Hardware · IEEE Spectrum AI ·
Accelerating Chipmaking Innovation for the Energy-Efficient AI Era
A sponsored Applied Materials article argues that energy-efficient AI will depend on system-level chip engineering across logic, memory, and advanced packaging. It says data movement is becoming a major energy cost, making lower energy per bit as important as higher compute performance.