Denmark Rations Grid Access as AI Buildout Floods Nordic Systems
Transmission operator halts new large-load agreements amid 60 GW queue
Denmark's national transmission operator, Energinet, has imposed a temporary moratorium on new grid-connection agreements for large-scale consumers. The move comes as a massive surge in requests from AI data centers and other industrial projects threatens to overwhelm the country's electricity infrastructure.
Key details
Energinet revealed that the national queue for large-load grid connections has swelled to approximately 60 GW of pending requests. This figure is nearly nine times Denmark's peak national electricity demand of 7 GW. Data centers, primarily driven by hyperscale AI buildouts, account for roughly 15 GW—about a quarter—of these pending requests.
In response, the operator has abolished the "first-come, first-served" principle for grid access. New applications will now be evaluated in "pools" based on project maturity, technical progress, and "grid-friendliness." Some developers are being warned of timelines stretching 5 to 10 years before full grid access can be secured, as transmission infrastructure expansion struggles to keep pace with AI-driven demand.
Why this matters
This rationing marks a significant turning point for AI infrastructure in Europe. The Nordic region, once prized for its abundant renewable energy and cool climate, is hitting physical infrastructure limits. The shift from encouraging growth to active triage demonstrates that even renewable-rich grids cannot scale fast enough to meet the immediate power requirements of massive AI clusters.
Context
Denmark's decision mirrors similar moves in other global AI hubs, including Ireland, Singapore, and parts of the United States. While Denmark generated over 80% of its electricity from renewables in 2025, the bottleneck is no longer just generation but the physical capacity of the high-voltage transmission lines to deliver that power to concentrated AI "factories."
What happens next
Energinet is currently designing a new framework for evaluating project maturity, with a transition to the pooled review process expected over the next three months. Developers of AI data centers in the region may be forced to migrate workloads to other markets or invest significantly in on-site generation and grid-balancing technologies to secure a place in the new prioritization queue.
Source: Data Center Knowledge Published on AI Usage Global, author: AUG Bot
