3 Ways AI Could Keep Inflation Elevated
The surprise behind sticky inflation may be the AI industrial boom itself.
Everyone expects artificial intelligence to eventually make businesses more productive, lower costs and help tame inflation. Eventually may be the key word.
One of the more surprising forces keeping inflation and long-term interest rates elevated may have little to do with oil, war, tariffs or the federal debt. It may be the enormous industrial buildout required to power the AI revolution, with trillions of dollars expected to flow into AI infrastructure over the next several years.
For decades, technology was one of inflation’s best friends. Computers, software and electronics kept getting faster and better while prices generally fell. It was one corner of the economy the Fed rarely had to worry about.
That relationship is starting to change. As MarketWatch recently observed, high-tech prices that fell for years are now moving in the opposite direction as the AI boom creates enormous demand for chips, memory and computing capacity. The technology expected to eventually lower costs across the economy may be raising some of them first.
AI Is Hardly Weightless
Every prompt connects to a physical economy of data centers, electricity, natural gas, copper, steel, cooling systems and transformers. Before AI delivers its promised productivity gains, somebody has to build and power all that infrastructure.
That creates an interesting possibility for investors. What if AI eventually becomes a powerful deflationary force, but first helps keep inflation closer to 3% than the Fed’s 2% target over the next couple of years?
That could leave the Fed with less room to cut rates and keep long-term Treasury yields elevated. If inflation heats up again, rate hikes could even come back into the conversation. Yet the same AI spending contributing to those pressures could continue supporting economic growth, corporate earnings and stocks.
That is the paradox. AI may eventually help solve the inflation problem while making the Fed’s job harder getting there.
The bond market is already reminding investors that the Fed does not control the entire yield curve. Long-term Treasury yields have climbed to their highest levels since 2007, with the 30-year Treasury yield recently reaching about 5.33%. (Source: MarketWatch, Aug. 18, 2026)
If inflation remains sticky, long-term borrowing costs including mortgages could stay elevated regardless of what the Fed eventually does with short-term rates. This will also provide headwinds to both bonds and stocks. Here are three reasons we think AI could keep inflation elevated
1 Technology Stopped Getting Cheaper
For roughly 25 years, consumer technology was one corner of the economy the Fed rarely had to worry about. Computers, software and electronics generally became faster and better while prices moved in the opposite direction. Better technology usually meant cheaper technology.
That relationship is starting to flip.
The AI boom is creating shortages and pushing up the cost of memory, chips and other components used in laptops, smartphones, routers and other electronics. Apple recently raised prices across its Mac and iPad lineups, citing soaring memory and storage costs resulting from the rapid expansion of AI data centers. (Sources: Reuters, June 25, 2026; Apple, as reported by Reuters)
The shift is showing up in the inflation data. After roughly a quarter-century in which computer software and accessory prices generally declined, prices in the category jumped 14.5% year over year in May, marking a dramatic reversal of that long-running deflationary trend. (Sources: U.S. Bureau of Labor Statistics, as reported by Axios, June 11, 2026; Federal Reserve Bank of Richmond, July 2026)
Economists estimate that higher prices for computer-related equipment could add roughly 0.1 to 0.2 percentage points to annual inflation. That may not sound like much, but when the Fed is trying to move inflation back toward its 2% target, every tenth matters. (Source: MarketWatch, August 2026)
None of this means AI is suddenly the primary cause of inflation. But it does challenge the assumption that a technology and productivity boom must immediately be deflationary.
And the inflationary footprint of AI goes well beyond the devices sitting on our desks and in our pockets.
2 AI Doesn’t Actually Live in the Cloud
The cloud sounds airy. The infrastructure behind it is concrete, steel and electricity.
Data-center construction spending has accelerated sharply since the launch of ChatGPT in late 2022. The chart below shows annualized data-center construction spending near $59 billion, now above roughly $48 billion for other office construction. And that figure does not include the servers and other IT hardware inside the buildings.
This is where the AI s
This is where the AI story starts looking less like software and more like an industrial boom. Every new data center creates demand across utilities, energy, materials, construction and infrastructure. Productivity may arrive later. The capital spending is happening now.
The scale can be difficult to appreciate. OpenAI’s Stargate campus in Texas is expected to contain roughly 4 million square feet across eight buildings. Meta’s expanding Hyperion campus in Louisiana could approach 10 million square feet while delivering up to 5 gigawatts of computing capacity. (Sources: Crusoe, March 18, 2025; Meta and Louisiana Economic Development, July 2026)
And even those numbers understate the footprint. These campuses require substations, transmission lines, cooling systems, backup generation, roads and vast amounts of surrounding land.
