Key Takeaways
-
Stock prices for Amazon and Alphabet have faced declines this year due to concerns about overspending on AI infrastructure.
-
Despite these worries, both companies have strong justifications for investing heavily in this area.
Analyst Nick Jones from BNP Paribas recently issued an optimistic outlook for Alphabet (NASDAQ: GOOGL) and Amazon (NASDAQ: AMZN), asserting that concerns regarding AI infrastructure spending are “exaggerated.” The issue of spending in data centers has garnered significant attention this year as investors are questioning its financial viability.
Notable investor Michael Burry, featured in the film The Big Short, voiced skepticism last fall, arguing that the durability of graphics processing units (GPUs) and other AI chips is limited. He believes this means that major players like Alphabet and Amazon may not realize financial gains from their investments.
Rising Capex Sparks Investor Concerns
Nonetheless, this skepticism did not prevent Alphabet and Amazon from significantly increasing their capital expenditure (capex) plans for the year. In February, Alphabet announced a capex increase to between $175 billion and $185 billion, up from $91.4 billion in 2025. Shortly thereafter, Amazon revealed it would raise its capex to $200 billion from $131.8 billion the previous year. These announcements prompted a decline in stock prices for both firms.
Amazon and Alphabet are key players in the cloud computing market alongside Microsoft. Amazon boasts the largest market share, having pioneered the infrastructure-as-a-service sector with Amazon Web Services (AWS), which remains its most profitable and fastest-growing segment. Alphabet’s Google Cloud ranks third in market share, following Microsoft’s Azure.
In Jones’s analysis, he mentions that the backlog-to-capex ratios for both companies indicate they are not overspending and that their investments align with surging demand. He also highlighted that both companies are enhancing their efficiency, as evidenced by increasing revenue per employee metrics. Jones has set a price target of $390 for Alphabet, suggesting about a 30% increase, and a $320 target for Amazon, indicating a potential 50% gain.
Advantages of Custom Chips
In my opinion, Alphabet and Amazon are well-positioned to ramp up their AI infrastructure expenditures for two main reasons. First, both have created specialized AI ASICs (application-specific integrated circuits) tailored for specific AI tasks. These purpose-built chips generally offer greater energy efficiency and lower total ownership costs, thereby giving both companies a structural cost advantage.
Alphabet leads the field with its tensor processing units (TPUs), developed over a decade ago, integrated throughout its hardware and software ecosystem. Amazon has also crafted cost-effective chips, such as Trainium and Inferentia. Both companies have seen increased customer adoption of these chips in their data centers.
