China’s thirsty data centres and the rapid growth of artificial intelligence (AI) could dramatically increase demand on the country’s water resources, according to a new report by think tank China Water Risk.
The Hong Kong-based non-profit estimated the annual water consumption of data centres in China to be around 1.3 billion cubic metres (343 billion gallons) – enough for residential use for 26 million people. By 2030, the figure could reach over 3 billion cubic metres as more data facilities are expected to open, equivalent to the demand of a population greater than that of South Korea.
Data centres consume water directly to prevent information technology equipment from overheating. They also consume water indirectly from coal-powered electricity generation.
The team projected that by the end of the decade, China would have more than 11 million data centre racks, which house servers, cables and other equipment. That is nearly triple the number it had in 2020 of around four million.
The boom in generative AI technology is also expected to add to water demands from the information and communication technology (ICT) industry.
“Along with its huge computational power, AI chatbots drink staggering amounts of water to cool themselves down,” said the report released on Friday.
It pointed to a preprint study – meaning it has not yet been peer-reviewed – from last year by researchers in the United States that showed the large language model GPT-3 consumed 500 millilitres (16.9 fluid ounces) of water for every 10 to 50 responses it generates. That is 20 times more than it takes to produce 50 Google searches.
The think tank’s report noted that chatbots have a growing user base, and Chinese tech giants including Baidu, Tencent and Alibaba launched their own AI services last year, adding to the potential impact on water consumption. Alibaba is the owner of the South China Morning Post.
The report said that if 100 million users had a conversation with ChatGPT, the chatbot “would consume 50,000 cubic metres of water – the same as 20 Olympic-sized swimming pools – whereas the equivalent in Google searches would only consume one swimming pool”.
CT Low, co-author of the report and geospatial risk lead at China Water Risk, said the accelerated development of generative AI would add pressure to the country’s already stressed water resources.
“Almost half of China’s data centre racks are located in water-scarce regions, which are as dry as the Middle East,” he said.
Lead author Debra Tan, director and head of China Water Risk, said improving energy and water efficiency with existing technology were simple solutions to address water risks.
“We recommend corporates and the financial sector to assess fast-evolving climate and water risks and curate cohesive climate strategies to survive them,” Tan said.
“For the ICT sector, the time to tackle water risks is now – we must get on top of these before the explosion of AI,” Tan said.
She said China’s ICT giants are encouraged to become “water neutral” or “water positive” – goals their counterparts in Silicon Valley such as Meta and Google are also pursuing.
A water neutral company is one that offsets its water footprint, while a water positive one replenishes more than its water footprint, ideally after reducing water use as much as possible, according to the report.
Strategies to minimise and offset water use include watershed restoration, improving water efficiency in existing facilities, reusing waste water and collecting rainwater.
Tan added that the Chinese government was taking steps to manage rivers holistically, “from source to sea”, and could be expected to adopt tighter regulations and water usage effectiveness standards for the ICT sector.
According to the report, more than three-quarters of China’s data centre racks are located in the basins of three rivers: the Yellow River, the Yangtze River and the Pearl River. – South China Morning Post