AI Data Center Boom Faces Power and Infrastructure Bottlenecks Despite Trillion-Dollar Spending

The artificial-intelligence boom is driving enormous investment in data centers, but the industry's biggest challenge may not be money.

AI data centers require massive amounts of computing equipment, electricity and cooling infrastructure as technology companies accelerate investment.

Power, computing equipment, construction capacity and skilled labor are emerging as major constraints, Julie Hyman reported for Yahoo Finance.

Goldman Sachs estimates that global AI-related investment could reach about $1 trillion in 2026, including roughly $581 billion in the United States. JPMorgan forecasts about $697 billion in U.S. spending, while Bank of America sees a path toward roughly $1.2 trillion in AI investment by 2027. 

The spending covers far more than GPUs. Companies need land, buildings, servers, cooling equipment and electrical infrastructure. Shortages of memory chips and other components can also delay projects.

Construction capacity and skilled labor are additional constraints.

At the same time, communities and regulators are increasingly challenging data-center developments because of their enormous electricity requirements.

Power may be the biggest obstacle. BloombergNEF estimates that U.S. data centers could consume about 20% of the country's electricity by 2035.

It projects data-center power demand of 194 gigawatts by that year and estimates a potential 19-gigawatt shortfall under its base-case scenario. 

The figures underscore the central challenge facing the AI infrastructure boom: enormous amounts of capital can fund new facilities, but money alone cannot immediately create electricity capacity, transmission infrastructure, construction crews or semiconductor supply.