Corporate AI Adoption Fuels Demand Surge For World's Largest Data Center Provider

Data center REIT Equinix is emerging as one of the main beneficiaries of a fundamental shift in how major companies are deploying their artificial intelligence dollars.

On Wednesday, Equinix reported the largest quarter of leasing activity in its history, with first-quarter revenues of $2.3B representing a 10% year-over-year increase. While revenues were slightly below Equinix’s own projections, the firm attributed the earnings miss to a major hyperscale lease closing a quarter later than expected.

The company raised its full-year revenue expectations by $21M. 

Equinix is already the world’s largest data center provider, with 281 data centers in operation across 77 markets. But the REIT's leadership said it is about to enter a period of accelerated growth as corporate adoption of AI hits an inflection point that will send a greater share of demand flowing in its direction. 

Despite previous concerns that corporate AI-driven data center demand could slow if early spending failed to deliver returns, enterprise customers are instead doubling down on AI adoption, Equinix CEO Adaire Fox-Martin said Wednesday.

Far from pulling back, corporate customers have shifted from deploying AI infrastructure in a limited, experimental capacity to deploying it at scale as they embrace AI as a central element of their operations. 

Fox-Martin said Equinix is uniquely positioned to capitalize on this expected surge in enterprise demand. 

“Over the course of the past year, my conversations with customers have changed,” Fox-Martin said on the REIT's Q1 earnings call. “A year ago, they were about piloting AI. Now our conversations are focused on enterprisewide adoption — adoption at scale.”

Corporate adoption of AI is already driving a sharp rise in demand for inference computing, the systems through which users interact with AI models in real time. Inference workloads now account for roughly 60% of Equinix’s new deployments, a share it expects to continue growing as new enterprise use cases emerge. 

A major driver is the growing adoption of agentic AI, autonomous systems that can plan and take initiative to achieve defined goals instead of simply responding to prompts.

These tools are beginning to be implemented by large organizations for a range of tasks, from onboarding employees by coordinating meetings, provisioning accounts and arranging equipment to representing buyers and sellers in multistage negotiations.

Fox-Martin said early adopters are already deploying such systems in Equinix data centers, including a pharmaceutical company using AI agents for molecular simulations in drug discovery and a construction firm leveraging AI to conduct worksite risk assessments based on drone footage.

Most companies implementing these AI tools are doing so using hybrid infrastructure that utilizes cloud computing and self-managed data centers. But traditional corporate data center infrastructure, whether self-operated or leased from colocation providers, is typically not well suited to supporting these new use cases.

Inference workloads typically need to be close to large population centers where the users are and in facilities where data can be transferred easily between different cloud providers and networks. Often, the high-performance computing required for these AI systems isn't compatible with older data centers and needs to be hosted in facilities with liquid cooling and other AI-focused design elements.

Yet these requirements align closely with Equinix’s core business model, as well as that of fellow REIT Digital Realty. The companies’ leaders have long argued that their data center portfolios are positioned to dominate the emerging inference marketplace. 

While much of the AI data center boom has taken the form of massive campuses far from major metro areas, Equinix’s portfolio has remained focused on blocks of capacity close to major population centers with strong interconnection to cloud providers, multiple fiber networks and other customers. Few other companies can offer this ecosystem tailor-made for hybrid AI deployments in major markets around the world. 

Equinix also has facilities capable of hosting high-performance computing. The company reported 50% growth in liquid cooling deployments last quarter, with 36 liquid-cooled facilities across the firm’s data center footprint. 

Equinix is also seeing demand from hyperscalers that need to deploy cloud computing edge nodes to serve these same enterprise inference customers.

While companies like Amazon, Google and Microsoft — as well as neocloud firms like CoreWeave — are building massive data center campuses on their own, they are turning to Equinix and other major colocation providers to deploy capacity for inference that has to be close to customers in major population centers. 

“This is a market that we are built to serve,” Fox-Martin said. “We believe there is meaningful upside to come, given we are still in the early days of the agentic AI wave and inferencing adoption.”

Equinix is pursuing a development strategy with inference in mind. The firm has 46 major projects underway across 32 markets, and 70% of that expansion capex is being spent in major metro areas. 

The company has prioritized purchasing land and power near major population centers. Fox-Martin has touted what she says is the firm’s unique ability to deliver new capacity in the markets facing significant power constraints, pointing to 3 gigawatts of powered land under the firm’s control. 

“We continue to meaningfully grow our pipeline for new powered land and capacity expansion opportunities that can enhance our long-term growth prospects in key metros,” Fox-Martin said Wednesday.