Amidst growing artificial intelligence (AI) demand and rising costs for chips and data centers, four major US technology giants did not signal any reduction in investment in their Wednesday earnings reports. Instead, they further raised their capital expenditure forecasts. Data shows that Microsoft, Amazon, Meta, and Google have increased their combined capital expenditure for this year to $725 billion—primarily allocated to AI data center equipment—surpassing the pre-earnings market expectation of $670 billion.
Prior to these earnings releases, reports that OpenAI failed to meet its internal user growth and sales targets had raised concerns about whether major tech firms would firmly execute their previously announced massive capital expenditure plans for building AI infrastructure. The latest expansions in AI-related capital spending by these tech giants have, to some extent, reassured investors.
All four tech giants reported on the same day, signaling continued commitment to AI investment. Specifically, Google indicated during its earnings call that it expects full-year 2026 capital expenditures to be between $180 billion and $190 billion, raising both the upper and lower ends of the range by $50 billion. The company also anticipates that capital expenditures will "increase significantly" in 2027.
Microsoft similarly stated during its earnings call that it expects 2026 capital expenditures to reach $190 billion, with approximately $25 billion attributed to rising component costs. This figure far exceeds the market's average expectation of $117.5 billion. Microsoft expressed confidence that these investments will yield returns, citing stronger demand signals, increased product usage, and ongoing platform efficiency improvements.
Amazon noted that its capital expenditure plans remain "broadly consistent" with the expectations provided in January. The company had previously projected 2026 capital expenditures to approach $200 billion, a 56% increase from 2025—primarily directed toward data centers, including those customized for AI services.
Meta raised its full-year 2026 capital expenditure forecast to a range of $125 billion to $145 billion, increasing both ends by $10 billion. The company attributed the upward revision to expected increases in component prices this year, as well as rising data center costs to support future capacity.
Lee Sustar, an analyst at Forrester Research, commented, "With the potential rewards of AI leadership so compelling, these companies continue to place their bets, forcing investors and customers to assess how their interests are affected."
Market reactions to the increased capital expenditures varied among the four tech giants. Following its earnings release, Meta's stock fell more than 8% in pre-market trading on Thursday. This reflects investor caution regarding short-term cost expansion and return cycles, despite general acceptance of the long-term AI narrative.
For Meta, the impact of AI-driven ad targeting was evident in its advertising revenue, which grew 33% year-over-year. However, this was not sufficient to justify the increased capital spending in the eyes of investors. Unlike the other three cloud hyperscalers, Meta does not sell AI computing services to customers, and its consumer AI application rollout has been relatively slow. Analyst Mandeep Singh noted, "Meta’s standalone apps haven’t garnered enough engagement."
Moreover, the market paid closer attention to Meta’s justification for raising capital expenditures—citing "component price increases and additional data center costs"—rather than citing increased demand as the reason for expansion. These two explanations send very different signals: the former implies paying more for the same infrastructure. Barclays estimates that Meta’s free cash flow could decline by nearly 90% this year.
Although Meta CEO Mark Zuckerberg expressed confidence in the decision to increase investment, his responses to analyst questions were somewhat vague. During the earnings call, he stated that Meta does not have a "very precise plan" for developing each AI product, adding, "I think we have a general sense of the direction things are going," while acknowledging that his answer might be "disappointing."
In contrast, Google’s stock rose nearly 6% in pre-market trading on Thursday after its earnings release. A key driver was the stronger-than-expected growth of its Google Cloud business, providing a more solid foundation for continued AI infrastructure investment. Google Cloud revenue reached $20 billion in the first quarter, significantly exceeding analyst expectations of $18.4 billion, with year-over-year growth accelerating to 63%. The company attributed the "growth acceleration" to strong demand for its AI software and infrastructure.
CEO Sundar Pichai stated during the earnings call, "Our AI models have strong momentum. We are bringing useful AI to billions of people every day through our products and platforms."
Meanwhile, Microsoft’s stock fell nearly 2% in pre-market trading. Although revenue growth for Azure and other cloud services accelerated to 40%, and its enterprise software business remained stable, Microsoft’s capital expenditure growth slowed significantly in the first quarter—decelerating from 66% in the previous quarter to 49%, against analyst expectations of approximately 65% growth.
Analysts suggested that the pre-market decline in Microsoft’s stock does not reflect weak earnings but rather a reassessment of capital expenditure, cloud profit margins, and future guidance uncertainties. While Microsoft’s fundamentals remain strong, the market has grown more sensitive to cost growth and future guidance amid crowded AI trades and high valuation expectations. Investors are not only focused on whether quarterly earnings beat expectations but also on when AI investments will translate into sustainable free cash flow and higher margins.
Amazon’s stock initially fell after its earnings release but later turned positive, rising more than 2% in pre-market trading on Thursday. Analysts noted that initial investor concerns centered on Amazon’s AI capital expenditures significantly exceeding expectations and squeezing free cash flow. Attention later shifted to reaccelerating AWS growth, strong second-quarter revenue guidance, and near-term growth certainty driven by an earlier Prime Day event.
The AI investments of these four tech giants are primarily directed toward high-performance chips, data centers, and related infrastructure, with Nvidia remaining a key beneficiary. Combined with Nvidia’s strong quarterly results released in late February and recent signals from chip manufacturers like TSMC, there are few signs of major cloud providers slowing AI investment.
At the same time, tight supply of memory chips and other low-power semiconductors has revitalized interest in some traditional tech hardware companies. Firms like Sandisk, Western Digital, and Intel have recently performed strongly, while Seagate’s stock rose about 11% after its Wednesday earnings report—the latest example of this trend. This indicates that the AI investment chain is no longer limited to GPUs and advanced process chips but is expanding to memory, servers, data center equipment, and related supply chains.
However, the continued expansion of AI capital expenditures is not without costs. Companies like Meta and Microsoft are streamlining certain teams. Additionally, after years of relying on strong free cash flow to fund expansion, major tech firms are increasingly turning to debt markets to finance AI investments.
In recent months, the structure of AI-related trades has shifted rapidly. Sentiment toward software stocks has cooled noticeably, and market enthusiasm for AI startups like OpenAI and Anthropic has fluctuated. Yet, for the world’s largest tech companies, the strategic outlook remains clear: the window of AI opportunity is still open, and the greatest risk is not overspending but underinvesting.
Based on the latest moves by Microsoft, Amazon, Meta, and Google, the race in AI capital expenditure is far from over. For the market, this means that the AI infrastructure supply chain continues to have earnings support. It also helps alleviate investor concerns sparked by OpenAI’s growth slowdown—where its CFO, Sarah Friar, indicated that if sales growth does not accelerate sufficiently, the company may struggle to afford necessary computing costs.
At the same time, however, investors are increasingly focused on a more practical question: when will this multi-hundred-billion-dollar AI gamble truly deliver returns?
Comments