The combined capital expenditure of the world's nine largest cloud service providers (CSPs) is anticipated to hit $886.7 billion, approximately 1,263 trillion Korean Won, according to recent market expectations. This year's spending by hyperscalers on AI data expansion is up about 90% from the previous year. Looking ahead, total capital expenditure is projected to grow to $1.3 trillion (approximately 1,849 trillion Korean Won) next year as competition to expand AI data centers intensifies.
Despite the significant investment, profitability outcomes among these tech giants are diverging. Microsoft, for example, saw its cloud business revenue surpass $100 billion for the first time and its AI business run rate exceed $37 billion, leading to a nearly 9% surge in its stock. This demonstrates concrete, direct monetization from its capital expenditures, reassuring investors.
In contrast, Google and Meta have experienced challenges. Meta's free cash flow plunged 91% to $784 million due to massive AI infrastructure investments and heavy legal charges, causing its stock to tumble roughly 9%. Similarly, Google saw its free cash flow plunge, with market skepticism noted despite a rebound in its stock. Analysts highlight that companies' ability to manage profitability is diverging even amidst expanding AI investment, with some firms struggling to generate cash flow from their heavy AI spending.
Overall, the four largest players in the data center race—Alphabet, Meta, Microsoft, and Amazon.com—have committed nearly $2.4 trillion in spending over the coming years on leases, buildings, energy, and other equipment to build fleets of data centers. However, there is growing scrutiny regarding these massive AI capital expenditures, with concerns about the return on investment and the potential for escalating depreciation charges to weigh on future earnings. While some analysts believe the return on incremental invested capital is still healthy, others question the sustainability and ultimate benefit of such extensive spending.
