Within market development, efforts can potentially focus on supporting data centres in becoming more active grid participants. Additionally, broader opportunities for added benefits could be explored, such as using automated energy management technologies to enhance grid flexibility, demand response and peak shaving. Encouraging advanced clean energy procurement (for instance, by matching hourly consumption with local clean energy) could also be prioritized. This approach recognizes the importance of where and when clean power goes onto grids, and how electricity is consumed (an element that could be beneficial to the energy transition).Additionally, identification of data centre development zones could be valuable, especially if integrated with grid planning and supporting policy efforts. Lastly, equipment upgrades and responsible IT asset recycling incentives could be explored to address growing e-waste generation (which could reach 2.5 million tonnes annually by 2030).25 By adopting these strategies, companies could reduce environmental impacts while still advancing AI solutions.3.7 Market development enablersFuture innovations could include specialized processors to reduce electricity use, with emerging technologies like quantum and neuromorphic computing enhancing efficiency. Quantum computing offers faster solutions, while neuromorphic computing enables low-power AI processing, transforming data centres for next- generation applications. Technological innovations in sustainable data storage can also support sustainable AI. Breakthroughs like biological data storage using synthetic DNA could revolutionize storage and computing, enabling massive scalability without overwhelming energy supply. Competitions rewarding energy-efficient data centre solutions can drive innovation, while case studies of data centres transitioning to renewable energy can inspire broader adoption of sustainable practices by showcasing economic, operational and environmental benefits. As these enablers progress, they offer various pathways for sustainable AI, balancing performance with environmental footprint.3.6 Technological innovation enablers Artificial Intelligence’s Energy Paradox: Balancing Challenges and Opportunities 17

Artificial Intelligences Energy Paradox 2025