3:19 am - February 14, 2026
The UAE launches a pioneering AI pilot project to reduce energy consumption and enhance resilience in its expanding data centre network, signalling a shift towards smarter, greener infrastructure.
The UAE has kicked off some testing of artificial intelligence, aiming to lower energy consumption and boost resilience across its growing data center network. This move seems to signal a kind of push to balance the rapid growth in AI with the country’s goals around climate and efficiency. So, recently, the Ministry of Energy and Infrastructure, along with Khazna Data Centers and logistics company Agility, revealed a pilot project during the World Governments Summit in Dubai. They’re testing out software from Seattle-based company Phaidra in various data centers and district cooling setups.
According to Agility, this trial will utilize Phaidra’s reinforcement-learning control agents to coordinate power, cooling, and workload management. The partners say their main goal is to squeeze out better energy efficiency from these operations, while still meeting the rising performance demands driven by AI workloads. Khazna will oversee the deployment and will test the system at a few selected sites. They’ll be measuring how much energy can be saved, whether IT capacity grows, and if system reliability improves, especially under hot climate conditions.
Phaidra describes its platform as an adaptive AI tool that works in real-time to optimise complex physical systems. As reported by Data Centre Dynamics, the company claims its technology could cut cooling-related energy use by up to 40%, and also trim capital expenses on cooling infrastructure. If such results hold true when scaled up, it could significantly change the energy demands of high-density data centers.
This pilot ties into a broader cooperation that was initially signalled through a memorandum of understanding signed between the Ministry and Agility at Abu Dhabi Sustainability Week. Officials framed the project as part of the UAE’s strategy to stay at the forefront as a global hub for AI and digital investments, all while aiming toward its Net Zero 2050 goals and national efficiency targets.
Khazna’s recent expansion gives a good testing ground, really. The company has announced several hyperscale projects, including AUH6 in Masdar City and a 100MW AI-optimized campus in Ajman, expected to be up and running by late 2025. Khazna reports that AUH6 supplies about 31.8MW of IT power and integrates renewable energy via a solar PV plant developed with Emerge. Their bigger development plans include AUH4 and AUH8, which will add another 60MW altogether, with targets for completion in 2026. They say their system designs lean on modular architecture and adiabatic free cooling, aiming for low PUE figures typical for the region.
People in the industry have been quick to note that the timing is pretty significant. The global demand for AI compute is skyrocketing, and energy use at hyperscale sites is climbing as data centers add more GPUs and accelerators. The International Energy Agency, which Khazna referenced, predicts that AI-related power needs could double by 2026, putting extra pressure on utilities and cooling systems, especially in hot climates.
Public-private collaboration seems to be central to tackling this challenge. As per Data Centre Dynamics, the UAE’s trial exemplifies how government bodies, data center operators, and tech vendors are teaming up to pair smart control software with infrastructure upgrades. Advocates argue that intelligent controls can deliver quick operational improvements without needing huge upfront capital investments immediately.
Of course, skeptics caution that vendor claims tend to reflect the best possible outcomes. The assertion of up to 40% cooling savings is based on previous implementations, but independent verification across different sites will be essential to really prove the point. The partners themselves have described the pilot as a kind of proof-of-concept, with Agility indicating that the results will help guide decisions about wider adoption in the UAE.
The project also addresses resilience, which is a big deal given the scorching temperatures. Higher heat means more stress on cooling systems and power infrastructure. By dynamically balancing workloads and adjusting cooling in real-time, adaptive control systems aim to reduce thermal stress, lower failure rates, and provide more stable service during heatwaves or power shortages.
There’s also a potential ripple effect on energy markets. Lower cooling loads can lead to reduced peak power demand and lessen strain on the grid. The UAE has committed to emission reductions and efficiency goals under the Net Zero 2050 plan, so improving data center operations fits squarely within that agenda. But, scaling these efficiency gains across all hyperscale and colocation facilities will depend on hardware setups, local power generation, and regulatory policies.
Khazna’s overall strategy aligns with this ecosystem shift. The operator emphasizes that their new builds are AI-ready and has partnered with NVIDIA to develop next-gen infrastructure. They position their facilities in Ajman and Masdar City as key components of a national AI platform, aiming to attract investments and create jobs, while also trying to keep their environmental footprint as low as possible.
For stakeholders involved in climate tech in the UAE, this pilot offers a pretty interesting glimpse into how software-driven optimization might ease the energy impacts of a rapidly expanding sector. If the trial confirms Phaidra’s claims under Gulf conditions, it could speed up the adoption of AI-based control systems across district cooling and data centers. On the other hand, if the results are more modest than hoped, it would emphasize the need for additional measures, like more efficient hardware, more renewable energy, and policies to manage demand better.
Looking ahead, the partners plan to use the results of the pilot to evaluate larger-scale deployment. People across the Middle East, where high temperatures, quick digital growth, and ambitions for AI dominance all intersect, will be watching closely. Ultimately, this initiative suggests that the UAE is exploring whether smart, AI-driven operations can help reconcile its goal of expanding high-density computing with its sustainability commitments.
