AI's vast water, energy, and land use poses global environmental threat

A comprehensive report from UN University (UNU) reveals the escalating environmental costs associated with the rapid expansion of Artificial Intelligence (AI), extending beyond carbon emissions to critically impact water, land, and waste management systems globally. The study, which focuses on the energy consumption of AI infrastructure, including data centers, highlights that current sustainability assessments often overlook these broader environmental footprints. By 2030, the report projects that AI-related water consumption will be equivalent to the basic annual domestic needs of 1.3 billion people. This significant demand, primarily for cooling data centers, exacerbates water scarcity in regions already facing such challenges. For instance, data centers in the US are projected to consume nearly 400 billion gallons of water by 2030, accounting for 7% of the state's total water usage in Texas. Similarly, India's data centers are estimated to consume 150 billion liters of water annually, projected to more than double by the end of the decade, potentially worsening an already critical water scarcity situation in many parts of the country. In terms of energy, global data centers powering AI are expected to consume approximately 945 terawatt-hours (TWh) of electricity annually by 2030. This figure is nearly triple the combined annual electricity use of Pakistan, Bangladesh, and Nigeria, countries home to over 650 million people. The International Energy Agency (IEA) projects that data center electricity consumption could reach 945 TWh by 2030, representing almost 3% of total global electricity consumption. The increasing demand is driven significantly by AI workloads, with accelerated servers, mainly used for AI, projected to grow by 30% annually. This surge in electricity demand poses challenges for power grids, particularly in regions like the United States, where data center electricity consumption is expected to increase by 130% by 2030. The land footprint associated with AI infrastructure is also a growing concern, projected to exceed 14,500 square kilometers by 2030—roughly twice the size of the Jakarta metropolitan area. This includes the land required for data centers, power generation, and associated supply chains. Furthermore, the rapid growth of AI infrastructure contributes to a significant electronic waste (e-waste) challenge. By 2030, AI infrastructure is projected to generate up to 2.5 million tonnes of e-waste annually. Much of this burden often falls on low-income countries lacking adequate disposal capacity. The report criticizes the narrow focus on greenhouse gas emissions from AI model training, arguing that it overlooks other critical environmental costs. Solutions that reduce carbon emissions in one area might increase water consumption or land use elsewhere, highlighting the interconnectedness of these environmental impacts. For example, transitioning to certain renewable energy sources might lower carbon emissions but could concurrently increase water usage and land demand. The UN Secretary-General, António Guterres, has called for greater transparency from AI companies, urging them to measure and publicly disclose their full environmental footprint, including carbon, water, and land usage. He also advocated for all major AI companies to commit to powering their data centers with renewable energy by 2030. The report proposes a framework for a "responsible AI ecosystem" based on principles of transparency, efficiency, equity, lifecycle responsibility, global cooperation, and sustainable use. Governments are urged to integrate AI infrastructure into energy, water, and land-use planning, while companies are encouraged to design resource-minimizing systems, and users are advised to choose lower-impact applications. For India, the implications are particularly significant. The nation is rapidly expanding its AI capabilities and data center infrastructure, driven by initiatives like the IndiaAI Mission. However, this growth occurs in a country already grappling with water scarcity. Data centers, while crucial for economic growth, are water-intensive, and their concentration in water-stressed cities like Bengaluru, Mumbai, and Chennai raises concerns about long-term water security. Weak regulatory frameworks and a lack of transparency regarding water consumption further complicate the situation. Experts emphasize the need for mandatory environmental impact assessments, renewable energy mandates for data centers, and transparent reporting on power and water usage to ensure sustainable AI development in India.

Frequently Asked Questions

Read Full Story on Quick Digest