Technology

Microsoft Wants Its Big AI Buildings to Be Better Neighbors

Martin HollowayPublished 2month ago3 min readBased on 11 sources
Reading level
Microsoft Wants Its Big AI Buildings to Be Better Neighbors
source:microsoft.com

Microsoft wants its huge data centers for AI to fit into nature and be less of a problem for neighbors. The idea is called biomimicry, learning from nature. It was described in reporting published Oct. 2, 2026, and centers on the very large sites needed for AI The Verge.

Microsoft describes the approach as "blending datacenters into the natural landscape to create resilient ecosystems that support biodiversity, strengthen community ties, and help datacenters become better neighbors." In plain terms, that means more gardens and local trees on its sites, plus money for projects meant to fix damaged nature where it operates.

San Antonio is the test case. Councilmember Ric Galvan remembers data centers first arriving in the 2000s looking like plain office buildings. That shape is gone. In recent years they have grown into huge buildings covering millions of square feet that hold the computers for AI.

Galvan's district covers about 55 square miles and now holds more than a dozen data centers, grouped along two main roads. He expects about 20 in the next few years if planned projects are finished. Galvan represents District 6 in San Antonio.

Cooling and site engineering

In 2024, Microsoft launched a new datacenter design built for AI that uses zero water for cooling Microsoft Sustainability Report. It cools the computer chips directly instead of cooling large rooms of air.

The company states its design for future large datacenters addresses reliability, efficiency, and sustainability Microsoft Azure. In other places it is growing projects that give water back, including AI-powered precision irrigation in Chile and rainwater harvesting at schools in Malaysia Microsoft Local.

Six new datacenters planned nearby are being designed with biomimicry from the start, according to Microsoft Microsoft Source. Those six are an early test, not a rebuild of older sites. Local plants and nature repair are part of the building plan, along with the cooling system.

Packed AI computers make a lot of heat in a small space. Operators must choose between cooling with evaporated water or with sealed liquid pipes, much like a car radiator moves heat with liquid. Microsoft uses direct-to-chip pipes with special fluid or water-glycol mixes, plus dry coolers that use air. That cuts water use but needs more electricity and more complex machines.

Worth flagging for operators, the open questions are breakdowns, upkeep at large size, and how the system holds up during long AI training and inference, the everyday running of AI models to give answers.

Permitting has become the gating factor

On Aug. 5, 2026, Galvan requested a Special Meeting of the San Antonio City Council concerning a Data Center Moratorium, a pause on new building. San Antonio is not alone. A growing number of governments, regulators and cities around the world were moving to freeze, restrict or ban new data centre construction as of August 2026 Reuters.

Opposition from towns and counties in the U.S. Midwest and Northeast led to cancellation of data center projects in the months before March 2026. Microsoft's president said winning the trust of U.S. communities is paramount to building data centres Reuters. New York became the first U.S. state to impose a data center moratorium, in an action dated July 14, 2026 Reuters.

Power, water, noise, building traffic, and large size are local issues. They are decided by councils, not standards groups.

The broader context here will be familiar to anyone who watched the cloud buildout. Plans once centered on cables, power deals, and tax breaks. Community approval is now in the critical path. A design that works on cost still fails if it cannot be zoned, powered, and cooled without long local fights.

In my view, biomimicry is less about biology and more about getting along with neighbors. Gardens do not change power use. They change the view, heat at the edge of the land, and whether people feel the building is an industrial intrusion or cared-for landscape.

Looking ahead, measurement will decide if this grows. Claims about nature and wildlife are hard to check without starting records, species lists, and checks over years. Fixing damaged nature is harder still. Tech teams will want clear borders, outside audits, and public reports, the same as for power and water. Without that, even green sites may be seen as cover. The hopeful side stays. If no-water cooling works under heavy AI use, and green design speeds permits, large sites can grow with a smaller local footprint.