Is the AI infrastructure frenzy driving inflation higher? The Fed’s June meeting minutes release key signals

On July 9, 2026, Beijing time, the U.S. Federal Reserve released the minutes of the Federal Open Market Committee (FOMC) meeting held on June 16–17. The record of the first rate-setting meeting chaired by the newly appointed Chair Kevin Warsh showed that all members unanimously agreed to keep the target range for the federal funds rate unchanged at 3.50% to 3.75%—a level that has remained unchanged since December 2025.

However, what truly drew widespread market attention was not the rate decision itself, but a brand-new variable first introduced into the inflation discussion. According to the minutes, AI investment was listed as one of the three major forces that could stoke inflation, alongside the Middle East war and tariffs. This is the first time in Fed history that AI infrastructure investment has been explicitly included in the inflation risk assessment framework. Just a few months ago, AI infrastructure investment was still rarely seen in Fed discussions as a main driver of inflation; now, multiple officials have pointed out that the surge in data center construction and computing spending has become a new source of demand, while the economy’s capacity to supply appears increasingly tight.

Three Transmission Channels: How AI Drives Up Inflation

In the minutes, Fed officials identified three main transmission channels through which AI investment pushes up inflation.

First, chip and hardware costs rise. Tech giants are racing to build AI infrastructure, buying and manufacturing millions of specialized AI computing chips, and constructing data centers equipped with liquid-cooling systems. Strong demand is driving up the prices of semiconductors and electronic components. In the late period of last month, due to shortages of electronic components and a surge in prices, Apple announced that it would raise prices for its MacBooks and iPads by $100 to $300, which at one point caused its share price to plunge by 6%; Microsoft also raised the price of its Xbox consoles by $100 to $150 because component prices skyrocketed. The logic chain in which corporate cost pressure is passed through to end consumers has already become clear.

Second, surging electricity consumption increases energy costs. Running data centers requires a large amount of electricity, putting persistent upward pressure on electricity prices. Multiple Fed officials believe that strong demand for AI infrastructure construction could push up prices of technology products and electricity, thereby intensifying short-term inflation pressure. A survey by the National Association for Business Economics found that 81% of surveyed economists expect AI infrastructure to exacerbate inflationary pressure over the coming year.

Third, investment spending is large in scale and sustained. Unlike one-off shocks from tariffs or oil prices, AI demand is viewed as a structural shock that could last for several years, and much of the spending has yet to be carried out. Hyperscale cloud providers (Alphabet, Amazon, Meta, Microsoft, Oracle) are expected to have capital expenditures of $741 billion in 2026, up nearly 75% year over year. Economists at Columbia University estimate that by the end of 2032, total AI construction spending could reach $8 trillion—nearly five times the total value of real estate in New York City.

The minutes stated: “Several participants commented that price pressures have become more widespread, with most goods and services… seeing substantial increases.” More officials believe that strong business investment spurred by AI infrastructure could become a new force sustaining price pressures.

Triple Inflation Overlap: The Fed’s Policy Dilemma

AI investment is not an isolated source of inflation. The minutes show that Fed officials are facing the combined effects of three sources of price pressure: the Middle East conflict pushing up energy costs, tariff policies raising prices of imported goods, and AI infrastructure investment bringing a new demand shock. These three forces are arriving in overlapping waves, and each one tests the central bank’s policy instinct to ignore one-off price shocks.

Timiraos noted that a year ago the Fed could treat tariff-driven price increases as a one-off shock and remain patient because the labor market was weak at the time. But now hiring is more stable, and energy and AI are simultaneously bringing new cost pressures; continuing to wait entails greater risk—above-target inflation could become entrenched.

The minutes show significant disagreement within the Fed about the future policy path. Among 18 participants, 9 expect at least one rate hike by December 2026; in March 2026, that number was zero. The number expecting rate cuts fell from 12 in March to 1. Another 9 officials expect rates to remain unchanged or for rate cuts to occur. The committee is almost split into two equal halves.

