“By 2030, AI could represent half of all data center workloads with inference becoming the primary driver. AI only represented about a quarter of all data center workloads in 2025, with training driving most of the demand; a significant shift is anticipated in 2027, when inference workloads could overtake training as the dominant AI requirement.”

Published alongside JLL's forecast that global data center capacity nearly doubles from 103 GW to about 200 GW by 2030; the workload-share milestone cannot be judged until 2030.