Artificial intelligence is often presented as a technology that could make the U.S. economy more productive, reduce business costs and eventually help bring down prices. But in the short term, the rapid expansion of AI infrastructure is creating a different economic effect: stronger demand for electricity, semiconductors, construction and other resources that are already under pressure.

That has raised a question for economists and Federal Reserve officials: Could the AI investment boom itself be adding to inflation?

Recent comments from Chicago Federal Reserve President Austan Goolsbee have brought the issue into sharper focus. Speaking in London on September 21, Goolsbee said strong demand was becoming a more important contributor to U.S. inflation, including demand associated with booming AI investment. He specifically pointed to the possibility that data-center construction could push economic activity beyond what the economy can currently absorb.

Building AI Is Creating a New Wave of Demand

The AI boom requires far more than software. Technology companies are spending heavily on data centers, advanced chips, power systems, cooling equipment and electricity.

That surge in spending is already showing up in producer prices. Research from the Federal Reserve Bank of Richmond found that prices for semiconductor and electronic components had reached unusually high levels, while prices for equipment used to generate and transmit electricity were also rising. The research also found elevated prices for some software and engineering services connected to AI and data-center development.

Electricity is another potential pressure point. A Federal Reserve Bank of Dallas analysis estimates that a plausible data-center expansion could increase annual U.S. PCE inflation by between 0.04 and 0.13 percentage points by 2030, largely through higher electricity prices. The researchers note that the effect could be considerably larger if data centers expand faster than expected or renewable energy development fails to keep pace.

The pressure is not necessarily limited to technology companies. When AI companies compete for chips, electricity and specialized equipment, they can increase costs for other businesses competing for the same resources.

The Inflation Story Could Eventually Change

The unusual part of the AI economy is that the same investment creating inflationary pressure today could potentially reduce it in the future.

AI is expected to allow businesses to produce more with fewer resources, potentially increasing productivity and reducing production costs. Research from the U.S. Bureau of Economic Analysis has already found an association between greater AI intensity and lower prices charged to customers, with some of the reduction linked to lower labor and materials costs.

But economists caution that those productivity benefits may take time to become large enough to offset the enormous spending required to build the AI infrastructure.

The Minneapolis Federal Reserve recently described the current situation as essentially a race between investment-driven demand and future productivity gains. Its analysis notes that AI-related investment is creating price pressures now, while the broader productivity benefits remain uncertain in timing and scale.

That distinction matters for monetary policy. If AI investment continues increasing demand faster than the economy can expand its supply of electricity, equipment, construction capacity and other resources, inflationary pressure could persist. If productivity eventually accelerates significantly, the effect could move in the opposite direction.

What It Means for the U.S. Economy

The AI boom is therefore not simply an inflation story or a productivity story. It could be both, but at different stages.

For now, Federal Reserve researchers and officials are watching whether AI-related investment begins spilling into broader prices and demand. Recent research from the Minneapolis Fed found that AI-driven demand for computer hardware has already contributed to higher goods inflation, while other research points to potential future pressure from electricity costs.

The central economic question is whether the productivity gains eventually arrive quickly enough to outweigh the costs of building the AI economy.

If they do, AI could ultimately help ease inflation by expanding the economy's productive capacity. Until then, America’s AI boom may be adding another layer of demand to an economy that is already dealing with elevated prices.