Labour’s first AI growth zone will be created near the UK’s first new reservoir in 30 years, raising fears that AI will put “severe pressure” on the region’s water supply.
Keir Starmer announced on Monday that he would significantly increase AI capacity and reduce planning restrictions on companies that want to build data centers by creating “growth zones” with fewer restrictions.
The first would be in Culham, Oxfordshire, just seven miles from the reservoir planned by Thames Water in Abingdon, which was supposed to provide water to people in severely water-stressed south-east England. This is the area most at risk of running out of water in the country, according to the Environment Agency. Oxfordshire has faced particular problems, with areas relying on bottled water during heatwaves.
AI data centers use a large amount of water, as their servers generate heat. To prevent computer systems from overheating and shutting down, the centers use cooling towers and outdoor air systems, both of which require clean, fresh water. Amnesty International It consumes between 1.8 and 12 liters of water per kilowatt hour of energy Usage across Microsoft’s global data centers. one He studies It is estimated that global AI could account for up to 6.6 billion cubic meters of water use by 2027 – equivalent to nearly Two-thirds of annual consumption in England.
Even without a significant increase in AI data centres, by 2050, England will face a shortfall of nearly 5 billion liters of water per day between available sustainable supplies and expected demand. This represents more than a third of the 14 billion liters of water currently allocated to public supplies. The Southeast faces a potential deficit of more than 2.5 billion litres day in the same period.
AI could undo the gains companies have made in reducing their water consumption; The government seeks to reduce non-domestic (business) consumption by 9% by 2037-38 compared to 2019-20 levels, and businesses are currently on track to achieve a 6.1% reduction.
Adrian Ramsay, MP and co-leader of the Green Party, said: “While communities will face heatwaves, droughts and water shortages over the coming decades, this strategy prevents us from pumping huge amounts of water into AI data centres. One estimate said that AI-related infrastructure could It will soon consume six times as much as Denmark, a country of 6 million people. What does this mean for residents in water-stressed communities like Culham in Oxfordshire?
Professor Hannah Cloke, from the University of Reading, added: “Southern England is already under severe pressure on water resources, which is becoming more acute as we build more homes and look to grow high-tech industries, all of which require more energy and water. We know that with Climate Change In the UK, there will be a more variable supply of water from the sky with hotter and drier summers which will exacerbate the demand on the cooling systems that data centers need.
AI data centers also use a lot of power. The International Energy Agency estimates that data centers’ total electricity consumption could occur Doubling from 2022 levels to 1,000 TWh in 2026approximately the level of electricity demand in Japan. Artificial intelligence will lead to the use of data centers 4.5% of global energy generation By 2030, according to calculations by research firm SemiAnalogy. In January, Amazon, the world The largest corporate buyer Renewable Energy, announced this Bought more than half Offshore wind farm production in Scotland.
A spokesman for Thames Water said it was not a legal adviser to data centres, meaning the company did not need to be asked about water availability. Timms intends to ask the government whether sustainable sources, such as surface water, could be used to cool data centres, rather than water directly from a reservoir.
After promoting the newsletter
The government is working to establish a dedicated energy council for artificial intelligence, headed by the Ministers of Science and Energy. The government said it intends to work with “energy companies to understand energy requirements and challenges” that would promote the development of artificial intelligence. It hopes that small nuclear reactors can power data centers.
Professor Gopal Ramshorn, Professor of Artificial Intelligence at the University of Southampton, said: “The expansion of AI has been a concern for National Grid, but the speed at which demand for AI computing is growing has caught everyone by surprise. Unless we balance the above trade-offs right, with the right policies, All our cheap, green energy will be used by big tech companies, pricing out households already suffering from energy poverty.
Some scientists have said there are environmental opportunities from artificial intelligence. Dr Sean Fitzgerald, Teaching Fellow at the University of Cambridge, said: “The increased energy use of AI should not be considered in isolation, but should be coupled with the potential energy reduction that intelligent control of our energy systems could enable. There are amazing opportunities to achieve more of what we have.” Already in our energy system, AI-integrated control has the potential to unleash these opportunities.
A government spokesperson said: “We recognize that data centers face sustainability challenges such as energy demand and water use. Many newer data centers are already addressing these issues, using advanced cooling systems that significantly reduce water consumption.
“Through the AI Energy Council, we will also build on this progress by exploring bold, clean energy solutions – from next-generation renewables to small modular reactors – to ensure our AI ambitions are aligned with the UK’s net zero goals. We also We are delivering £104 billion in water infrastructure over the next five years, which includes supporting the resilience of water supplies in and around data centres.