importantly, data centres and AI infrastructure should be recognized as critical national infrastructure deserving strategic policy attention comparable to ports, airports, telecommunications networks, and transportation corridors.

The Future of Intelligence Depends on the Future of Energy

For much of the digital age, electricity was treated as an operational consideration. In the age of artificial intelligence, it is becoming a determinant of national power.

The coming decade will witness intense competition among nations seeking leadership in AI, advanced computing, digital services, and technological innovation. While discussions will continue to focus on chips, algorithms, and talent, the decisive factor may lie elsewhere.

The countries that can provide abundant, reliable, affordable, and sustainable energy will increasingly shape the future of intelligence.

India’s aspirations to become a leading AI nation are entirely achievable. The country possesses the talent, market scale, entrepreneurial capability, strategic vision, and technological foundation necessary for success.

But ambition alone will not power data centres. Electricity will.

The defining strategic question for policymakers is therefore no longer, “How many data centres can India build?” Rather, it is “How much affordable and reliable energy can India guarantee to those data centres for the next thirty years?”

The answer to that question may determine whether India becomes one of the world’s foremost producers of artificial intelligence or merely one of its largest consumers.

As the world enters the age of machine intelligence, one truth is becoming increasingly evident: the most important fuel for the digital economy is not data. It is power.

(Major General Dr. Dilawar Singh, IAV, is a distinguished strategist having held senior positions in technology, defence, and corporate governance. He serves on global boards and advises on leadership, emerging technologies, and strategic affairs, with a focus on aligning India’s interests in the evolving global technological order.)