AI Data Centres Fuel Unseen Water Crisis as UK Faces Shortage Predictions
AI data centres in England consume large amounts of water for cooling, complicating water shortage forecasts. Without reporting regulations, future shortages may be more severe than expected.
AI Challenges Water Shortage Predictions in England
The UK Environment Agency warns that advances in artificial intelligence are creating uncertainty around future water shortages in England. Data centres supporting AI operations use large volumes of fresh mains water for cooling, but current regulations do not require these centres to report their water consumption. This lack of data limits the government’s ability to forecast water demand accurately.
The Environment Agency explained that its five-year water deficit projections were difficult to complete this year because AI infrastructure growth was not factored into previous models. As a result, future water shortages could be more severe than anticipated.
Water Use in AI Data Centres
Data centres cool servers and protect internal components like pipes, pumps, and heat exchangers from contaminants using fresh water. A Cornell University study estimates global AI water use could reach 6.6 billion cubic metres by 2027—nearly two-thirds of England’s current annual water consumption.
Looking ahead, England could face a public water supply shortfall of 5 billion litres per day by 2055. Additional demands from agriculture, power generation, and emerging technologies could increase this deficit by another 1 billion litres daily.
Alan Lovell, chair of the Environment Agency, highlighted the stakes: “The nation’s water resources are under huge and steadily increasing pressure. This deficit threatens not only tap water availability but also economic growth and food production.”
Expansion of AI Infrastructure and Water Demand
The UK government’s AI Opportunities Action Plan calls for increased computational infrastructure, including new data centres to train AI models. While the government expects not to own all the required computing resources, it recognizes a need for significant expansion.
Prime Minister Keir Starmer announced plans to accelerate AI capacity growth by easing planning approvals and establishing AI Growth Zones (AIGZs). The government encourages regional cooperation to create these zones to support local needs.
Calls for Sustainable AI Development
The UK’s National Engineering Policy Centre recently urged government investment in sustainable AI practices. Their report emphasized the importance of transparency from private firms regarding electricity and water consumption in data centres.
Major technology companies like Google and Microsoft have reported rising water consumption in their data centres since 2020, underscoring the need to monitor and manage environmental impacts.
For government professionals working with AI infrastructure, understanding water usage implications is critical for sustainable planning. Further training on AI’s operational impacts can be found at Complete AI Training.
