The rapid expansion of artificial intelligence is creating growing pressure on water resources, forcing governments to reconsider how water is managed, allocated, and regulated. While public debate around AI often focuses on energy consumption, the water required to support data centers has become an increasingly important concern. As AI infrastructure expands, competition over access to water is intensifying, particularly in regions already facing environmental stress and climate-related challenges.
Data centers depend on large amounts of electricity, and the heat generated by computing systems must be controlled through cooling processes that frequently require significant water use. Although companies have improved efficiency and introduced some water-saving technologies, the demand continues to grow alongside AI development. As a result, technological progress is increasing pressure on resources that many communities already struggle to access sustainably.
This tension is especially visible at the local level. Governments promoting AI as a driver of economic growth are simultaneously supporting infrastructure projects that may increase water demand in vulnerable areas. In the United Kingdom, for example, plans for large-scale AI expansion coincide with concerns that many future data centers will be located in regions expected to face water stress within the coming decades. These decisions reflect a broader political calculation that prioritizes long-term economic growth, even as environmental risks continue to accumulate.
The challenge extends far beyond the facilities themselves. AI systems depend on global supply chains linked to mining, manufacturing, and semiconductor production, all of which require substantial amounts of water. This creates what researchers describe as a largely hidden water footprint. Water consumption therefore occurs not only where data centers operate, but also across international networks that supply the materials and technologies needed to sustain the digital economy.
Semiconductor production illustrates these vulnerabilities clearly. Taiwan produces most of the world’s advanced chips, yet the industry relies heavily on ultrapure water and operates in a region increasingly exposed to drought risks. Climate change has made rainfall patterns less predictable, while geopolitical tensions involving China continue to raise concerns about supply-chain stability. Water insecurity and geopolitical conflicts are therefore becoming closely interconnected.
The discussion also challenges the way governments traditionally evaluate infrastructure projects. Measuring only the direct water use of data centers provides an incomplete picture because it ignores the broader environmental costs embedded throughout supply chains. Focusing exclusively on local consumption can obscure how resource pressures are transferred across borders and concentrated in other communities.
Alongside, questions of fairness increasingly shape the debate. Access to water is recognized internationally as a human right, yet the growth of AI infrastructure creates difficult dilemmas about allocation priorities when water resources are scarce. Expanding digital industries may generate economic benefits, but those benefits are not always experienced by the communities that bear the environmental costs. This creates ethical disputes between technological development, commercial interests, and equitable access to essential resources.
Greater transparency is presented as a necessary step forward. Current estimates of AI-related water consumption remain incomplete, making it difficult to assess risks accurately or establish effective regulations. The broader concern is not simply that AI uses water, but that its expansion is exposing weaknesses in existing approaches to resource management. Technological growth is accelerating within a world already facing mounting environmental problems and water scarcity. Without stronger oversight and long-term planning, the pursuit of digital innovation may deepen existing inequalities and place additional strain on resources that many regions can no longer afford to treat as unlimited.
Reference: Morris-Iveson, L. (2026, June 1). AI water usage requires governments to rethink their approach to water. https://www.chathamhouse.org/2026/06/ai-water-usage-requires-governments-rethink-their-approach-water