Microsoft's AI Datancenter Strategy Is Hurting Its Climate Goals

Microsoft's goal of becoming carbon-negative by 2030 has hit a roadblock: the AI gold rush.

In 2020, Microsoft announced it's spending $1 billion to reduce its carbon emissions over the next 10 years. However, its 2024 Environmental Sustainability Report, released earlier this week, indicates that it has fallen behind on some of its targets.

Specifically, its indirect emissions (i.e., emissions produced through its supply chains) increased by over 30 percent in fiscal 2023. These indirect emissions, also called "Scope 3" emissions, account for 96 percent of Microsoft's total carbon footprint.

Microsoft attributed the miss, in part, to the thousands of acres of servers needed to support today's generative AI boom. Microsoft is a solid AI front-runner thanks to its commitment to market darling OpenAI and its own Copilot ecosystem of generative AI tools, but getting (and staying) there has meant undertaking an unprecedented datacenter buildout.

In just the past few months, for instance, Microsoft has announced major infrastructure expansion plans in the state of Wisconsin (worth $3.3 billion), France ($4.3 billion), Japan ($2.9 billion), Germany ($3.4 billion) and Spain ($2.1 billion). There's also a joint Microsoft-OpenAI supercomputer datacenter that's allegedly in the works ($100 billion). That's in addition to the already 300-plus Microsoft datacenters in operation around the world.

Each of these datacenters -- from their construction to their maintenance -- represents a significant challenge to Microsoft's carbon-negative goal.

"New technologies, including generative AI, hold promise for new innovations that can help address the climate crisis. At the same time, the infrastructure and electricity needed for these technologies create new challenges for meeting sustainability commitments across the tech sector," wrote Microsoft's president Brad Smith and chief sustainability officer Melanie Nakagawa in the report's foreword. They continued:

The rise in our Scope 3 [indirect] emissions primarily comes from the construction of more datacenters and the associated embodied carbon in building materials, as well as hardware components such as semiconductors, servers, and racks. Our challenges are in part unique to our position as a leading cloud supplier that is expanding its datacenters. But, even more, we reflect the challenges the world must overcome to develop and use greener concrete, steel, fuels, and chips. These are the biggest drivers of our Scope 3 challenges.

To mitigate the impact of its datacenter expansion plans, Microsoft said it's making an effort to source more "clean" construction materials. It's also trying to make its existing datacenters more energy-efficient, for instance by putting underused and unallocated servers into low-power states. To reduce water consumption in its datacenters, it's harvesting rainwater or using recycled water in regions that support it.

"Our new datacenters are designed and optimized to support AI workloads and will consume zero water for cooling," the report said. "This initiative aims to further reduce our global reliance on freshwater resources as AI compute demands increase."

Microsoft is also looking for ways to use AI to fix a problem that AI, at least partially, wrought. For instance, it's using AI to measure the energy efficiency of its datacenters, as well as their impacts on surrounding ecosystems.

"Our science, research, and AI for Good teams are also working to accelerate solutions and develop climate resilience with AI," Smith and Nakagawa added. "Through our AI for Good team, we are collaborating with the United Nations to research the use of AI to advance the Early Warning for All Initiative, with a goal of better understanding the populations that may be at risk of extreme weather events and other threats."

More information on Microsoft's 2024 sustainability report is available here.

About the Author

Gladys Rama (@GladysRama3) is the editorial director of Converge360.