UK Government Disagrees on AI Data Centre Energy Use

The UK government's energy estimates for AI data centres have changed dramatically, with one department revising its figures up by over 100 times. This highlights a major challenge for the UK's green goals.
By Newsroom | Technology, Environment |

A stark disagreement has emerged between UK government departments regarding the projected energy consumption of artificial intelligence data centres, casting a shadow over the nation's ability to meet its climate commitments. The Department for Science, Innovation and Technology (DSIT) apparently revised its published emissions figures for the AI data centre sector upward by over a hundredfold, shortly after inquiries from The Guardian. This adjustment followed a period where DSIT and the Department for Energy Security and Net Zero (DESNZ), which manages the UK's carbon budget, appear not to have reached a consensus on the scale of this burgeoning energy demand.

This internal disarray highlights a potential struggle for control and presents a concerning picture of corporate influence over government policy, with industry players shaping not only the 'AI value chain' but also the narrative around its environmental impact.

Read More: Capgemini Hiring GCP Data Engineers for API Work in USA

Divergent Estimates Fuel Uncertainty

The immediate friction surfaced when a spokesperson for DSIT deferred questions to DESNZ, indicating a lack of unified understanding or reporting on the issue. The rapid, post-inquiry revision of DSIT's figures suggests an attempt to align or perhaps obscure the true scale of AI's energy appetite.

The exponential growth of AI data centres poses a significant challenge to the UK's net zero targets, with projections indicating a substantial increase in national electricity demand. This surge could intensify competition for grid connections, delaying the deployment of renewable energy schemes and other industrial electrification projects already facing headwinds.

AI Energy Council: A Collaborative Facade or a Genuine Attempt?

In response to these escalating concerns, the UK government launched the 'AI Energy Council' in April 2025. This cross-sectoral body, comprising representatives from 14 key organisations in the energy and technology fields, was established with the stated aim of ensuring the UK's energy infrastructure is "ready for the AI revolution."

Read More: Chernobyl wildlife faces new threats from military activity

The council's mandate includes exploring how sustainable energy sources, such as renewable and nuclear power, can fuel AI advancements. It also seeks to foster collaboration between digital and energy sectors, aligning AI expansion with the government's ambition to become a "global clean energy leader." Industry giants like Google have publicly committed to sustainable infrastructure and joined the council.

However, concerns linger about the practical effectiveness of such initiatives. Reports from February 2026 highlighted that if the UK were to prioritize renewable energy for AI, other industries might be forced to rely more heavily on fossil fuels. This points to a potential zero-sum game where AI's energy needs could inadvertently push other sectors away from clean energy.

Fossil Fuels Loosely Tethered to AI's Rise

The push for rapid AI infrastructure expansion also appears to be, in practice, leaning towards fossil fuels. Reports from August 2025 indicated that data centre developers were exploring connections to the UK's main gas pipelines. Environmental groups have voiced warnings that the current wave of AI data centre projects risks entrenching vast energy and water demands without adequate oversight, potentially locking the UK into unsustainable practices.

Read More: New App Organizes Catalysis Data for Faster Green Tech

Grid Strain and Industry Innovation

The UK's power grid is already showing signs of strain. Limited grid capacity, particularly in areas like London's M25 corridor, has already forced some data centre operators to seek locations elsewhere. This dispersal could spread economic benefits but also poses risks to established data centre hubs.

Despite these pressures, the data centre industry itself is exploring various strategies to mitigate its environmental impact. These include advances in liquid cooling, modular data centres, and AI-driven energy management systems. However, the fundamental increase in demand, largely driven by AI applications requiring higher power densities and more intensive cooling, remains a persistent challenge.

Read More: Georgia Wildfires Destroy 120 Homes Near Brantley County

A Looming Energy and Water Crunch

Expert insights from August 2025 pointed to a looming "twin crunch" for the UK, encompassing both energy and water resources, exacerbated by the growth of AI and data centres. The construction timelines for data centres, typically three to five years, significantly outpace those for large-scale renewable energy projects, which can take much longer to bring online. Similarly, commitments for new nuclear power plants are unlikely to yield results before the mid-2030s, creating a temporal mismatch between burgeoning AI demand and the availability of clean energy solutions.

Frequently Asked Questions

Q: Why are UK government departments disagreeing about AI data centre energy use?
The Department for Science, Innovation and Technology (DSIT) changed its energy use figures for AI data centres significantly after being asked about them. This suggests there wasn't agreement with the Department for Energy Security and Net Zero (DESNZ) on how much energy these centres will need, which is important for meeting climate goals.
Q: How could AI data centres affect the UK's net zero targets?
AI data centres need a lot of electricity. If their energy use is much higher than expected, it could make it harder for the UK to reach its goal of having zero carbon emissions by a certain date. This might also slow down new green energy projects because of high demand.
Q: What is the UK government doing about AI energy needs?
The UK government started the 'AI Energy Council' in April 2025. This group includes people from energy and technology companies. They are trying to make sure the UK's energy system can support AI growth using clean energy like renewables and nuclear power.
Q: Could AI data centres lead to more use of fossil fuels in the UK?
There are worries that building many AI data centres quickly might mean using more fossil fuels, like gas, because new clean energy sources take a long time to build. Some reports from February 2026 suggested that if AI uses renewable energy, other industries might have to use more fossil fuels.
Q: Is the UK's power grid ready for more AI data centres?
The UK's electricity grid is already under pressure. Some areas, like around London, have limited capacity, which has made it difficult for data centres to get connected. This could force them to build in different places, spreading out the impact.
Q: What are data centres doing to reduce their environmental impact?
The data centre industry is looking at new ways to be more eco-friendly. This includes using better cooling systems like liquid cooling, building smaller, modular centres, and using AI to manage energy use more efficiently. However, the overall increase in energy demand from AI is still a big problem.