The Underappreciated Need to Enable AI and Data Center Growth

Data Centers for the storage and computation of data began in earnest in the U.S. in 2006 as the concept of performing computations and storing data off premises (The Cloud) began to take shape. More recently, data centers have become synonymous with realizing the promise of generative artificial intelligence (AI). It is generally understood that for the U.S. to continue to lead unabated in the AI revolution, the country must build data centers at an unprecedented, accelerated pace. At least three general types of data centers are needed: 1) Hyperscale training centers for AI development; 2) Hyperscale cloud centers for data storage and computation; and 3) Colocation and edge centers for interconnection to users. This paper focuses on the first two categories related to hyperscale data centers.