Microsoft Vice Chair and President Brad Smith shared details on the company’s latest updates on datacenter water efficiency as it expands AI and cloud infrastructure. In a post on X, Smith said water management is a key part of Microsoft’s Community-First AI Infrastructure plan, which focuses on expanding datacenter capacity while improving resource efficiency.
When we launched our Community-First AI Infrastructure plan in January, we made a promise: we would meet the growing demand for datacenters without straining the resources communities depend on. Water was central to that commitment.
Today we shared our progress. Since our first… pic.twitter.com/WNr1Ab1DCp
— Brad Smith (@BradSmi) June 24, 2026
According to Microsoft, the company has improved its water-use effectiveness (WUE) by nearly 90% since its earliest datacenter designs in the early 2000s, reducing average WUE from 2.3 litres per kilowatt-hour to 0.27 litres per kilowatt-hour in 2025. The company said the improvement was driven by advances in cooling technologies, operational optimisation and the increased use of alternative water sources.
Microsoft also said it has reduced datacenter water-use intensity by 25% as of 2025, putting it more than halfway toward its goal of improving water-use intensity by 40% by 2030. The company added that it replenished more water than it withdrew across its global operations in FY2025.
New datacenter designs introduced in 2024 for AI workloads use a zero-water cooling system during operations through a closed-loop direct-to-chip cooling design. Microsoft also said approximately 90% of its owned datacenter fleet now operates using low- to zero-water cooling systems.
In addition to cooling innovations, Microsoft said it is expanding the use of recycled, reused and non-potable water across several locations, alongside rainwater harvesting systems and on-site water treatment technologies that allow water to be reused multiple times in cooling operations.
Microsoft also said it continues to invest in community water infrastructure and restoration projects, including AI-enabled leak detection and wetland restoration, as part of its broader water stewardship efforts.
The update reflects how major technology companies are working to improve the efficiency and sustainability of AI infrastructure as demand for cloud and AI services continues to grow.
Also Read: NVIDIA Claims 100% Liquid Cooling and Near-Zero Water for Rubin AI Data Centres
