Next Generation Data Centres: Scotland’s Global Opportunity
Scotland can become a global leader in sustainable AI data centres by harnessing its renewables, cool climate, and engineering expertise to tackle energy and cooling challenges.
The explosive growth of AI is reshaping the digital infrastructure landscape.
Hyperscale data centres—facilities with massive processing capacity for training and running large language models and other AI workloads—are at the heart of this transformation.
Globally, data centre electricity consumption stood at around 415–460 TWh in 2024 (roughly 1.5% of global electricity use) and is projected to roughly double by 2030, reaching 945–1,000+ TWh, with AI driving much of the surge through accelerated servers growing at ~30% annually.
Power demand from AI data centres in the US alone could rise more than thirtyfold by 2035 to 123 GW. Individual campuses are scaling to gigawatt levels, creating concentrated, 24/7 loads that strain grids, supply chains, and resources.
Scotland is uniquely positioned to capture a significant share of this market—not just as a host for facilities, but as a global leader in sustainable, next-generation data centre expertise.
The Core Challenges of Large-Scale AI Data Centres
Energy Supply and Grid Integration: AI workloads demand high power density. Traditional data centres might use 5–20 kW per rack; AI-optimised ones push toward 50 kW+ and far higher in some cases. Cooling can account for 30–40% of total energy use, and the largest new facilities require hundreds of MW to GW of reliable power.
Globally, grid constraints are the primary bottleneck. Interconnection delays, transmission limitations, and the need for constant baseload power complicate deployment. In high-demand areas, this has led to higher energy costs, reliance on fossil backups during low-renewable periods, and public backlash over impacts on residential bills and reliability.
Cooling and Water Use: High-performance computing generates intense heat. Evaporative cooling is efficient but water-intensive; a single large facility can consume millions of gallons annually. In Scotland, existing (smaller) data centres already use enough tap water yearly to fill 27 million half-litre bottles, with usage quadrupling since 2021. Scaling to hyperscale AI facilities without innovation risks local strain, even in a wet climate.
Sustainability and Environmental Impact: Hyperscalers face ESG pressure to achieve climate neutrality. Co-location with renewables helps, but intermittency requires storage, flexible demand response, or backups. Heat waste offers opportunities for district heating, while biodiversity, land use, and construction emissions add complexity.
Other Hurdles: Supply chain constraints for chips and equipment, skilled labour shortages, planning delays, and community concerns over noise, visual impact, and resource competition.
These challenges create barriers in traditional hubs (e.g., parts of the US, Ireland, Southeast England) but open doors for locations that can address them innovatively.
Scotland’s Competitive Advantages
Scotland combines abundant renewables, a cool climate, strategic connectivity, and policy momentum:
- Renewable Energy Leadership: Scotland generates more renewable electricity than it consumes (often >100% of domestic needs from sources like wind). It has ~15+ GW installed capacity with tens of GW in the pipeline. Offshore wind, hydro, and emerging tidal resources provide a strong foundation. Data centres can absorb curtailed or excess renewable output, reducing constraint costs (hundreds of millions paid annually) and supporting grid stability. Up to 1–2 GW of flexible data centre demand could be strategically placed to maximise system benefits.
- Cool Climate and Efficiency: Lower ambient temperatures reduce cooling energy needs compared to hotter regions, improving Power Usage Effectiveness (PUE).
- Land, Connectivity, and Cost: Cheaper land than in Slough or Dublin, subsea fibre cables, and regeneration sites (e.g., Ravenscraig) offer advantages. Talent and operational costs are competitive.
- Policy Support: National Planning Framework and green data centre ambitions favour renewable-powered facilities. Proposals like North Lanarkshire’s AI Growth Zone and projects such as ILI Group’s “The Stoics” (a multi-site hyperscale network) signal momentum. Planning policy supports AI data centres that leverage Scottish renewables.
A pipeline of 17+ hyperscale projects (200 MW to 1 GW+) exists, though total proposed demand (several GW) exceeds current peak needs and requires careful management.
Cultivating Global Expertise: Beyond Hosting
Scotland’s real opportunity lies in moving up the value chain—from hosting facilities to exporting world-class expertise in sustainable data centre design, operation, and integration.
- Energy System Integration: Leverage oil & gas sector skills in engineering, project delivery, and complex energy systems for co-located renewables, battery storage (BESS), and demand flexibility. Data centres as flexible loads can help balance the grid.
- Sustainable Cooling and Heat Recovery: Innovate in liquid cooling, free-air cooling, water reuse, and waste heat utilisation for district heating or industry. Scotland could pioneer low-water, circular designs.
- Skills and Workforce: Transition expertise from traditional energy sectors. Invest in training for critical facilities engineering, AI infrastructure management, cybersecurity, and data centre-specific trades. Partnerships with universities and colleges can build a pipeline, creating high-value jobs in supply chains.
- R&D and Innovation: Build on strengths in AI (Scottish AI Alliance), marine energy (EMEC), and digital connectivity. Export solutions in green data centre standards, modular construction, and net-zero operations.
Globally, the addressable market is enormous. Data centre infrastructure spending could exceed $1 trillion annually by 2030. Demand for sustainable solutions will grow as more regions face power and water constraints. Scotland can position itself as the “Nordics of the UK” or better—a European (and transatlantic) hub for green AI infrastructure.
Strategic Recommendations
- Streamline Planning with Standards: Fast-track approvals for projects meeting strict green criteria (renewable matching, water efficiency, heat recovery, biodiversity net gain).
- Grid and Spatial Planning: Use Strategic Energy Planning to optimise locations near renewables and transmission. Accelerate targeted reinforcements and co-location.
- Skills Investment: Scale apprenticeships, university programmes, and cross-sector retraining. Attract global talent while upskilling locals.
- Innovation Ecosystem: Fund R&D in advanced cooling, AI-optimised efficiency, and circular economy models. Support pilot projects demonstrating exportable tech.
- Community and Economic Benefits: Mandate local content, skills funds, and shared infrastructure (e.g., heat networks) to build public support.
Conclusion
The AI boom’s energy and resource intensity poses real challenges, but it also creates a once-in-a-generation opportunity. Scotland’s renewable riches, temperate climate, engineering heritage, and forward-looking policy give it a genuine chance to lead in next-generation data centres. By addressing challenges head-on and investing in expertise, Scotland can attract investment, create skilled jobs, support its net-zero goals, and export solutions to a world hungry for sustainable digital infrastructure.
The window is open. With decisive action, Scotland can turn data centres from a potential strain into a cornerstone of its green industrial future and a global export success story.
