AWS is the latest hyperscaler reported to be scaling back its plans for leasing new colocation capacity — but the moves should not affect enterprises for now, analysts say.
A slowdown in data center lease talks by two of the larger public cloud service providers — AWS and Microsoft — is unlikely to have an immediate effect on enterprises running advanced workloads in the cloud.
“The pause in data center lease talks is unlikely to significantly disrupt enterprise ability to scale generative AI, agentic AI, or optimize AI workloads in the short term,” said Ron Westfall, research director at Futurum Group.
Questions about a data center slowdown arose earlier this week when various media outlets quoted Wells Fargo analysts as saying: “AWS has paused a portion of its leasing discussions on the colocation side (particularly international ones).”
The analysts compared the AWS move to a slowdown in data center spending at Microsoft in recent months. In late February and March, reports emerged that the hyperscaler was not moving forward with some of its leasing plans.
AWS Vice President for Global Data Centers Kevin Miller took to LinkedIn to reassure customers that the company is always exploring more options than it needs to meet demand, and some of those options inevitably don’t work out. “This is routine capacity management, and there haven’t been any recent fundamental changes in our expansion plans,” he wrote.
The moves followed bullish statements by both companies about their data center investment plans.
In early February, AWS had said that it intended to invest $100 billion in ramping up infrastructure for AI cloud services this year, surpassing the spending plans of rivals Microsoft and Google.
The previous month, Microsoft president Brad Smith had reiterated that the company was on track to invest around $80 billion this fiscal year to build out AI-enabled data centers buoyed by near term demand of AI workloads.
If the pace of those investment plans is slowing, it could be for a number of reasons, said Matt Kimball, vice president analyst at Moor Insights and Strategy. He pointed to a combination of potential factors: the hyperscalers being over-aggressive in recent months in terms of meeting demand for AI workloads; the rise of local cloud providers in Europe and elsewhere; demand for these local providers to meet data sovereignty requirements; and companies waiting for US tariffs to stabilize.
A sign of trouble to come?
But, said Forrester senior analyst Alvin Nguyen, it could also be a sign that the hyperscalers are positioning themselves for a larger market fallout, possibly driven by energy and cooling requirements.
In case of a larger market fallout, Nguyen said that enterprises will see limited options, forcing them to use more AI cloud services or partners.
“Costs will increase for those global enterprises in terms of supporting more partners or vendors even if the cost of the cloud AI services cost the same,” Nguyen said.
Brian Alletto, director of West Monroe’s technology and experience practice, didn’t expect the leasing pause at AWS to hurt the hyperscaler’s business.
“AWS has traditionally done a good job of anticipating customer needs through various signalling mechanisms — account team engagement, discount programs, reserved instances, and savings plans — as well as its core capacity management and expansion practices,” he said.
And even if AWS does somehow find itself unable to meet customer demand, there are workarounds for enterprises, said Futurum Group’s Westfall.
“Other ways to bypass localized infrastructure challenges would be to use spot instances (unused EC2 compute capacity) for cost-effective scaling, or optimizing workloads to reduce resource demands,” he said, adding that CIOs are unlikely to worry about simpler AI workloads, such as inference of lightweight machine learning (ML) models as the impact is likely negligible due to lower compute requirements
An opportunity for Google, Oracle, and Meta?
While the larger public cloud service providers are slowing down their data center expansion plans, other players, such as Google, Oracle and Meta are still expanding aggressively and analysts don’t seem them slowing down.
“Oracle, Google, and Meta are unlikely to significantly slow down their data center footprint expansion in 2025, since their strategies are driven primarily by strong AI workload demand, cloud services, and digital infrastructure capabilities,” Westfall said.
Oracle, according to Westfall, is committed to investing $10 billion in 2025 to build 100 new data centers and expand 66 existing ones, aiming to double its capacity this year.
Likewise, Google is investing $75 billion in 2025 for data center construction, focusing on AI and cloud infrastructure, with projects such as a $600 million facility in Mesa, Arizona, and a $2 billion data center in Fort Wayne and Indiana underway, Westfall said.
Meta, too, plans to spend up to $65 billion in 2025, a sizable bump up from $40 billion in 2024, primarily for data center expansion to support AI (Llama models, Meta AI) and metaverse workloads, Westfall added.
However, these expansion plans will not result in the relatively smaller players catching up with AWS and Microsoft.
“For smaller players like Google and Oracle, catching up with AWS and Microsoft would require historically large capital investments that likely aren’t justified by their current growth rates,” Alletto said.
