Microsoft Plans 38GW Data Center Expansion by 2032 Amid AI Demand

Microsoft is planning to expand its global data center capacity to more than 38 gigawatts by 2032, more than tripling its current footprint of roughly 12 gigawatts. The planned expansion covers both Microsoft-owned and leased facilities and is aimed at addressing shortages that have limited the company’s ability to provide additional AI and cloud computing capacity.

Only about 2 gigawatts of Microsoft’s current capacity is dedicated to systems using AI-specific chips, according to the reported figures. That share is expected to rise to roughly one-third of the planned 38GW network, pointing to a substantial increase in infrastructure built specifically for AI workloads.

AI Demand Drives the Expansion

The planned buildout comes as Microsoft tries to keep pace with demand for Azure cloud services and AI applications. The company has previously said it is adding data center capacity at an unprecedented scale and expects to roughly double its overall footprint over a two-year period.

The new target indicates that the longer-term expansion will be considerably larger. Microsoft has faced periods when limited computing capacity prevented it from accepting some AI and cloud business, making access to land, electricity, servers and advanced AI chips increasingly important to its infrastructure plans.

Power Becomes a Major Constraint

Adding 26GW or more of capacity over the coming years will require a substantial increase in electricity supply. Data centers running AI workloads consume large amounts of power, particularly when high-density GPU systems operate continuously for training and inference.

Microsoft has already been working to secure additional energy and improve the efficiency of its existing infrastructure. The company's global data center network currently spans more than 500 facilities, with more than 40GW of new renewable energy contracted across 26 countries, according to Microsoft.

Microsoft Is Also Optimising Existing Capacity

The company is not relying solely on new construction to address the shortage. Microsoft has been improving the performance of its existing infrastructure through changes to its chips, servers, networking and software.

Microsoft said it added another gigawatt of capacity in its latest reported quarter and remains on track to roughly double its overall capacity within two years. It has also been increasing GPU efficiency and inference throughput to get more computing output from the infrastructure already in operation.

AI Infrastructure Race Intensifies

The planned 38GW network would put Microsoft among the world's largest individual buyers and operators of data center infrastructure. The expansion also shows how the AI boom is shifting competition beyond model development toward access to electricity, computing hardware and physical infrastructure.

The immediate challenge is execution. Microsoft will need to secure enough power, land, servers and AI accelerators to bring the planned facilities online. For now, the 38GW figure remains a reported long-term target rather than a formally announced company commitment, but it illustrates the scale of infrastructure Microsoft believes will be needed to support its cloud and AI businesses through 2032.