AI and Data Centers Are Creating New Sustainability Challenges

The rapid growth of artificial intelligence technologies is accelerating hyperscale data center investments worldwide, while the environmental impacts of this expansion are becoming a major topic of discussion in terms of sustainability. International organizations, regulatory authorities, and energy experts warn that the increasing electricity consumption, water usage, and carbon emissions of data centers could reshape the global sustainability agenda over the next decade.

According to the International Energy Agency (IEA), electricity consumption by data centers increased significantly in 2025, particularly due to AI workloads and GPU-intensive infrastructure investments. The agency estimates that investments by major technology companies in data center and artificial intelligence infrastructure exceeded USD 400 billion in 2025 and could increase by approximately 75% in 2026.

AI-focused data centers require substantially greater computing power compared to traditional cloud infrastructures. Advanced AI chips can consume several times more energy than conventional processors, creating significant pressure on electricity grids and renewable energy systems. According to analyses by S&P Global, global data center electricity demand is expected to nearly double between 2024 and 2030, with AI applications serving as the primary driver of this growth.

The increasing energy demand is also reshaping energy and infrastructure strategies. In the United States, one of the key factors behind the merger discussions between NextEra Energy and Dominion Energy is believed to be the rising energy demand generated by AI data centers. Pressure on electricity infrastructure is increasing rapidly, particularly in regions hosting hyperscale data center clusters.

In addition to energy consumption, water usage has become one of the most controversial sustainability issues surrounding AI infrastructure. Many data centers rely on evaporative cooling systems that require large amounts of freshwater to maintain optimal operating temperatures.

According to a recent study published in the journal Water Research, the global water footprint of artificial intelligence systems could reach between 4.2 and 6.6 billion cubic meters annually by 2027. The study also reveals that nearly two-thirds of newly established data centers are located in water-stressed regions. This indicates that pressure on water resources may intensify further in the future. At the same time, local opposition to data center projects continues to grow. In May 2026, the city of Millville in the U.S. state of New Jersey decided to ban large-scale new data center investments due to concerns over water consumption, energy costs, noise pollution, and environmental impacts.

Technology companies and public authorities are working to develop alternative solutions aimed at reducing the environmental impacts of AI infrastructure. One of the most notable recent developments is China’s launch of the world’s first offshore wind-powered underwater data center near Shanghai. The facility uses offshore wind energy to power AI workloads while utilizing seawater for passive cooling. As a result, it is reported to achieve significantly lower energy consumption compared to conventional data centers.

Other sustainability solutions gaining prominence globally include liquid cooling systems, closed-loop water recycling applications, waste heat recovery, renewable energy integration, battery energy storage systems, and advanced energy efficiency optimizations for AI workloads. The European Commission also emphasizes that waste heat generated by data centers could be utilized in carbon capture and water purification processes, highlighting that data centers may evolve from being merely high-consumption facilities into integral components of circular sustainability systems.

Investors, regulators, and sustainability stakeholders are increasingly demanding more transparent reporting regarding the carbon emissions, water consumption, and climate risks associated with data centers. Major technology companies including Google, Microsoft, Amazon, and Meta are facing growing pressure regarding the compatibility of their AI investments with long-term net-zero emission targets. According to experts, the sustainable future of artificial intelligence will depend not only on advances in processor efficiency, but also on the decarbonization of energy systems, responsible water management, and more transparent ESG reporting practices.