Data centers have existed for decades, but they have become much more prominent recently. Is the current growth in data center construction mainly being driven by AI, or are other technologies—such as video surveillance, cloud services, data collection, streaming, and expanding online platforms—also major contributors?
4 Answers
AI is a major reason for the recent surge, especially because training and running large models requires huge numbers of GPUs. Compared with ordinary web traffic, these systems need far more electricity, cooling capacity, and specialized hardware. The demand is not just for floor space—it is heavily constrained by power and heat management.
AI is important, but it is not the only factor. Cloud computing, video streaming, online services, machine-learning systems, and the growing amount of recorded data all contribute to the need for more storage and processing capacity. AI is probably the biggest new source of growth, while these older workloads continue expanding in the background.
Large-scale video systems and surveillance archives can also add substantial storage demand, particularly when footage is retained for long periods. Storage-heavy applications are different from AI training, but both contribute to overall data center expansion.
The key difference with AI-focused facilities is resource intensity. Large models use dense GPU clusters, which consume much more electricity and generate more heat than many traditional server workloads. That can require major investments in cooling systems, transmission infrastructure, and sometimes new power generation.
That distinction makes sense: data storage and conventional cloud workloads were already growing, but AI has sharply increased the amount of compute and power needed per workload.