NERC Flags AI Data Center Grid Risks in Report
NERC’s 2026 report emphasizes that as individual campuses approach gigawatt scale and cluster regionally, these abrupt disconnections now threaten grid stability and require new modeling standards, operational coordination, and regulatory frameworks.
Get smart on it
The North American Electric Reliability Corporation documented incidents in 2025 where gigawatt-scale AI data center campuses rapidly disconnected from the power grid during transmission disturbances, shedding over 1 GW of demand within moments. This matters because unlike traditional industrial customers, large computational loads can disconnect suddenly rather than gradually, and as these facilities grow larger and cluster regionally, their abrupt disconnections now threaten overall grid stability and frequency. Utilities historically planned for how much power customers would consume, but must now model how data centers behave during grid disturbances and coordinate operations before new facilities connect. NERC responded by developing new modeling guidance, standards, and registered entity categories for computational load facilities, while industry groups work to improve data sharing and operational coordination between utilities and hyperscale developers.
Related news
The Switch Is the Bottleneck: Why AI Infrastructure Has a Network Problem
AI workloads are outpacing network capabilities, leaving expensive chips idle. Mark Rushworth explains why the switch is the bottleneck and how to fix it.
Texas Tests New Rules for AI Campuses Behind Existing Power Plants
PUCT staff backs ERCOT’s proposed operating conditions in an early test of Texas’ new framework for colocated data centers.
