Almost 40% of U.S. data center projects are at risk of falling behind schedule, the Financial Times reported Thursday (April 16), citing data from satellite and artificial intelligence (AI) analytics group SynMax.
The company also found that 60% of projects scheduled for next year have not yet begun construction, according to the report.
SynMax uses satellite imagery to track construction progress and cross-checks this information against benchmarks compiled by industry research group IIR Energy to assess likely completion dates, per the report.
Industry executives interviewed by the FT said delays are being caused by problems around permitting, local opposition and shortages in labor, power and equipment.
However, four companies involved in the development of data centers told the FT their projects are on track. When asked about their projects by the FT, OpenAI, Oracle, digital infrastructure and energy platform SB Energy and AI cloud company Nebius said the projects are on schedule.
This report comes at a time when the market for AI infrastructure is booming.
Advertisement: Scroll to Continue
It was reported April 1 that as the demand for AI services continues to surge, companies are facing a shortage of data center space that is restraining their business.
As a result, companies such as Meta, Google and Amazon recently said they plan to spend tens of billions of dollars more this year than they expected to meet the demand of AI.
It was reported on April 10 that alternative asset manager Blackstone is considering a $2 billion initial public offering (IPO) for an acquisition company that would buy data centers.
PYMNTS reported in October that as demand for AI training large models rises, traditional providers such as Amazon Web Services (AWS), Microsoft Azure and Google Cloud are facing limits in available capacity.
This gap has created an opportunity for smaller infrastructure firms known as “neoclouds” that lease clusters of graphics processing units (GPUs) to AI developers and enterprises that need computing resources on short notice.
PYMNTS reported in November 2024 that tech firms and other companies that depend on AI for their daily operations were facing infrastructure bottlenecks cause in part by America’s outdated power grid.
While foreign markets move quickly to build power systems for next-generation computing facilities, American utilities have lengthy deployment pipelines for new electrical capacity.
