India’s groundwater crisis is becoming more severe with each passing year, yet it has long remained neglected in public policy and governance. In January 2026, the National Green Tribunal expressed strong dissatisfaction with a report submitted by the Central Ground Water Authority. The report failed to provide key information sought by the tribunal, avoided explaining the criteria used for granting permissions for projects in groundwater-stressed regions, and remained vague on several regulatory issues.
When questioned, the authority argued that framing such criteria falls under the jurisdiction of state-level regulatory bodies. The tribunal then raised a more troubling question: if many states have not even established such bodies, how are permissions being granted at all? The authority was unable to clarify whether audit reports for commercial groundwater extraction are publicly available, whether no-objection certificates are regularly reviewed, or how much environmental compensation has been imposed on violators. The episode reflected not just administrative inefficiency, but a deeper institutional failure in managing one of the country’s most critical natural resources.
According to the 2024 report of the Central Ground Water Board, India’s average groundwater extraction rate has risen to 60.47 percent, up from 59.26 percent in 2023. In 2024 alone, 245.64 billion cubic metres of groundwater were extracted. Agriculture accounted for 87 percent of this usage, or 213.29 billion cubic metres, while domestic use constituted 11 percent and industrial use around two percent. Of the country’s 6,746 groundwater assessment units, 751 have been categorised as “over-exploited,” meaning groundwater extraction exceeds annual recharge. In Punjab, Haryana, Rajasthan, Delhi, and Dadra and Nagar Haveli, extraction levels have crossed 100 percent, indicating that more water is being removed than nature can replenish. This is not merely an alarming trend but a warning of long-term ecological and social instability.
The crisis is not limited to declining groundwater levels. Water quality is also deteriorating rapidly. Citing the 2024 Annual Groundwater Quality Report, the National Green Tribunal noted increasing levels of salinity, fluoride, and heavy metals in groundwater across Haryana, Rajasthan, Gujarat, Punjab, and western Uttar Pradesh.
Even more concerning is the growing presence of uranium in drinking water sources. A nationwide survey conducted in 2019-20 tested 14,377 samples and found uranium concentrations in some locations to be as high as 96 times above the prescribed limit. The World Health Organization recommends a limit of 30 micrograms per litre for uranium in drinking water. In Punjab, 24.2 percent of wells exceeded this limit, while the figure stood at 19.6 percent in Haryana, 11.7 percent in Delhi, and 10.1 percent in Telangana. Despite the seriousness of the issue, the Bureau of Indian Standards has yet to establish a national standard for uranium in drinking water. Across 151 districts in 18 states, millions of people continue to consume contaminated water without adequate awareness or protection.
The most visible and painful face of the water crisis is borne by women and girls. According to United Nations estimates, women in water-stressed regions shoulder nearly 80 percent of the responsibility for collecting water. Carrying pots, buckets, and containers over long distances, they collectively spend an estimated 250 million hours every day securing water for their households. This is time that could otherwise be devoted to education, livelihoods, healthcare, or rest. Water scarcity directly affects girls’ education, as many are forced to miss school to assist with domestic water collection. Insufficient access to clean water also undermines sanitation and maternal health. Yet despite their central role in water management at the household level, women remain largely excluded from decision-making processes, with their participation in water governance remaining disproportionately low. The labour of securing water falls on women, while policy decisions continue to be dominated elsewhere.
Against this backdrop, another less visible but rapidly expanding water challenge is emerging through the growth of Artificial Intelligence. AI is often perceived as a virtual technology existing within mobile apps, chatbots, or image-generation platforms. In reality, the infrastructure behind these services is highly physical. AI systems rely on vast data centres filled with thousands of servers operating continuously. The chips powering these systems consume enormous amounts of electricity and generate significant heat. To prevent overheating and system failure, these facilities require extensive cooling systems, many of which depend heavily on water.
A medium-sized data centre can consume around 110 million gallons of water annually for cooling alone, roughly equivalent to the yearly water requirements of nearly one thousand households. Large data centres may use as much as five million gallons of water per day, placing their annual consumption on par with that of small towns. In several developed countries, the rapid expansion of digital infrastructure has already forced local administrations to allocate increasing amounts of water to industries and data facilities, often competing with domestic needs.
The water footprint of AI extends far beyond cooling systems. It begins with the manufacturing of semiconductor chips, a process that requires extremely pure water for production and cleaning. By the time a single chip reaches a data centre, thousands of gallons of water may already have been consumed during its manufacturing cycle. The electricity required to operate AI systems further intensifies the burden. In many countries, power generation still depends heavily on coal and gas-based thermal plants, which themselves consume large volumes of water.
Reports by global energy and water agencies indicate that in countries such as the United States, China, and France, between 30 and 40 percent of total water use is linked to energy production. Water discharged from cooling systems often returns to rivers and lakes at elevated temperatures, disturbing local ecosystems, while a substantial portion evaporates entirely from the water cycle.
Research on large language models suggests that even a short AI interaction involving a question and response may indirectly consume approximately one bottle of water. This usage remains invisible to most users, but when multiplied across billions of interactions worldwide, the cumulative impact becomes enormous. Studies such as “Making AI Less Thirsty” warn that the indirect water consumption of global AI systems could soon rival the annual water usage of the bottled water industry itself.
This reveals one of the starkest contradictions of the contemporary world. On one side is a girl who walks several kilometres each morning to collect water before attending school, often at the cost of her education and health. On the other side is a data centre consuming massive quantities of water each day to sustain digital processes, including many that may be trivial or unnecessary. Both depend on the same finite freshwater reserves, yet the distribution of benefits, power, and decision-making remains profoundly unequal.
The benefits of AI-driven technologies are largely concentrated among wealthier societies, corporations, and urban middle and upper classes. The burden of water scarcity, however, falls most heavily on communities that already struggle for access to basic resources. This imbalance is not simply a technological or economic issue; it is an ethical and political question. The issue is not whether AI itself is inherently harmful, but rather how it is developed, deployed, and regulated. AI can undoubtedly contribute positively in fields such as healthcare, education, climate research, agriculture, and disaster forecasting. However, such progress cannot be considered equitable if it deepens water insecurity for already vulnerable populations.
This demands greater accountability from governments and technology companies alike. Firms developing AI infrastructure should publicly disclose the total water consumption associated with their operations. Data centres should increasingly rely on recycled or non-potable water for cooling purposes wherever possible. Greater public awareness is also needed regarding the hidden environmental costs of digital technologies. Responsible use of AI does not require rejecting innovation, but it does require recognising that digital systems are tied to material resources with real ecological consequences.
The time has come to ask not only how intelligent AI systems are, but also how much water they consume. If society wishes to ensure that no child is forced to sacrifice education or health in pursuit of water, then the expansion of AI infrastructure cannot be allowed to take precedence over the basic needs of vulnerable communities. The debate around Artificial Intelligence is therefore not only about algorithms or computational power. It is equally about water, justice, and social equality. Water remains a fundamental human necessity, and any future technological development must place that reality at its centre.
---
*Independent writer
Comments
Post a Comment
NOTE: Hateful, abusive comments won't be published. -- Editor