Beyond the Code, Why India’s AI Future Depends on Energy, Rare Earths, and a Legal Framework
The artificial intelligence revolution is often discussed in the language of code—algorithms, models, datasets, and compute. But as the technology matures from lab curiosity to industrial backbone, a different set of challenges is coming into focus. These challenges are not about how to make AI smarter, but about how to power it sustainably, where to source the materials that build its hardware, and how to govern its environmental impact.
India stands at a critical juncture. The AI market is growing rapidly, with companies integrating the technology across sectors from healthcare to agriculture. The future of AI in India, as independent researcher Niranjana KarthigaiRajan argues, will not be shaped by algorithms alone. It will be shaped by sustainable infrastructure, strategic foresight, and the ability to navigate a global landscape where critical resources are increasingly concentrated in the hands of a few players.
The Hidden Cost of Intelligence
AI systems are computationally intensive. Training large language models requires vast arrays of servers running 24/7, consuming enormous amounts of electricity and generating corresponding amounts of heat. The environmental footprint is staggering.
According to a July 2024 study by the United Nations Environment Programme (UNEP) titled “Navigating New Horizons,” a single request made to ChatGPT consumes up to 10 times more energy than performing a standard Google search. When multiplied by billions of queries, the energy demand becomes a significant factor in global electricity consumption.
Water consumption is another concern. Data centres generate immense heat and require cooling systems that often rely on water evaporation. In water-stressed regions, the presence of large data centres can strain local resources. The carbon footprint of AI, if powered by fossil fuels, adds to the very climate crisis that technology might otherwise help solve.
The direction of AI infrastructure is shifting toward sustainability. Google’s Project Suncatcher, which aims to launch data centres into low-earth orbit and run them entirely on solar energy, is one example of out-of-the-box thinking. But such futuristic solutions are years away. In the near term, strengthening AI infrastructure through enhanced energy efficiency and greater integration of renewable energy sources will be crucial to ensuring long-term competitiveness and technological leadership.
The Regulatory Landscape
The environmental impact of AI has not gone unnoticed by regulators. The United States and the European Union are among the leading jurisdictions addressing the issue through proposed and adopted measures. The US Artificial Intelligence Environmental Impacts Act of 2024 and the EU’s Artificial Intelligence Act, which establishes harmonised rules for AI governance, both include provisions related to sustainability and transparency.
China has introduced a series of AI-specific regulations, while South Korea enacted the AI Basic Act in January 2026, making it one of the first countries to implement a comprehensive national AI law. These frameworks are not just about controlling AI; they are about shaping its development in ways that align with national priorities.
India has taken certain measures, such as amending its IT Rules in 2026 to mandate AI content labelling and address online harms. But it has yet to introduce a comprehensive AI framework that systematically regulates risk assessment, transparency, liability, safety, and ethical standards across sectors. The environmental dimension is particularly neglected.
Without a legal framework that mandates transparency about AI energy usage, encourages greener data centres, and ensures sustainability standards, India risks falling behind. Not because its AI models are less sophisticated, but because its infrastructure is less sustainable and its regulatory environment less predictable.
The Rare Earths Reality
There is another dimension to AI infrastructure that is often overlooked: the materials required to build it. Rare Earth Elements (REEs) are critical components in clean energy technologies like electric vehicles and in high-tech industries including semiconductors and AI hardware.
China has emerged as a monopoly over REEs, contributing to 60% of global production in the last five years and controlling 92% of global refining capacity. With its dominance across reserves, production, and exports, China has gained the bargaining power to shape the global flow of these critical resources.
This concentration poses a strategic risk for any country seeking to build a domestic AI and semiconductor industry. Without access to REEs, or without the capacity to process them, India could find itself dependent on a single supplier for materials essential to its technological future.
China’s trade surplus crossed a record $1.2 trillion in the first 11 months of 2025, driven by its dominance in global value chains in high-technology manufacturing sectors such as electric vehicles, batteries, and solar panels. This is not an accident; it is the result of decades of strategic investment in building processing capacity and value-chain integration.
India must similarly prioritise securing a stable supply of REEs. This means increasing funding for mineral exploration, nurturing domestic manufacturing processes, strengthening the human resource pool, and building strategic partnerships with other countries that have REE reserves.
