A message reading 'AI artificial intelligence,' a keyboard and robot hands are seen in this illustration
Credit: Reuters Photo
Artificial intelligence has flooded the marketplace. From search engines and social media to office software and complex research tools, AI has rapidly embedded itself in daily life. While much has been debated about AI’s impact on work, society, and productivity, one dimension remains dangerously overlooked: its environmental footprint. What can India learn from the adoption of AI for environmental justice?
AI can arguably help fight climate change with wide-ranging and often promising applications. For instance, the United Nations Environment Programme’s World Environment Situation Room uses AI to analyse multilayered data and inform predictions on CO2 atmospheric concentration, methane emissions, changes in glacier mass, sea levels, and more. These insights help shape credible, data-backed policy decisions in government offices, classrooms, and corporations around the world.
In California, US, AI supports the state’s power grid by identifying weaknesses, predicting failures, and enabling faster recovery after outages. Microsoft, in partnership with Vestas Wind, is using AI to maximise turbine efficiency. As the NITI Aayog document, National Strategy for Artificial Intelligence, highlights, AI technologies can help address unforeseen challenges in agriculture through real-time advisory to farmers and support the development of smart, efficient cities to meet the demands of a rapidly urbanising population.
The applications are manifold, and India’s future remains wide open to this transformative adoption. Yet, this is not the complete story.
AI and its rapid adoption leave behind a trail of overconsumption, waste, and environmental harm, disproportionately borne by those least connected to its rewards.
A 2023 study published in Nature Computational Science estimates that AI could generate between 1.2 million and 5 million metric tonnes of electronic waste by 2030. This is in addition to the 62 million tonnes of e-waste already produced globally, as documented in 2022. AI e-waste is largely composed of CPUs, GPUs, servers, and storage devices that are replaced every two to five years in pursuit of faster, more powerful models (MIT Tech Review).
Conversely, to run these powerful models, there is an ever-growing need for energy. The International Energy Agency reports that the electricity demand from AI is set to more than double by 2030, surpassing the entire electricity consumption of the country of Japan today.
These devices also require the extraction of critical and often hazardous materials such as gold, nickel, copper, silver, lead, mercury, and chromium. These materials are also vital for decarbonisation technologies, making it a fierce competition for these finite resources (Harvard Business Review). Besides, these processes are energy-intensive, contribute to biodiversity loss, and often displace local communities.
Worse still, much of the world’s e-waste ends up in the Global South, including India. Weak environmental regulations and a steady supply of workers extract valuable metals from discarded electronics using dangerous methods like open-air burning and acid leaching—practices that release toxic pollutants. With India’s trajectory geared towards
Viksit Bharat 2047 and increasing use of tech solutions,
it is important to ensure
sustainable growth without further deepening the environmental degradation.
A more deliberate, transparent, and responsible approach to AI development is imperative—one that considers its entire lifecycle, from mineral extraction to disposal.
India, as of June 2025, does not have dedicated AI regulations. NITI Aayog’s National Strategy for Artificial Intelligence, while outlining a wide range of potential applications and government-led missions, remains a strategic vision
rather than a binding roadmap. There is no clear implementation framework, financial backing, or formal adoption by
the government.
To this end, UNEP recommends that governments and technology companies take responsibility for AI’s environmental impact by developing assessment tools, enforcing mandatory disclosures, and promoting efficient design. This includes creating leaner algorithms, reducing energy and water use in data centres, and reusing components. Data centres should run on renewable energy and offset emissions. Crucially, AI regulations must be integrated into broader climate and environmental policies, not treated in isolation.
These suggestions reflect a fundamental truth: AI is not a neutral force. It holds the potential to be both a climate solution and a sustainability crisis, depending on how we govern its growth. As India grows, we must ask: What kind of future are we engineering?
(The writer is pursuing a master’s degree at the London School of Economics)