Economic projections released after the meeting showed that 6 of the 19 officials believed two rate hikes are needed. Warsh, who has consistently been critical of forward guidance, refused to submit his own rate forecast. At the post-meeting press conference, Warsh described the policy split as an “internal debate,” but emphasized the importance of restoring price stability and did not release any signal of “remaining patient.”

A June survey by the New York Fed showed one-year inflation expectations rising to 3.7%, the highest since September 2023; three-year expectations touched 3.3%, reaching the June 2022 peak. The Fed’s preferred inflation measure—the PCE price index—is currently near 4%. Participants noted that inflation has risen further and remains far above the committee’s 2% long-run target.

Crypto Market Reaction: Heightened Macro Uncertainty

On the day the Fed minutes were released (July 9, Beijing time), the crypto market did not continue the sharp volatility of the previous trading day and instead showed an overall pattern of stabilization and recovery. According to Gate market data, Bitcoin (BTC) was at $62,610.5, up 0.11% over 24 hours. During the day it fell to a low of $61,546.6 and rebounded to a high of $62,935.1. Its market cap remained at $1.25 trillion, and market dominance rose to 55.42%. Ethereum (ETH) was at $1,750.73, essentially flat over 24 hours (+0.01%), with an intraday trading range of $1,713.48 to $1,758.71, and a market cap of approximately $21.1284 billion. The total global cryptocurrency market cap was about $2.22 trillion. Market sentiment indicators rebounded from yesterday’s “extreme fear” range back to “neutral.”

On a longer-term basis, over the past 7 days Bitcoin fell 7.63%, the 30-day decline was 10.73%, and it has pulled back about 33.74% year-to-date from its historical high. Over the past 7 days Ethereum fell 7.38%, the decline over the past 30 days reached 20.92%, and over the past year it is down 31.14%. Over the past week, both major assets experienced technical rebounds from lows of $69,950 (BTC) and $1,635 (ETH), but resistance overhead remains significant. BTC’s 7-day high was only $69,950.9, far below the 30-day high of $82,828.2, indicating that short-term downside pressure has not been fully released.

From a technical perspective, the BTC upside resistance area is concentrated around $62,935 (the 24-hour high) to $63,137. After an effective breakout, the next target would be $64,546. The short-term support below is at $61,546 (the 24-hour low). If that level is lost again, $60,976 will become the key line of defense for bulls. For ETH, short-term resistance lies in the $1,758 to $1,810 area, and short-term support is in the $1,713 to $1,635 range. After the minutes were released, the market did not show a directional breakout, and trading volume did not expand significantly, reflecting investors’ wait-and-see stance as the policy path becomes clearer.

Market wait-and-see sentiment has deeper logic. The divergence in the policy path revealed by the Fed minutes—whether there will be rate hikes, a hold, or rate cuts—means the future liquidity environment could follow three distinctly different directions. For risk assets, expectations of rate hikes usually imply rising valuation pressure; if inflation pressures eventually ease, keeping rates unchanged or even cutting could bring expectations of improved liquidity.

It is also worth noting that the minutes mentioned a geopolitical variable. Ahead of the June meeting, a temporary agreement to reopen shipping through the Strait of Hormuz had briefly eased concerns about energy prices. But this week, as the U.S. again launched attacks on Iran, the outlook for the Middle East is once again becoming full of uncertainty. The repeated nature of geopolitical risk further exacerbates uncertainty in the inflation outlook.

Latest pricing from federal funds rate futures shows that the market expects the probability of a rate hike at the Fed’s July 28–29 meeting to be about 30%, while the probability of a rate hike at the September meeting is above 50%. Goldman Sachs’ base case is that rates will remain unchanged throughout 2026, but it acknowledges some risk of additional hikes. Citigroup’s view is more dovish, arguing that the market’s pricing for a July hike is “too hawkish relative to the Fed’s reaction function.”

AI’s Inflation Effect: A Battle Between Short-Term Pressure and Long-Term Suppression

Regarding AI’s ultimate impact on inflation, there are two opposing views in the economics community.

