Indonesian Political, Business & Finance News

How AI Affects the Earth's Water Supply

| Source: Antara (Indonesian) Translated from Indonesian | Technology
How AI Affects the Earth's Water Supply
Image: Antara (Indonesian)

Jakarta (ANTARA) - Artificial intelligence (AI) is now increasingly woven into everyday life. From searching for information, generating images and assisting with work, to serving as a place to pour out one’s heart. The technology relies on vast numbers of computers housed in data centres.

Behind this convenience lies an environmental impact that is perhaps not widely known: water usage.

So, does AI really reduce the amount of water on Earth?

Simply put, AI does not make water vanish from the Earth. Water continues to follow the hydrological cycle - evaporating, forming clouds, falling as rain, flowing into rivers and seas, and then evaporating again.

The problem is not the loss of water from the planet, but the reduced availability of fresh water usable by humans at particular locations and times.

AI requires computers with high processing power. These computers generate large amounts of heat when in use, particularly when training or running large AI models.

To prevent the equipment from overheating, data centres require cooling systems. Some facilities use water to help dissipate that heat.

In other words, the greater the demand for AI computing, the greater the demand for energy and cooling, which has the potential to increase water usage.

One of the easiest ways to understand this issue is to imagine computers on a very large scale.

If a computer is used continuously, its temperature rises. Data centres running thousands to millions of computing devices face the same problem, but on a far larger scale.

Certain cooling systems use water that then evaporates to carry heat out of the facility. That evaporated water does not immediately become available again to communities in the same area.

Research on AI’s water footprint shows that water usage can come from two main sources: direct use for data centre cooling, and indirect use from the power plants supplying energy to the data centres.

This means AI’s water footprint does not only come from water visibly used inside data centre buildings.

Electricity also has a relationship with water usage

AI requires large amounts of electricity. Data centres running AI services must operate around the clock and demand high computing power.

The problem is that power generation itself can also require water. The scale of usage depends on the generation technology and the energy source used.

Therefore, when someone uses an AI service, its water footprint comes not only from the server cooling process. There is also water used indirectly along the electricity supply chain.

Data from the International Energy Agency (IEA), cited in a study on the water footprint of data centres, shows that indirect water consumption from electricity generation can be a larger component than direct cooling use. In 2023, total global water consumption by data centres was estimated at around 560 billion litres, with roughly 373 billion litres coming from indirect use for power generation and around 140 billion litres used directly at facilities.

These figures cover data centres as a whole, not only facilities used for AI. However, the development of AI is one of the factors driving the growth in computing and data centre demand.

So, does every question posed to an AI consume a lot of water?

The answer is not that simple.

Several studies have attempted to calculate the water footprint of AI activity, but their results vary. Reasons include the type of model used, the length of the request, the data centre location, the weather, the cooling system and the source of electricity used.

One study led by researchers from the University of California, Riverside, for example, estimated that training a large AI model such as GPT-3 at certain data centres could directly evaporate around 700,000 litres of fresh water. The study also estimated that global AI demand could reach 4.2 to 6.6 billion cubic metres of water usage by 2027 if growth continues.

However, these figures are estimates based on particular methods and assumptions. So water usage figures for a single question or conversation cannot be treated as definitive figures applicable to all AI services.

The impact is felt more acutely in certain regions

The biggest issue is not merely how much water AI uses globally.

Equally important is where that water is used.

Recent research on AI’s water footprint also shows that many data centres are located, or potentially being built, in regions facing pressure on water resources. For this reason, the location of data centre construction is an important factor in assessing the environmental impact of AI.

Thus, a single data centre facility may have a relatively small impact on global water availability, but it could pose a serious problem for the water system of surrounding communities if its water needs increase at certain times.

AI can also be made more water-efficient

The good news is that AI usage does not always have to be synonymous with large-scale water consumption.

Data centre operators can use more efficient cooling technologies, including air-based cooling systems, liquid cooling, the use of recycled water, and closed-loop cooling systems that allow water to be reused.

Some modern data centre designs have even been developed with systems that reduce or eliminate the need for water for direct cooling. However, these options come with other consequences, such as potentially higher electricity demand for certain cooling systems.

Therefore, making AI more environmentally friendly requires more than just reducing water usage inside data centres. The source of electricity, equipment efficiency, facility location and cooling technology all need to be taken into account.

More from 8 September 2026 →

Cite this article
Source: jawawa.id (https://jawawa.id/newsitem/how-ai-affects-the-earths-water-supply-1788858194)
Originally published by Antara (Indonesian) (https://www.antaranews.com/berita/5731308/bagaimana-ai-mempengaruhi-jumlah-air-di-bumi), translated from Indonesian
Print