Source: Noah Wire Services
- https://container-news.com/uae-ministry-khazna-and-agility-launch-ai-pilot-to-boost-energy-efficiency/ – Please view link – unable to able to access data
- https://agility.com/news/uae-ministry-of-energy-and-infrastructure-khazna-and-agility-announce-pilot-to-implement-phaidra-ai/ – The UAE Ministry of Energy and Infrastructure, Khazna Data Centers, and Agility have announced a pilot project to implement Phaidra’s AI technology across data centers and district cooling operations in the UAE. The initiative aims to enhance energy efficiency and economic performance of energy-intensive digital infrastructure, reinforcing the UAE’s position as a global hub for AI and digital investment. The pilot will test Phaidra’s advanced AI control systems, which use reinforcement learning agents to manage power, cooling, and workloads in AI data centers, aiming to maximise energy efficiency while meeting rising performance demands. Khazna will lead the operational deployment, evaluating the technology across selected data center campuses to assess energy savings, increased IT capacity, and improved system resilience, particularly in high-temperature environments. The project also builds on Khazna’s collaboration with NVIDIA to develop next-generation AI infrastructure across the region. The partners will use the results to assess wider deployment across the UAE, supporting the country’s Net Zero 2050 Strategy and national energy efficiency goals. ([agility.com](https://agility.com/news/uae-ministry-of-energy-and-infrastructure-khazna-and-agility-announce-pilot-to-implement-phaidra-ai/?utm_source=openai))
- https://www.datacenterdynamics.com/en/news/uae-launches-ai-efficiency-pilot-across-khazna-data-centers-in-partnership-with-phaidra-and-agility/ – The UAE Ministry of Energy and Infrastructure, Khazna Data Centers, and Agility have launched a pilot project to deploy Phaidra’s AI system across Khazna’s data center portfolio. The initiative aims to improve energy efficiency in data centers and district cooling systems in the UAE. Under the terms of the pilot project, Phaidra’s AI agents will be tested across selected Khazna data center campuses to optimize power, cooling, and workload management. Phaidra claims that its platform has proven effective at reducing capital expenditures for data center cooling and lowering energy consumption by up to 40 percent. The primary aims of the pilot will be to reduce cooling energy use, increase IT capacity, and improve reliability, particularly in high-temperature environments prevalent in the Middle East. This initiative underscores the importance of effective public-private collaboration as AI reshapes global digital and physical infrastructure. ([datacenterdynamics.com](https://www.datacenterdynamics.com/en/news/uae-launches-ai-efficiency-pilot-across-khazna-data-centers-in-partnership-with-phaidra-and-agility/?utm_source=openai))
- https://khaznadatacenters.com/press-release/khazna-announces-the-uaes-first-ai-optimized-data-center-and-the-largest-in-its-portfolio-powering-the-nations-drive-toward-ai-leadership/ – Khazna Data Centers has announced the launch of the UAE’s first AI-optimized data center in Ajman, marking the largest in its portfolio. The 100MW facility, featuring 20 data halls, each with 5MW capacity, is set to be operational by Q3 2025. The Tier 3 data center is designed to meet the rising power demands in the AI era, which are anticipated to double by 2026, according to the International Energy Agency. The facility’s electricity supply will be primarily provided by Etihad Water and Electricity Company (EtihadWE). The data center is being constructed with an energy-efficient, modular design to ensure effective resource utilization. With a multi-billion-dirham investment, Khazna’s AI-focused data center is set to positively impact the local economy by boosting highly skilled jobs in AI and positioning the UAE as an investment hub for business, AI, and innovation. ([khaznadatacenters.com](https://khaznadatacenters.com/press-release/khazna-announces-the-uaes-first-ai-optimized-data-center-and-the-largest-in-its-portfolio-powering-the-nations-drive-toward-ai-leadership/?utm_source=openai))
- https://khaznadatacenters.com/press-release/khazna-expands-infrastructure-to-accelerate-uaes-leadership-in-artificial-intelligence/ – Khazna Data Centers is expanding its infrastructure to accelerate the UAE’s leadership in artificial intelligence. The establishment of AUH4 and AUH8, as well as the strong progress on QAJ1, reaffirms Khazna’s position as a key enabler of the future economy in the UAE. AUH4 and AUH8 will substantially expand the region’s cloud hosting capacity, providing a combined 60MW of capacity. Due for completion in December 2026 and August 2026 respectively, Khazna is using a modular design architecture to improve efficiency during the build phase, with minimal waste and faster construction timelines. By employing adiabatic free cooling to improve cooling efficiency, these data centres are set to deliver industry-leading Power Usage Effectiveness (PUE) figures for the region. The QAJ1 facility in Ajman will be the region’s first AI-optimized data centre. With a 100MW capacity, it has been designed specifically to deliver the high-density compute infrastructure required for the AI-powered applications reshaping the economy. With large-scale AI infrastructure a strategic priority for the UAE, the aggressive construction timelines of all three facilities will help the UAE meet its ambitious digital transformation targets. ([khaznadatacenters.com](https://khaznadatacenters.com/ar/press-release/khazna-expands-infrastructure-to-accelerate-uaes-leadership-in-artificial-intelligence/?utm_source=openai))
