UK Departments Clash Over AI Data Centre Energy Forecasts
AI data centre power demand is quickly becoming a contentious issue within the UK government, as various departments offer conflicting forecasts for the sector’s energy requirements by 2030. As the government sets its sights on both becoming a global AI leader and achieving ambitious net zero targets, a lack of consensus raises crucial questions about the nation’s energy planning and environmental commitments.
Conflicting Visions: Net Zero Versus AI Superpower
The UK government’s dual ambitions are setting up a complex challenge. On one hand, the Department of Science, Innovation and Technology (DSIT) is pushing forward with plans to transform the nation into an ‘AI superpower’, expecting AI data centres to consume as much as 6GW of electricity by the end of the decade. On the other, the Department of Energy Security and Net Zero (DESNZ) offers a drastically lower estimate, projecting less than a tenth of that consumption for the sector.
This stark difference in projected AI data centre power demand spotlights a broader issue: the lack of alignment in long-term planning between departments responsible for technology advancement and those tasked with environmental stewardship. According to Tim Squirrell, head of strategy at the NGO Foxglove, “The government’s cluelessness over the environmental impact of data centres would be laughable, if it weren’t so alarming.”
Environmental Impact and Corporate Influence
Experts remain divided on the root of these conflicting forecasts. Cecilia Rikap, a researcher at University College London, suggests either incompetence or wishful thinking is at play, noting that such discrepancies highlight how large tech companies increasingly influence not only the AI value chain but also government decision-making. DESNZ is charged with delivering the UK’s carbon budget growth and planning for international climate targets, but it appears that AI data centre power demand has not been fully integrated into these critical projections.
Earlier this year, the NGO Foxglove filed a request with DESNZ for an environmental impact assessment regarding AI data centres. The department’s response referred researchers to broader energy use forecasts for Britain’s “commercial services” sector, with no separate data for anticipated data centre growth. Their projections estimate a sector-wide increase of just 528MW between 2025 and 2030, equivalent to the consumption of 1.7 million homes—significantly less than what DSIT predicts for AI data centres alone.
The UK Compute Roadmap and Growth Zones
DSIT’s “UK Compute Roadmap,” published in 2025, outlines a bold vision to expand the national compute ecosystem, including the creation of AI growth zones across the country. Each zone would require at least 500MW of power, nearly matching DESNZ’s entire forecasted energy increase for the commercial sector. This reveals a substantial mismatch in departmental planning over AI data centre power demand and its implications for the UK’s energy grid and carbon emissions.
Rapid Revisions and Emissions Concerns
The lack of clarity became even more evident when, following media inquiries, DSIT revised its public figures for the AI data centre sector’s total emissions, increasing them more than a hundredfold. Initial estimates placed additional AI computing capacity emissions at between 0.025m and 0.142m tonnes of carbon equivalent (MtCO₂). After scrutiny, DSIT updated these numbers, stating that cumulative emissions over the next decade could range from 34 to 123 MtCO₂—representing 0.9-3.4% of the UK’s projected emissions.
DSIT noted that if the nation’s grid decarbonisation efforts succeed, emissions from AI data centres could be kept at the lower end of this estimate. Meanwhile, DESNZ insists that data centre emissions are included in their modelling, and the AI Energy Council is actively seeking to attract investment and support clean energy solutions for data centres. Details are expected to emerge with the release of the upcoming Carbon Budget 7 this summer.
Implications and the Path Forward
The dispute over AI data centre power demand is more than a bureaucratic quibble; it underscores the urgent need for cohesive and transparent planning as the UK attempts to balance digital innovation with climate responsibility. As AI infrastructure expands, so too does its potential impact on national energy grids and the environment. Clear, unified data is essential for policymakers, businesses, and the public to make informed decisions about the future of AI in the UK.
With both environmental groups and technology advocates watching closely, the government’s next steps will be critical in determining whether the UK can achieve its twin goals of technological leadership and environmental sustainability.
This article is inspired by content from Original Source. It has been rephrased for originality. Images are credited to the original source.