# EU: AI to Transform Data Center Operations – But Not Overnight — retained record

> Retained news record from news:datacenterknowledge.

Canonical: https://highsignal.app/data/records/447aa1dd0694961f

[← sources](https://highsignal.app/data)


source: [news](https://highsignal.app/data/news) (news:datacenterknowledge)  published 2026-05-20 08:55 ingested 2026-05-21 09:06

[Alphabet Inc.](https://highsignal.app/entities/GOOGL)[original source ↗](https://www.datacenterknowledge.com/ai-data-centers/eu-ai-to-transform-data-center-operations-but-not-overnight)

## extract

EU: AI to Transform Data Center Operations – But Not Overnight
We spoke to the EU agency managing critical IT systems about its vision for AI in government data center operations.
AI is emerging as a “transformative force for optimizing operations,” according to a report from the organization that oversees much of the European Union’s digital infrastructure.
The Energy-Efficient Data Centers report, from the EU Agency for the Operational Management of Large-Scale IT Systems in the Area of Freedom, Security and Justice (EU-LISA), covers a range of issues from the regulatory landscape to current best practice.
Areas of innovation, including the use of AI to improve the management and operations of data centers, are also examined in the report. Generative AI is a critical workload for data centers, driving key technological shifts, such as the adoption of liquid cooling and the use of graphics processing units (GPUs). But the use of other forms of AI inside the data center has arguably received less attention.
According to the report: “In large-scale cloud and hyperscale facilities, AI-driven control systems can continuously optimize infrastructure operation, improving key efficiency metrics such as power usage effectiveness (PUE), and enabling more sustainable management of computing resources in real time”.
However, EU-LISA, which manages several large-scale IT systems including the EU entry and exit system, cautioned that despite the potential of AI “to become a very powerful tool in data center energy management,” the technology is still nascent.
“[A] key finding from research, expertise and experience is that in most cases AI-based solutions are not mature enough to be considered as a first line fix for optimizing data center energy efficiency,” the report reads.
With demand for data center capacity growing and physical constraints becoming a pressing issue, we spoke to several stakeholders in the EU who discussed the Energy-Efficient Data Centers report. They shared insights into the opportunities and challenges AI might hold for the EU’s government data center industry.
The EU is leveraging AI to enhance data center efficiency, with hyperscalers leading innovative applications. (Image: Alamy)
AI’s Role in Transforming Data Center Operations
Recent research from data center certification body Uptime Institute, aligns with EU-LISA’s findings. The organization found that while AI tools have the potential to transform data center management, there is still considerable caution around full-scale adoption.
“AI-based products and services for data center management and operational support are available, and most of the industry is entering a period of careful adoption, testing and validation,” Uptime Institute stated in its 2025 Global Data Center Survey. “Data centers are slow and careful in adopting new technologies, and AI will not be an exception.”
Tor Björn Minde, director, head of the ICE Data Center unit at RISE Research Institutes of Sweden, which contributed to the EU-LISA report, agreed that some forms of AI – such as machine learning – could be used to improve the management and operation of standalone pieces of equipment or other discrete actions.
“It can be used for example to find optimums in a limited range of parameter settings or predict weather for smart grid operations or predictive maintenance – for example error detection,” Minde said.
However, managing an entire data center remains a far more complex challenge, likely beyond the current capabilities of full AI automation. “A complete data center is a complex machine,” Minde said. “A large data center is a critical environment you don’t want to mess around with (i.e., operate in non-safe or non-optimized modes).”
Minde explained that one of the key challenges was that AI management tools would need to be trained in what to do in the event of failure – data which may not be readily available. “If you don’t have failure data or non-optimized operations data, how does the model account for that? The model has not been there,” he said.
AI Technologies Shaping the Future of Data Centers
Despite expressing caution around maturity, the EU report highlights progress made by hyperscalers in using AI for data center management, and related tasks.
Examples include:
Google, which used Deep Mind to adjust system settings for HVAC systems resulting in a 40% reduction in cooling energy and a 15% improvement in PUE.
IBM Watson, which uses real-time data analysis to anticipate failures of aging equipment such as fans and UPS systems.
Meta, which used AI to schedule non-critical workloads based on the availability of solar and wind, improving renewable-energy utilization.
Other innovative technologies featured in the report include:
High voltage direct current (HVDC): Using HVDC for power distribution inside the data center eliminates several AC to DC conversion steps, reducing energy losses. Suppliers including Nvidia are backing 800V HVDC for AI data centers. However, as the report notes “There is likely to be a significant delay before the widespread adoption of a new standard around 800V HVDC”.
Cooling chillers with magnetic bearing compressors: These reduce friction and improve energy efficiency, but downsides include higher costs and long payback times.
Hydrogen storage: The report suggests hydrogen production and storage may become a cornerstone of the future of smart energy grids or small buildings. However, hydrogen can be carbon intensive to produce and is expensive to store and distribute.
EU-LISA: Managing Critical Data Centers On-Prem
Javier Galbally, research and innovation officer at EU-LISA, told Data Center Knowledge that the organization has three data centers across the EU with the main site in Strasbourg, France, and additional facilities in Austria and Estonia.
Legislation around the critical data that EU-LISA controls and manages mandates it must operate its own on-premises data centers and is forbidden from using third party service providers. “There is a very clear regulation for each of the systems, and the regulation does not allow any of the data to be put in the cloud. Let's say that EU-LISA is the cloud. We are the cloud,” he said.
Some of the motivation for producing the report is that EU-LISA’s main site is facing capacity constraints, so it was important to understand best practice for improving data center efficiency and to also pass that insight on to other organizations, Galbally explained.
“We are getting to the physical space limit to host any more racks,” he said. “We will need at some point to do an expansion.”
AI holds immense potential to transform data center operations, offering opportunities to optimize efficiency, reduce energy consumption, and improve sustainability. However, as highlighted by EU-LISA and other industry experts, the technology is still in its early stages, requiring careful adoption and validation.
Still, as demand for data center capacity continues to grow, organizations like EU-LISA continue to explore innovative solutions while balancing operational maturity and safety. The journey toward fully AI-optimized data centers may be gradual, but the progress made by hyperscalers and advancements in supporting technologies signal a promising future for the industry.
Read more about:
Europe

## retained text

No document body was retained for this record — only its metadata was stored.