The Influence of AI on Data Centre Design: From the Power Grid to Environmental Demands 

Over the past few years, the UK has witnessed exponential growth in AI, and the industry continues to expand. The UK AI market is expected to grow to £1 trillion by 2035 (International Trade Administration, 2023) and with generative AI growing in popularity, we are also witnessing an increased demand for AI Data Centres. So, how do we design for now, and the future?  

The rise of AI and its impact on data centers means increased power demands, cooling challenges, and the need for resilient infrastructure, there is also currently an international demand for AI-ready data centres, and these spaces require the computing power to train and develop Generative AI models in a safe and resilient environment.  

AI’s Growing Demand on Data Centres 

Global spending on AI infrastructure is expected to double between 2023 and 2026, highlighting the rapid adoption of AI across various industries. Recently, OpenAI secured a $40 billion funding round, valuing the company at $300 billion. More locally, Westminster unveiled an ambitious Artificial Intelligence Opportunities Action Plan, aiming to establish the nation as a global leader in AI technology. This initiative has spurred over £25 billion in private sector investments into new UK data centers since July 2024, further emphasising AI’s swift growth and increasing integration into everyday life. 

However, this expansion comes with significant engineering challenges. Unlike traditional computing, AI workloads rely on GPU-powered racks instead of CPU-based servers. These GPUs consume far more power and generate substantially more heat, demanding enhanced power distribution and advanced cooling solutions to maintain efficiency and operational stability in data centres. 

Because AI servers generate more heat, cooling is a major design challenge. Higher-density AI server racks demand modern solutions, which is why we are seeing the adoption of liquid cooling, direct-to-chip cooling, and immersion cooling. These cooling techniques are becoming essential for AI-ready data centres, ensuring operational efficiency while still reducing energy waste. With this increased cooling demand, the opportunities associated with locating data centres in cooler climates could lead to significant energy savings when coupled with elevated water temperatures. 

Beyond cooling, the nature of AI workloads also means resilient energy strategies are essential to prevent downtime. Since data centres operate 24/7 365 days a year, power reliability, redundancy, and sustainable energy solutions must be prioritised to support AI Data Centre growth.  

Adapting to AI & Cybersecurity Demands 

This shift into the ‘AI evolution’ demands continuous adaptation from not only our teams across MEP & Sustainability, but also across disciplines and in other fields, as we need to meet the growing energy needs and heightened cybersecurity requirements. The rise of AI also introduces new cybersecurity vulnerabilities, making it essential for data centres to integrate both physical and digital security measures. 

At Wallace Whittle, our experience in MEP and Sustainability across sectors allows us to apply solutions and cross-disciplinary knowledge to our data centre projects. As technology advances and energy systems grow more complex, our designs continually evolve to meet these changing demands. 

We are also well-positioned to support our clients in integrating security measures beyond digital assets. Our expertise extends to electronic security and vehicle mitigation strategies, ensuring protection that addresses both physical and digital vulnerabilities in AI-ready environments. 

The Sustainability Dilemma: Can the Power Grid Keep Up with AI Demand? 

With the rise of generative AI, there’s growing public concern over the energy usage and sustainability of AI data centres. These workloads require vast computational power, which is driving a significant surge in energy consumption across the industry. In Ireland, data centres already account for one-fifth of all electricity consumption (CSO 2023), and in the UK, the National Grid anticipates a six-fold increase in commercial data centre demand by 2034. This presents not only a climate challenge through increased emissions, but also a technical one as electricity grids become increasingly stretched. 

The pressure on grid infrastructure is already shaping how data centres are developed. A lack of available grid connections is forcing developers to consider alternative power sources. Off-grid solutions, including gas-powered systems, are making a comeback, although they raise serious questions about long-term environmental impact. In the UK, National Grid and NESO are actively working to expand grid availability. New NESO connection reforms aim to clear the backlog of projects awaiting connection, giving priority to viable developments and accelerating access to clean energy sources. 

In response to these challenges, the industry is evolving. Data centres are increasingly being designed with renewable energy in mind. Leaders such as Google are investing in geothermal energy in the US, while Amazon, Meta, and Microsoft are exploring nuclear energy to meet their growing power demands. In one of the most notable examples, Microsoft is planning to reopen Three Mile Island by 2028 to power its data centre operations — reflecting the scale of energy needs now associated with AI computing. 

Closer to home, Scotland-based operators like DataVita by HFD are integrating green energy and renewable technologies into their operations. The UK’s temperate climate is also a key asset. At a national level, the UK Government is making AI infrastructure a strategic priority, with plans to develop secure, resilient and sustainable facilities both domestically and abroad (Department for Science, Innovation & Technology, 2025). 

Looking ahead, there’s growing interest in how the byproducts of data centre operations can be repurposed. For example, capturing and reusing waste heat from server operations could help reduce overall energy consumption, lower carbon emissions, and even provide heating for nearby homes or businesses. As AI continues to grow, these types of circular, efficient solutions will be vital to balancing innovation with long-term sustainability. 

The Future of AI & Data Centre Design 

There will be significant and continued growth, driven by increasing demands of AI workloads, and as AI continues to evolve, data centres must adapt  

Sustainability will continue to play a huge role in future Data Centre design and as the demand for AI power grows, data centres will have to draw from renewable energy sources and find ways to repurpose waste heat, such as for local heating systems. Regulatory changes will also guide these transformations, ensuring energy efficiency and responsible operation. 

Lastly, the design and construction of AI data centres will require more collaboration and expertise across different sectors. MEP design will become even more crucial, with engineers focusing on energy management, cooling solutions, and sustainability to ensure smooth, reliable operations. 

As AI demands continue to shape the future of data centres, MEP and ESG considerations are key to ensuring efficient, sustainable, and resilient infrastructure. Our expertise in MEP design allows us to address the high power and cooling needs of AI technologies while integrating advanced systems that optimise energy use whilst minimising the environmental impact. 

At Wallace Whittle, we understand the influence that the growth in AI plays on Data Centre Design. If you’re looking to strengthen the resilience of your projects or would like to discuss how our expertise can support your engineering needs, reach out to Stephen Phimister or contact us at [email protected]