Amazon.com (NASDAQ: AMZN) latest earnings call delivered a strong signal that the demand for data centers from hyperscale cloud providers is far from peaking, which is a positive development for leading stocks in the data center infrastructure sector.

On July 30, after the U.S. stock market close, Amazon CEO Andy Jassy announced during the second-quarter earnings call that the company's 2026 cash capital expenditure guidance has been raised by $20 billion to $220 billion, marking a 71.4% year-over-year increase.

More notably, Jassy explicitly stated that even with this level of investment, this year's capacity will still fall short of meeting all demand, and he expects this supply-demand gap to persist into 2027.

According to a trading desk note, Barclays promptly released a research report arguing that these comments effectively counter the market's previous concerns about overinvestment in AI infrastructure. The report suggests that demand for hyperscale infrastructure could continue to grow even after the current AI training cycle matures.

However, Barclays pointed out that changes in the macroeconomic environment and a decline in AI enthusiasm are the main risks to the valuation of these stocks. Fluctuations in the U.S. dollar, interest rates, and energy costs could all impact the earnings and valuation outcomes of related companies.

Supply-Demand Gap Persists, Capacity Constraints Extend Through 2026 to 2027

Andy Jassy says the company will further increase its investment in AI, with this year's capital expenditure expected to reach $220 billion. In February, Amazon stated that this year's capital expenditure would be $200 billion, and it maintained this forecast unchanged in April.

Andy Jassy noted that rising memory prices have pushed up Amazon's capital expenditure expectations. He believes that the company's spending spree is unlikely to slow down in the near term. Andy Jassy stated:

"Even so, our capacity this year will still be unable to meet all demand. I believe this situation will also exist in 2027."

According to Barclays' analysis, this assessment confirms that demand from hyperscale cloud providers remains exceptionally strong, with the entire industry still operating in a capacity-constrained environment. This supports ongoing demand for data center development, power-available land, and wholesale capacity through 2027 and beyond.

In terms of expansion pace, Amazon said it is progressing with capacity building as planned. The company has previously disclosed a goal to double its capacity and power capacity by the end of 2027 compared to 2025, and current progress is on track.

2028 Demand Visibility Already Unusually High, Enterprise Inference Applications Still in Early Stages

Amazon's view on future demand extends beyond 2027. Jassy pointed out that demand signals from 2028 are already "compelling" and emphasized that enterprise customers are still in the very early stages of using AI inference on a large scale in their existing production applications.

According to Barclays' report interpretation, combined with the assessment that enterprise inference adoption is still in its early stages, these comments suggest that demand for hyperscale infrastructure could continue to grow even after the current AI training cycle matures.

This structural long-tail nature of demand provides data center operators with longer-term revenue visibility.

In terms of capital allocation logic, Jassy characterized data centers as long-duration infrastructure assets that transcend single technology cycles. He stated during the earnings call:

"Capital expenditure for data centers begins two years before servers are installed and start generating revenue. Once a data center is operational and servers are connected, we immediately begin generating substantial revenue, and we can continue to monetize it over a period of more than 30 years without needing to reinvest the initial startup capital."

He further elaborated on the economics of multi-generational servers:

"For a data center with a useful life of more than 30 years, we can capture the economic benefits of at least five to six generations of servers. The economics of each generation after the first are even better because there is no need to repeat the initial data center construction investment."

Barclays believes this framework helps explain why hyperscale cloud providers are willing to accept large upfront capital expenditures. Once a data center is built, it can generate returns across multiple server generations, with the marginal benefit of the initial investment increasing over time.