In the digital age, data centers and artificial intelligence (AI) technologies such as ChatGPT have become indispensable. They power everything from online services and cloud computing to advanced analytics and interactive AI platforms. However, these technological achievements come with significant environmental costs, particularly when it comes to water consumption.
The water footprint of AI and data centers: room for improvement
A recent study from 2025, published in the journal Patterns by Alex de Vries-Gao, specifically examined the environmental impact of AI systems. The study shows that AI systems are estimated to have consumed around 765 billion liters of water in 2025, an amount greater than the total global consumption of bottled water in that same year.
Asking 10 to 50 questions to ChatGPT consumes roughly the same amount of water as a 500 ml bottle, depending on the location and timing of the request.
Servers and other IT equipment in data centers must be cooled efficiently, which is why most companies opt for water-based cooling. Both traditional air conditioning and more advanced liquid cooling systems rely heavily on water. In addition, a significant amount of water is consumed indirectly through the production of electricity needed to run AI systems. Recent research indicates that this indirect water use, such as for power generation, can be up to four times higher than the direct water consumption for server cooling alone.
With the rise of new AI tools and the expansion of data centers, challenges related to water consumption are expected to intensify. For example, a search query using generative AI is estimated to consume ten times more energy than a regular internet search; for complex queries, this figure can easily rise to fifty times more. Within a few years, large AI systems are likely to consume as much energy as entire countries.
The impact of artificial intelligence on water: the need for transparency
Tech companies that use AI acknowledge the issue of water consumption and have established goals and initiatives to reduce it, such as Microsoft’s Datacenter Community Pledge.
However, obtaining precise and comprehensive data on the environmental effects of AI remains extremely difficult. The total global costs of generative AI are closely guarded corporate secrets. This highlights the urgent need for greater transparency in the sector.
It is only a matter of time before concrete laws are introduced that force companies to disclose more data about their environmental impact. Regular environmental audits by independent bodies could, for example, contribute to greater transparency and compliance with standards.
As data centers and AI technologies like ChatGPT continue to spread across the world, it is essential to recognize and address their environmental costs—particularly their impact on water resources and energy consumption.
Future-oriented companies must think strategically about efficient water management. Water reuse can offer part of the solution.
Ekopak Sustainable Water specializes in circular water solutions and is ready to tackle the challenges of today and tomorrow.
Sources
Generative AI’s environmental costs are soaring, and mostly secret | Nature