Innovation, Planning, and Coordination Across the Energy Sector
Paul Glanville, PE is based in Chicago and is a Senior R&D Director in GTI Energy’s Low Carbon Energy Solutions group. He leads the cross-cutting Heat & Power group that collaborates with industry and academia, resulting in several products, patents, and publications. He is active in ASME, ASHRAE, and ANSI/CSA, and is a registered Professional Engineer in Illinois.
The dramatic growth of artificial intelligence is creating a fundamental shift in how we think about energy infrastructure. As AI becomes embedded in everything from healthcare diagnostics to financial modeling, the data centers that power these capabilities are becoming essential utilities in their own right. Their rising power needs are challenging conventional approaches to power delivery.
The scale of this challenge is most visible in the pace of new development. Wood Mackenzie estimates the U.S. has more than ninety gigawatts of new data center capacity in the pipeline, with many individual sites requiring one hundred megawatts or more. These facilities aren’t simply larger versions of those built a decade ago. They represent a new class of energy customers with concentrated loads, minimal tolerance for downtime, and development timelines that often outpace traditional utility planning cycles.
In today’s build cycle, some hyperscale developers are turning to natural gas-powered on-site generation as a bridge solution, using it to secure predictable, high-capacity power while waiting for long-timeline infrastructure upgrades. These systems serve as a strategic complement to grid power, reinforcing reliability during both construction and full operation.