Europe is striving to become a leader in artificial intelligence (AI) to compete with the U.S. and China, but experts warn that rising energy prices could hinder these ambitions.
The region aims to enhance its computing capacity and develop essential infrastructure for AI, yet the energy-intensive nature of data centers makes them particularly vulnerable to energy costs, which have surged due to geopolitical tensions, including the U.S.-Iran war.
Analysts predict that data center projects may relocate to areas within Europe with lower energy costs, leading to disparities across the continent.
Michael Brown, a global investment strategist at Franklin Templeton, emphasized that the global energy cost differences will significantly impact investment decisions, suggesting that if he were to build a $7 billion data center, he would choose the U.S. or China over Europe.
Olivier Darmouni, an associate professor at HEC Paris, noted that the rapid expansion of data centers could increase electricity costs by 20-40% in high-demand regions like Texas and Virginia in the U.S. and parts of Europe. He stressed that without addressing energy system challenges, Europe cannot achieve affordability, competitiveness, or technological leadership in AI.
According to the International Energy Agency, energy prices for industries in Europe were approximately double those in the U.S. and 50% higher than in China and India last year. Data centers currently account for 2% of global electricity consumption, up from 1.7% in 2024, with concerns that community opposition may arise once their consumption exceeds 5% of national electricity use.
Chris Seiple from Wood Mackenzie identified three factors contributing to Europe's lag in data center development: high energy costs, the geographic location of data center companies, and the time required to build infrastructure. Despite Europe's plans to enhance data center capacity, Darmouni pointed out that matching U.S. levels would require substantial investment.
Vladimir Prodanovic from Nvidia indicated that regions like Germany and the U.K. have already fallen behind due to high electricity costs, citing average prices of $111.65 per MW in the U.K. and $88.97 in Germany, compared to $28 in the U.S. OpenAI's recent decision to pause its Stargate project in the U.K. was partly attributed to energy costs and regulatory concerns.
Darmouni also predicted that AI services might eventually reflect energy costs in their pricing, raising concerns about potential price discrimination. Conversely, the Nordics and France are positioned to benefit from AI investments due to lower electricity prices and a diverse energy mix.
Prodanovic highlighted Norway as a prime location for AI companies, with significant investments from Microsoft in the region. France's leadership in nuclear energy gives it a competitive edge in electricity pricing, but Darmouni emphasized the need for greater energy integration across Europe to ensure uniform pricing.
The geographical challenges in regions like the U.K. and the Iberian Peninsula complicate this integration. Additionally, a projected 12% increase in data center capacity costs in Europe's major markets by 2026 underscores the growing challenges in this sector