The Long March
KarthigaiRajan calls it the “long march towards technological sovereignty.” It is a fitting phrase. Technological sovereignty is not achieved overnight. It requires sustained investment, patient development of human capital, and strategic foresight that looks beyond the next election cycle.
For India, this means several things. First, increasing funding for mineral exploration to identify and develop domestic REE resources. Second, nurturing domestic manufacturing processes to build refining capacity and reduce dependence on foreign processing. Third, strengthening the human resource pool to build advanced AI applications, not just use them. Fourth, bringing in a comprehensive AI law that addresses not just content labelling but the full spectrum of governance issues, including environmental impact.
It also means thinking differently about infrastructure. Data centres should be sited where renewable energy is abundant. Cooling systems should be designed to minimise water use. Energy efficiency should be a design criterion, not an afterthought.
The Opportunity
The challenges are real, but so is the opportunity. India has a thriving tech sector, a deep pool of engineering talent, and a government that has demonstrated ambition in digital infrastructure. The UPI story shows that India can build world-class digital systems that serve hundreds of millions of users.
AI is the next frontier. But winning the AI race is not just about building better models; it is about building the entire ecosystem that supports them. That ecosystem includes energy, materials, talent, and law. Neglect any of these, and the entire edifice is at risk.
The countries that succeed in AI will not be those with the smartest coders alone. They will be those that can power their data centres sustainably, source their materials securely, and govern their technologies wisely. India has the potential to be among them. But potential must be translated into policy, and policy into action.
The long march begins now.
Q&A: Unpacking the Challenges of AI Infrastructure
Q1: What is the environmental impact of AI systems, and why does it matter?
A: AI systems are computationally intensive, requiring vast amounts of electricity to train and run models. According to a UNEP study, a single ChatGPT query consumes up to 10 times more energy than a standard Google search. This translates into a significant carbon footprint, especially if powered by fossil fuels. Data centres also consume large amounts of water for cooling, which can strain local resources in water-stressed areas. As AI adoption grows, managing its environmental impact becomes crucial for sustainability and for maintaining public trust in the technology.
Q2: How are other countries regulating AI’s environmental impact, and where does India stand?
A: The US has proposed the Artificial Intelligence Environmental Impacts Act of 2024, and the EU’s AI Act includes harmonised rules for AI governance. China has introduced AI-specific regulations, and South Korea enacted a comprehensive AI Basic Act in January 2026. India has taken some steps, such as amending IT Rules in 2026 to mandate AI content labelling, but lacks a comprehensive framework addressing risk assessment, transparency, liability, safety, and environmental standards. Without such a framework, India risks falling behind in sustainable AI development.
Q3: What are Rare Earth Elements, and why are they important for AI?
A: Rare Earth Elements (REEs) are a group of 17 elements critical for manufacturing high-tech products including semiconductors, AI hardware, electric vehicles, and clean energy technologies. China dominates the REE market, controlling 60% of global production and 92% of refining capacity. This concentration gives China significant bargaining power over global supply chains. For India to build a domestic AI and semiconductor industry, securing a stable supply of REEs—through domestic exploration, processing capacity, and strategic partnerships—is essential.
Q4: What does “technological sovereignty” mean in the context of AI?
A: Technological sovereignty refers to a nation’s ability to develop, control, and sustain critical technologies without dependence on foreign suppliers or vulnerabilities to external pressure. For AI, this means having domestic capacity in hardware manufacturing, semiconductor design, rare earth processing, energy infrastructure, and talent development. It also means having a legal and regulatory framework that reflects national values and priorities. Sovereignty is not about autarky but about ensuring that critical dependencies do not become strategic vulnerabilities.
Q5: What concrete steps should India take to build a sustainable AI ecosystem?
A: Several steps are essential. First, increase funding for mineral exploration to identify and develop domestic REE resources. Second, nurture domestic refining and manufacturing processes to build processing capacity. Third, strengthen the human resource pool through education and training in AI and related fields. Fourth, enact a comprehensive AI law that addresses environmental impact, transparency, and ethical standards. Fifth, design data centre infrastructure with sustainability in mind—siting them near renewable energy sources, minimising water use, and prioritising energy efficiency. Sixth, build international partnerships with countries that share strategic interests in resilient supply chains. The “long march” requires coordinated action across multiple fronts.
