How AWS is Using AI to Make Buildings More Energy Efficient

The buildings where people work, shop and live are responsible for more carbon emissions globally than anything else, which is why Amazon Web Services chose this year’s re:Invent conference in the US to spotlight a major push into AI-powered building efficiency.

At the summit in Las Vegas, the company revealed the ways in which it is using AI to transform its grocery fulfilment network into a network of smarter, lower carbon facilities that can respond in real time to changing conditions, all while using significantly less energy than before.

At the centre of this announcement is a new virtual assistant for commercial buildings that talks directly to heating, ventilation and air conditioning systems.

Developed through a collaboration between Trane Technologies and BrainBox AI and running on Amazon Bedrock, the system uses AI to continuously optimise temperature, airflow and energy use rather than relying on fixed schedules and manual adjustments.

Each site effectively gains its own digital operator that reads data from sensors, predicts demand and fine tunes equipment settings throughout the day, helping to maintain comfort while simultaneously slashing wasted energy.

From pilots to real energy savings

AWS and Trane Technologies first piloted the technology in three Amazon grocery fulfilment centres across North America, where refrigeration loads, automation and round the clock operations make energy management an especially complex task.

In those pilots the firm's AI assistant managed to reduce the facilities' energy use by nearly 15%, more than double what had initially been hoped for.

The success shows that smarter controls can deliver meaningful savings even before any physical retrofits to the buildings themselves.

Those reductions translate directly into lower operating costs and avoided carbon emissions, reinforcing the business case for applying advanced analytics to existing infrastructure.

Scaling across Amazon’s grocery network

On the back of the pilot results, Amazon plans to roll out the AI-powered efficiency system across more than 30 grocery fulfilment and distribution centres in the US.

This scale up will extend the virtual assistant from a small test group of sites to a substantial portion of Amazon’s grocery logistics footprint, multiplying the potential carbon and cost savings year on year as more facilities come online.

The partners also intend to start pilots in grocery stores from next year, moving beyond back end hubs into customer facing spaces where comfort, food safety and brand experience all depend on stable climate control.

Faster fault detection for facility teams

The system does more than tune set points.

By analysing large streams of building data, the AI can help facility managers and building operators locate, diagnose and fix problems faster than traditional building management systems, from stuck dampers to failing chillers.

That early warning capability reduces downtime and maintenance costs while ensuring that energy savings are not eroded by hidden faults that would otherwise go unnoticed for weeks or months.

Buildings as a frontier for decarbonisation

AWS frames the initiative as part of a broader effort to tackle emissions from the built environment, which is now widely recognised as one of the most important frontiers for climate action.

Because the new AI layer is software based and runs on Amazon Bedrock, it can be adapted and improved over time, learning from different facilities and potentially being extended to other types of buildings beyond grocery fulfilment sites.

If replicated across offices, warehouses, data centres and retail spaces, similar efficiency gains could deliver substantial carbon reductions at a global scale without waiting for full deep retrofit cycles.

For Amazon, re:Invent has become the annual showcase for its most important cloud and AI announcements and this year sustainable infrastructure is firmly part of that narrative.

"These announcements reflect a broader truth," says Kara Hurst, Amazon’s Chief Sustainability Officer.

"AI and cloud technology are helping us build better businesses and a better future, by enabling and accelerating climate action. And this is just the beginning!"