 "Working with AI has always been resource-intensive, especially when it comes to water and electricity. However, the recent AI boom has caused Microsoft to increase its water consumption by 34%, Google by 22% and Meta by 3% as of 2022")
In recent years, major technology companies such as Microsoft, Google, and Meta have significantly escalated their utilization of water for cooling down data centres, leading to apprehensions regarding the environmental consequences amidst the surge in generative artificial intelligence (AI) advancements, as per a Financial Times report.
With millions of users reliant on their online services, Microsoft, Google, and Meta have witnessed a surge in water consumption.
According to a study from the University of California, Riverside, published in Nature, there is an urgent need to uncover and address the undisclosed water footprint of AI models, particularly in light of escalating freshwater scarcity, prolonged droughts, and ageing water infrastructure.
The concern intensifies as leading tech companies vie to introduce products utilizing generative AI, powered by extensive language models capable of processing vast amounts of data. These models necessitate substantial computing power, driving the establishment of massive server farms that rely on chilled water for cooling systems, some of which evaporates in the process while some can be recycled.
Water, a crucial resource across various forms of energy generation, including hydroelectric power and thermal power stations, has seen notable spikes in consumption by tech giants.
In 2022, Microsoft had consumed about 22 million cubic metres of water, which is equivalent to 22 billion litres of water. That’s enough to fill an Olympic-regulation-sized pool 8800 times over.
Google, on the other hand, consumed about 6-7 million cubic metres or roughly 6-7 billion litres of water. Meta’s consumption was at a modest 2 billion litres.
Impact Shorts
More ShortsIn 2023, the consumption was even more. These companies have set targets to replenish water resources, aiming to return more water to systems like aquifers than they consume by 2030, through initiatives like enhancing irrigation infrastructure and restoring wetland systems.
Experts anticipate that the demand for AI will propel water withdrawal to unprecedented levels, estimated to range between 4.2 billion and 6.6 billion cubic meters by 2027, a volume nearly equivalent to half of the UK’s annual water consumption.
In a recent lawsuit, residents of West Des Moines, Iowa, raised concerns over a data centre cluster consuming a significant portion of the district’s water supply. Shaolei Ren, an associate professor at UC Riverside, likened the water consumption of popular chatbots like ChatGPT to “drinking” a 500ml bottle per 10 to 50 interactions, underscoring the potential water-intensive nature of AI models.
Calls for enhanced transparency and data disclosure from AI firms have emerged, urging detailed breakdowns of water consumption across different computing services. While some companies like OpenAI have expressed commitment to improving efficiencies, others like Google have refrained from commenting on the issue.
Experts emphasize the necessity of comprehensive reporting on the environmental impacts of AI models, particularly amid global concerns over climate change and dwindling water resources. Kate Crawford, a research professor at USC Annenberg, stresses the importance of understanding the true environmental consequences of generative AI tools amidst a climate crisis.