Grids for data centres: ambitious grid planning can win Europe’s AI race

Report an error

Summary

AI-generated

This summary is written by a language model reading the source document. It is not the publisher's words and is not a substitute for the original.

Learn more about AI enrichment

The report argues that proactive grid planning is essential for Europe to meet its goal of tripling data centre capacity and capturing the economic benefits of AI. It highlights a shift in investment from traditional hubs (FLAP-D markets) to regions with lower grid congestion, such as the Nordics and Southern Europe. The analysis suggests that innovative connection agreements, data centre flexibility, and strategic spatial planning can significantly reduce connection wait times and infrastructure costs.

Key insights

AI-generated

These insights are written by a language model reading the source document. They are not the publisher's words and are not a substitute for the original.

Learn more about AI enrichment
  • Electricity demand from European data centres is projected to grow from 96 TWh in 2024 to 236 TWh by 2035, representing an increase of nearly 150% over the decade.
  • Grid congestion is eroding the dominance of traditional data centre hubs (Frankfurt, London, Amsterdam, Paris, and Dublin), with their share of Europe's capacity expected to drop from 62% to 51% by 2035.
  • Investment is shifting toward markets with better grid availability; the Nordics and Southern Europe are expected to see data centre demand grow by 110% by 2030, nearly double the 55% growth rate of the FLAP-D markets.
  • Grid connection wait times in legacy hubs average 7-10 years and can reach up to 13 years, whereas countries with half that connection time are projected to attract 20% more growth by 2030.
  • Data centres often underutilise their contracted grid capacity, with European averages at 44% in 2024 and specific examples in Ireland (34%) and Norway (30%), suggesting that phased connection agreements could reduce queue times by up to 80%.
  • Implementing flexibility—where data centres adjust consumption during grid stress—could more than double the available grid capacity if facilities offered just 30 hours of flexibility annually.

Cite the original document

APA
Cremona, E., & Czyzak, P. (2025). Grids for data centres: ambitious grid planning can win Europe’s AI race. Ember. https://ember-energy.org/app/uploads/2025/06/Grids-for-data-centres-in-Europe.pdf
Chicago
Cremona, Elisabeth, and Pawel Czyzak. Grids for data centres: ambitious grid planning can win Europe’s AI race. Ember, 2025. https://ember-energy.org/app/uploads/2025/06/Grids-for-data-centres-in-Europe.pdf.
Wikipedia
{{cite report |last1=Cremona |first1=Elisabeth |last2=Czyzak |first2=Pawel |title=Grids for data centres: ambitious grid planning can win Europe’s AI race |publisher=Ember |date=19 June 2025 |url=https://ember-energy.org/app/uploads/2025/06/Grids-for-data-centres-in-Europe.pdf |access-date=17 August 2026 |via=Climate Insights Directory}}
BibTeX
@techreport{cremona2025grids, author = {Cremona, Elisabeth and Czyzak, Pawel}, title = {{Grids for data centres: ambitious grid planning can win Europe’s AI race}}, institution = {Ember}, year = {2025}, month = jun, url = {https://ember-energy.org/app/uploads/2025/06/Grids-for-data-centres-in-Europe.pdf}, urldate = {2026-08-17}, note = {Indexed by Climate Insights Directory} }

Full text

Collected · Record updated