In the short term, the surge in AI infrastructure investment is indeed pushing up the prices of specific goods and services. Comments by multiple officials in the minutes confirm this view. The AI data center construction wave is becoming a new structural inflation catalyst by pushing up memory chip prices and electricity consumption.

But in the long term, AI could also suppress inflation by improving productivity. Warsh previously said that AI will suppress inflation in the long run by increasing productivity. In his latest remarks on July 3, he further pointed out that AI model capabilities are growing exponentially, and the resulting expansion in supply capacity will become a new variable that monetary policy must watch. Improved productivity means the economy could achieve faster growth under lower inflation pressure.

However, UBS expects that the cooling effect of AI will take at least several years to become visible. In the short term, the demand shock from large-scale capital expenditures will still be the dominant force. Apollo economists noted that falling oil prices do not necessarily mean inflation will decline; instead, after consumers save on energy spending, they may shift spending to other consumption categories, further boosting demand in an already overheated economy.

Former St. Louis Fed President Jim Bullard said plainly in a CNBC interview that a single rate adjustment is meaningless, and that this round is likely to kick off a full tightening cycle. Bank of America has also raised its forecast, saying it believes the Fed in 2026 could deliver three consecutive rate hikes, each of 25 basis points.

Conclusion

The Fed’s June meeting minutes mark an important expansion of the monetary policy analysis framework. For the first time, AI infrastructure investment has been incorporated into inflation risk assessment, alongside Middle East geopolitical conflicts and tariff policy, making it one of the three major forces that could potentially push the Fed toward rate hikes.

The significance of this change lies not only in short-term fluctuations in inflation data, but also in the identification of structural sources of inflation. AI infrastructure investment is not a one-off shock; it is a structural source of demand that may last for several years. The $741 billion in annual capital expenditures by hyperscale cloud providers and the projected $8 trillion in total investment by 2032—these figures suggest that AI’s impact on the price system will be long-lasting and profound.

For participants in the crypto market, the Fed’s focus on AI inflation risk sends a clear signal: uncertainty in macro policy is rising. Divergence in the rate path, recurring geopolitical risks, and the introduction of AI as a new variable together create a more complex decision-making environment than ever before. Before the next FOMC meeting on July 28–29, the market will closely watch the latest changes in inflation data, the Middle East situation, and AI-related cost developments. The macro narrative is being rewritten, and the pricing logic of crypto assets will accordingly face new tests.

FAQ

Q1: Why did the Fed list AI investment as an inflation risk?

The Fed believes that the surge in AI infrastructure construction is pushing up prices of technology products (especially chips) and electricity, creating a new source of demand while supply capacity is tight. Data center construction requires a large number of computing chips and electricity, and related costs are being passed through to end products. Unlike one-off shocks from tariffs or oil prices, AI investment is viewed as a structural inflation factor that could last for several years.

Q2: What specific rate decision was made at the Fed’s June meeting?

The FOMC unanimously voted to keep the target range for the federal funds rate unchanged at 3.50% to 3.75%. This rate level has remained unchanged since December 2025. The next meeting is scheduled for July 28–29.

Q3: What disagreement do Fed officials have about the subsequent rate path?

Among the 18 participants, 9 expect at least one rate hike before the end of 2026 (no one held this view in March), and 9 expect rates to remain unchanged or for rate cuts. 6 believe two rate hikes are needed. Chair Warsh did not submit his personal rate forecast.

Q4: What impact could AI investment have on the crypto market?

If AI pushes up inflation, it could force the Fed to raise rates and tighten the liquidity environment, creating valuation pressure on risk assets (including cryptocurrencies). At the same time, AI infrastructure requires a large amount of computing power, which could indirectly increase demand for crypto mining chips and energy. The market is closely watching signals from the late-July meeting.

Q5: Will AI suppress inflation in the long run?

Fed Chair Warsh believes AI will suppress inflation in the long run by improving productivity. However, UBS expects that this cooling effect will take at least several years to become apparent. In the short term, the demand shock from large-scale capital expenditures remains the dominant force.