- https://khaznadatacenters.com/press-release/khazna-data-centers-fuels-uaes-digital-future-with-the-launch-of-hyperscale-ai-ready-data-center-in-masdar-city/ – Khazna Data Centers has launched a hyperscale AI-ready data center in Masdar City, named AUH6, which combines advanced technology, security, and sustainability. Standing at the forefront of the UAE’s digital transformation, AUH6 signifies Khazna’s commitment to shaping tomorrow’s AI reality. In an era driven by data and AI, AUH6 emerges as a hub for exclusive AI-related data development and hosting for its customers, offering 31.8-megawatt (MW) of IT power capacity. Located strategically in Masdar City, an urban development renowned for its commitment to sustainable practices and innovative technologies, AUH6 aligns with Khazna’s Future First Strategy, which outlines the company’s sustainability ambitions. The data center employs a modular and repeatable structure that not only enhances construction efficiency but also boosts performance, accelerates deployment timelines, and ensures seamless day-to-day operations. Through Khazna’s collaboration with Emerge, a ground-mounted solar photovoltaic (PV) plant has been developed to efficiently provide electricity to fuel the operations at AUH6. ([khaznadatacenters.com](https://khaznadatacenters.com/press-release/khazna-data-centers-fuels-uaes-digital-future-with-the-launch-of-hyperscale-ai-ready-data-center-in-masdar-city/?utm_source=openai))
- https://www.phaidra.ai/ – Phaidra develops AI software that optimizes power, cooling, and operational efficiency in data centers and other energy-intensive infrastructure. Founded by engineers with deep experience building large-scale AI and control systems, the company applies adaptive AI to manage complex environments safely and in real time. Phaidra is headquartered in Seattle and supports customers globally as demand for high-density computing continues to grow. ([phaidra.ai](https://www.phaidra.ai/?utm_source=openai))
Noah Fact Check Pro
The draft above was created using the information available at the time the story first
emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed
below. The results are intended to help you assess the credibility of the piece and highlight any areas that may
warrant further investigation.
Freshness check
Score:
8
Notes:
The announcement was made on 5 February 2026, aligning with the publication date of the source article. ([agility.com](https://agility.com/news/uae-ministry-of-energy-and-infrastructure-khazna-and-agility-announce-pilot-to-implement-phaidra-ai/?utm_source=openai)) The content appears original and not recycled from other sources. However, the article includes information about Khazna’s previous projects, such as the AUH6 data center in Abu Dhabi, which was completed in November 2023. ([red-eng.com](https://www.red-eng.com/projects/project-auh6?utm_source=openai)) This suggests that while the announcement is recent, some contextual information may be older.
Quotes check
Score:
7
Notes:
The article includes direct quotes from officials like His Excellency Engineer Sharif Al Olama and Tarek Sultan. ([agility.com](https://agility.com/news/uae-ministry-of-energy-and-infrastructure-khazna-and-agility-announce-pilot-to-implement-phaidra-ai/?utm_source=openai)) These quotes are consistent with the official press release from Agility. ([agility.com](https://agility.com/news/uae-ministry-of-energy-and-infrastructure-and-agility-sign-mou-to-accelerate-energy-efficiency-in-data-centers-and-district-cooling/?utm_source=openai)) However, the absence of independent verification of these quotes raises concerns about their authenticity.
Source reliability
Score:
6
Notes:
The primary source is Agility’s official press release, which is a corporate communication. ([agility.com](https://agility.com/news/uae-ministry-of-energy-and-infrastructure-khazna-and-agility-announce-pilot-to-implement-phaidra-ai/?utm_source=openai)) While Agility is a reputable company, press releases are inherently promotional and may lack objectivity. The article also references other sources like Data Centre Dynamics, but without direct links or citations, it’s unclear how independent these sources are.
Plausibility check
Score:
8
Notes:
The claims about the pilot project, including the use of Phaidra’s AI technology to enhance energy efficiency in data centers, are plausible and align with current industry trends towards AI-driven optimization. However, the assertion that AI-related power needs could double by 2026, as referenced by the International Energy Agency, ([agility.com](https://agility.com/news/uae-ministry-of-energy-and-infrastructure-khazna-and-agility-announce-pilot-to-implement-phaidra-ai/?utm_source=openai)) is a projection and may not be accurate.
Overall assessment
Verdict (FAIL, OPEN, PASS): FAIL
Confidence (LOW, MEDIUM, HIGH): MEDIUM
Summary:
The article presents information primarily sourced from Agility’s press release, lacking independent verification and third-party sources. While the content is plausible and timely, the reliance on a single corporate source and the absence of independent confirmation raise significant concerns about the accuracy and objectivity of the information. ([agility.com](https://agility.com/news/uae-ministry-of-energy-and-infrastructure-khazna-and-agility-announce-pilot-to-implement-phaidra-ai/?utm_source=openai))
