Since the rise of the generative AI trend, the four major technology companies in the United States have been heavily investing in AI-related technologies, with total capital expenditures reaching approximately $1.1 trillion. Analysts warn that Wall Street and investors’ patience may gradually be put to the test as concrete financial returns have yet to clearly materialize.
According to financial reports released by Alphabet (Google’s parent company), Amazon, Microsoft, and Meta (Facebook’s parent company) in the past two weeks, the cumulative AI-related capital expenditures of the four companies have reached about $1.1 trillion from the beginning of the AI trend in 2023 through June 2026.
In a bid for a leading position, these companies, originally focused on asset-light models, have rapidly shifted to becoming major investors in physical infrastructure. The four tech giants are projected to have a combined capital expenditure of $745 billion this year, primarily for building data centers, purchasing advanced chips, and maintaining the massive power required for operations.
Amazon has raised its annual capital expenditure forecast to $220 billion; Microsoft’s annual spending is expected to exceed $85 billion; Meta alone could invest over $140 billion this year and sign long-term data center and infrastructure leases worth hundreds of billions of dollars.
Furthermore, Google has disclosed a significant increase in financial commitments related to AI investments, with commitments related to new contracts involving technology infrastructure and long-term procurement of data center energy soaring by approximately $500 billion compared to three months ago.
This wave of investment has had a notable impact on corporate cash flow. Several executives have admitted to analysts that AI spending in the coming quarters will continue to compress free cash flow. Free cash flow, a key metric closely watched by the market to gauge how much cash a corporation has available for debt repayment, buybacks, or shareholder returns after deducting operating costs and capital expenditures, has dropped to a decade-low of only about $7 billion for the combined free cash flow of the four tech giants.
Alphabet, Google’s parent company, reported a negative $6 billion in free cash flow in a quarter with revenue of $118 billion, marking the first time in over twenty years since going public that the company has had negative free cash flow in a single quarter. Meta’s free cash flow is also around $784 million.
Amazon’s Q2 financial report for this year showed an operating cash flow of $161.4 billion over the past 12 months (a 33% increase year-on-year), but free cash flow turned negative at $7.6 billion (compared to a positive $18.2 billion in the same period last year).
Amazon’s CEO Andy Jassy mentioned that the group will have to endure pressure on free cash flow for some time as they are simultaneously constructing multiple data centers, with approximately a two-year gap from facility commissioning to server installation to ultimately being able to charge customers.
Although each tech giant has introduced its own AI chatbots or tools, most are currently in a situation where investment exceeds returns.
Market responses have varied, with Microsoft and Amazon receiving some investor approval due to solid growth in core businesses. On the other hand, Meta’s stock price has experienced a decline as a result of a lack of a clear AI profit timeline.
Dec Mullarkey, Managing Director at asset management company SLCManagement, told the Financial Times, “For investors now, it’s no longer just about growth at any cost; they want to see spending translate into real performance.”
Rishi Jaluria, an analyst at RBC Capital, cautioned, “These companies seem to see no end in sight for capital expenditures,” prompting investors to rethink their timelines, and companies must strike a balance between investing in AI and not jeopardizing core businesses.
Furthermore, what further concerns the market is the actual on-the-ground effectiveness of AI in the corporate world. According to Bloomberg data, among companies discussing AI productivity, only 11% can quantify the benefits brought about by AI, and only 2% of enterprises can concretely quantify its “profit impact”.