Europe AI data centres seek cheaper, quicker energy and land

European artificial intelligence data centre developers are targeting less conventional locations, seeking lower energy and land costs, and quicker connection times, according to a Reuters analysis. This shift could stimulate economic growth in underdeveloped regions, although it might face local opposition due to concerns over environmental impact and resource competition.

JLL, a leading commercial real estate and property services firm, reported plans for hyperscale data centres that will be up to three times further from major cities compared to those built in the previous three years. These new facilities do not need to be as close to end-users, allowing for more strategic placement.

The average distance between the new sites and major hubs is projected to be 175 kilometres (109 miles), up from 46 kilometres for projects completed between 2022 and 2025. This trend is primarily driven by the immense power requirements of AI training campuses, which also demand significant water for cooling. The decision-making process is now focused on securing sufficient power resources rather than just identifying high-demand areas.

According to JLL, the four major hyperscale cloud providers plan to invest $725 billion in 2026, a 77% increase from $410 billion in 2025, largely for AI computing and data centre infrastructure. By 2030, AI workloads could make up approximately half of global data centre capacity. While core markets such as Frankfurt, London, Amsterdam, Paris, and Dublin continue to be highly sought after, they are increasingly challenged by land shortages, planning restrictions, and lengthy grid connection waits.

Land costs vary greatly depending on the location, with powered land priced at €2.36 million per megawatt of IT load in core markets, €978,000 in secondary cities like Copenhagen, Warsaw, and Milan, and as low as €512,000 in tertiary areas like Bordeaux. Amsterdam remains the most expensive market at around €2.7 million per megawatt, followed by London and Frankfurt.

Despite the competition for space from other asset classes, JLL notes that Europe's core markets will remain crucial due to persistent enterprise demand for AI infrastructure.