Chip-for-Energy: Can a Grand Bargain Save AI From Itself?
An SDES Strategic Briefing — August 2026 The artificial intelligence race is hitting a wall. Not a wall of algorithms. Not a wall of talent. A wall of electrons. By 2030, global data center electricity demand is projected to more than double, reaching approximately 945 terawatt-hours per year — roughly the entire current electricity consumption of Japan — with artificial intelligence workloads accounting for nearly half of that growth [1]. In the United States alone, data centers are expected to consume between 35 and 60 gigawatts of power by 2030, up from roughly 25 gigawatts today, driven almost entirely by AI training and inference clusters [2]. Individual next-generation AI training facilities are already being designed in the 1 to 5 gigawatt range — the output of a mid-sized nuclear power plant, dedicated to a single campus [3]. Meanwhile, China has spent the past decade building the world’s most aggressive energy expansion pipeline. In 2024, China added 278 gigawatts...