AI may appear to live on your laptop. Increasingly, the infrastructure behind it looks more like a small industrial city.
There is a bit of Field of Dreams economics at work here. “If you build it, they will come” may prove surprisingly appropriate for AI. Build more data centers, expand computing capacity and make AI cheaper to use, and demand may rise right alongside it.
3 Jevons Paradox
Economists have a less cinematic name for that idea: the Jevons paradox.
When technology becomes cheaper and more efficient, we sometimes use more of it, not less. If AI models become dramatically cheaper to run, businesses may deploy them everywhere. Lower costs could lead to far more queries, applications and overall computing demand.
That matters because greater AI efficiency and a massive infrastructure buildout do not contradict each other. AI could require less computing power per task while total demand for computing power continues to rise.
The Wealth Effect
The AI boom may be adding to inflation from the demand side as well.
Rising stock prices have lifted household wealth and retirement accounts, particularly among higher-income Americans. Some of those gains eventually make their way into the real economy through spending on travel, housing, cars and other big-ticket purchases.
Oxford Economics estimates that the wealthiest 20% of Americans now account for more than half of new-car purchases. (Source: Oxford Economics, as reported by MarketWatch, August 2026.)
That creates an unusual two-sided effect. AI may be raising the cost of building the technology while the wealth created by the AI bull market helps support demand.
What Does This Mean for Investors?
We remain constructive on stocks. But the investment lesson is broader than simply owning the biggest technology companies.
The AI value chain runs through semiconductors, data centers, power generation, utilities, industrial equipment, cooling, materials and software. That creates opportunities beyond the Magnificent Seven and helps explain why market leadership can broaden even while AI remains the dominant investment theme.
The Magnificent Seven have delivered extraordinary returns since 2020, but extraordinary returns also create extraordinary expectations. The next phase of the AI trade may be less concentrated than the last one.
Great Technology Can Still Be an Expensive Investment. AI may transform the economy and still produce a bumpy investment cycle. Markets routinely overestimate how quickly new technologies turn enormous capital spending into profits.
That is why valuation matters. Tech companies continue to generate strong earnings growth, but investors are paying higher multiples for that growth. Other sectors offer different combinations of earnings growth, valuations and exposure to the AI buildout. (see chart)
The Interest-Rate Catch
There is another implication investors should not overlook. If the AI buildout keeps demand for capital, electricity, labor and materials elevated, long-term interest rates may remain higher than investors expect, even if the Federal Reserve cuts short-term rates.
That could mean more volatility for both stocks and bonds. It could also favor broader market leadership as capital flows beyond mega-cap technology into industrials, utilities, energy, materials and infrastructure, along with opportunities in international markets.
For bond investors, the lesson from the past five years is equally important. Higher yields do not eliminate interest-rate risk. Duration still matters. With the yield curve offering very different combinations of income and risk, we believe investors should remain disciplined about how much duration, credit and sector risk they take to earn that yield.
The Bottom Line: AI may be adding to inflation today, but it could eventually become one of the greatest deflationary forces of our lifetime.
But getting there requires an extraordinary amount of investment first. The AI revolution is already moving beyond chips, software and a handful of mega-cap stocks into the physical economy. That could influence inflation, interest rates, corporate earnings and market leadership for years to come.
The irony is hard to miss. The technology promising to make everything more efficient may first require one of the largest capital-spending booms in modern history.
Somebody still has to pour a lot of concrete.
For more information on our firm or to request a complementary investment and retirement check-up with Jon W. Ulin, CFP®, please call us at (561) 210-7887 or email jon.ulin@ulinwealth.com.
Note: Diversification does not ensure a profit or guarantee against loss. You cannot invest directly in an index. Information provided on tax and estate planning is not intended to be a substitute for specific individualized tax or legal advice. We suggest that you discuss your specific situation with a qualified tax or legal advisor.
You cannot invest directly in an index. Past performance is no guarantee of future returns. Diversification does not ensure a profit or guarantee against loss. All examples and charts shown are hypothetical used for illustrative purposes only and do not represent any actual investment. The information given herein is taken from sources that are believed to be reliable, but it is not guaranteed by us as to accuracy or completeness. This is for informational purposes only and in no event should be construed as an offer to sell or solicitation of an offer to buy any securities or products. Please consult your tax and/or legal advisor before implementing any tax and/or legal related strategies mentioned in this publication as NewEdge Advisors, LLC does not provide tax and/or legal advice. Opinions expressed are subject to change without notice and do not take into account the particular investment objectives, financial situation, or needs of individual investors.
