The UK’s data centre sector could inject an additional £44 billion into the economy by 2035, according to industry body techUK, as demand for the digital infrastructure that powers artificial intelligence, cloud computing and advanced analytics accelerates. The industry currently contributes around £4.7 billion a year and, with supportive investment and policy, could create more than 40,000 jobs over the next decade.
Economic opportunity and government backing
The government has already highlighted more than £28 billion of planned investment into AI infrastructure projects and data centres across the UK, Parliament’s research service notes. In September 2024, data centres were designated as Critical National Infrastructure, a move that underscored their strategic importance. A series of AI Growth Zones have since been announced to fast-track AI-focused facilities. The first such zone was established at Culham, with potential sites identified in Scotland, Wales, the North East, North West England, North Lincolnshire and Oxfordshire. The government’s UK Compute Roadmap aims for these zones to serve between 500 megawatts and 1 gigawatt of demand by 2030, forecasting a 5.7-times rise in national compute demand by 2035.
Major technology companies have committed significant sums. Amazon Web Services has pledged £8 billion over five years, Microsoft announced £2.5 billion over three years, CoreWeave has invested £1 billion, Blackstone is planning a £10 billion AI campus in Blyth, and Google has announced a £5 billion data centre programme. Collectively, more than £28 billion of planned investment has been cited by the government.
Strengths: connectivity and existing capacity
The UK is already one of Europe’s leading data centre markets. With more than 500 operational facilities, it ranks second in Europe behind Germany and third globally. London remains a critical connectivity hub, benefiting from extensive fibre infrastructure, subsea cable connections and proximity to major financial institutions and technology companies. John Booth, CEO of Teledata UK, said the country’s role as a gateway between North America and Europe gives it a valuable connectivity advantage as AI and cloud demand continues to grow. techUK has similarly argued that Britain’s combination of digital expertise, connectivity and investor interest positions it well to capture a significant share of future AI infrastructure investment.
Nearly 100 new data centre facilities are in the pipeline, representing a projected 20% increase in inventory by 2030. Oxford Economics estimates that new data centres will add around 6.2 GW of IT power capacity by 2030, more than doubling the UK’s current 2.9 GW. The UK data centre market is projected to reach USD 22.65 billion by 2030, with some forecasts suggesting the AI-focused segment alone could hit USD 48 billion by 2033.
The critical challenge: securing electricity and grid capacity
Industry experts increasingly believe that electricity, not land or computing hardware, will determine which nations succeed in the AI race. The International Energy Agency reports that global electricity consumption from data centres is expected to double by 2030, driven primarily by AI workloads, with demand growing at around 15% annually. In 2024 global data centre power consumption was estimated at 415 TWh; the IEA projects it could reach 945 TWh by 2030, accounting for about 2.5–3% of total global electricity consumption.
Within the UK, the trend is stark. Analysis from Electric Insights suggests AI-related infrastructure increased Britain’s electricity demand by 1.7% during 2025, and data centre demand is forecast to quadruple by 2030. Oxford Economics forecasts UK data centre electricity consumption will grow from 5 TWh in 2023 to 26.2 TWh by 2030, representing 8.8% of total UK electricity demand. The National Energy System Operator estimates data centres could drive up to 71 TWh of additional grid demand over the next 25 years.
The scale of the strain on the grid is immense. Large data centres can require power equivalent to tens of thousands of homes, with some proposed AI-focused facilities demanding hundreds of megawatts each. Legal and infrastructure specialists at CMS estimate that around 140 proposed UK data centre developments could collectively require approximately 50 GW of electricity — for context, Britain’s current peak electricity demand is roughly 45 GW.
Demand connection requests have surged. Between November 2024 and June 2025, total contracted offers in the demand queue rocketed from 41 GW to 125 GW, according to NESO figures. Grid connection delays of up to eight years have become commonplace, with some developers offered connection dates in 2037 and beyond. Tim Harper, founder of technology consultancy G2O Water Technologies, summed up the issue by saying “Power is the new planning permission,” arguing that access to reliable, low-carbon electricity is becoming the primary constraint on future data centre development.
The government has responded. A consultation launched earlier this year, “Accelerating electricity network connections for strategic demand,” aims to prioritise strategic demand projects and reform how data centre connections are managed. Reforms include creating new connection mechanisms to reallocate capacity and reserve specific connection points for strategically important projects. AI Growth Zones will also benefit from accelerated grid connections, alongside reduced electricity costs: from April 2027, discounts of up to £24 per MWh in Scotland, £16 per MWh in Cumbria and £14 per MWh in the North East will apply.
Beyond grid access, developers are exploring alternative power delivery such as co-located generation assets and private offtake arrangements. The UK’s surge in renewable power — wind and solar — could meet growing demand, but requires reliable energy storage to address intermittency. Nuclear power, including small modular reactors, is also being examined; the pipeline of conditional offtake agreements between data centre operators and SMR projects has grown significantly.
Space and planning: land is less of an obstacle
Physical land availability is generally viewed as a less significant obstacle than power. While London and the South East remain attractive due to connectivity and proximity to customers, developers are increasingly exploring regions with available land and stronger power infrastructure. The AI Growth Zones are expected to encourage development outside traditional technology hubs, spreading investment across the country.
However, planning delays remain a concern. Several major projects have faced lengthy approval processes. Data centre planning applications are subject to an average delay of 490 days, often due to objections related to community engagement, design, infrastructure and energy use. The government has introduced reforms: following December 2024 changes to the National Planning Policy Framework in England, local authorities must now consider data centre needs when setting policies and deciding applications. Data centres can also opt into the Nationally Significant Infrastructure Projects regime, where decisions are made by the Secretary of State. techUK has called for a dedicated National Policy Statement for data centres to further streamline planning. The creation of a government Data Infrastructure Forum has formalised a structured relationship between industry and the state.
The cost of energy: a competitive disadvantage
Even where power is available, cost remains a major issue. The UK faces some of the highest industrial electricity prices among G7 economies. Electricity prices for UK data centres are roughly four times those in the United States and 46% above the median of 31 IEA countries. Recent research from PwC warned that persistently high energy costs could undermine Britain’s ability to attract future investment across several sectors, including digital infrastructure. For hyperscale data centre operators investing billions of pounds, long-term energy pricing is becoming just as important as access to land and fibre connectivity.
AI-focused facilities present additional challenges. The shift towards high-density, AI-first campuses requires more power, generates more heat and necessitates GPU-powered racks. Water usage is also a consideration: techUK reported that 51% of surveyed data centre facilities used waterless cooling, and 64% use less than 10,000 m³ of water per year. Across the sector, data centres consume over 560 billion litres of water annually globally, though many are adopting waterless cooling methods.
The consensus among industry bodies, policymakers and infrastructure experts is that the UK possesses many of the ingredients needed to succeed — world-class connectivity, a strong technology ecosystem, significant investor interest and growing government support. Yet the country’s ability to become a genuine global AI infrastructure powerhouse may ultimately depend on how quickly it can solve its energy challenge. As data centres increasingly become the factories of the digital economy, access to electricity could prove every bit as important as access to talent, capital and innovation.
