AI capital expenditure report card is coming: Microsoft, Amazon, cloud computing, Meta, advertising

AI capital expenditure report card is coming: Microsoft, Amazon, cloud computing, Meta, advertising.

Four technology giants with a market capitalization of trillions will disclose quarterly financial reports one after another. Among them, and Meta are the core drivers of U.S. AI capital expenditures.

The progress of each company's huge investment and its ability to alleviate the market's concerns about cash flow will become the biggest focus of this financial reporting season.

Free cash flow contraction ≠ cash flow depletion Industry data shows that the total capital expenditure of the world's five largest cloud service providers is expected to exceed US$725 billion in 2026, a significant increase from last year's US$412 billion.

However, under accounting rules, only about US$211 billion of capital expenditures for the whole year will be included in the income statement for depreciation for the year, and the remaining more than US$500 billion will be included in the balance sheet and will be amortized year by year as long-term assets in the future.

This time difference has created a special market pattern in which the profit growth rate of technology giants exceeds 20% and large amounts of cash continue to flow out. It also triggers market caution, and funds continue to sell stocks of companies with high capital expenditures.

However, some analysts pointed out that betting on the long-term deterioration of corporate free cash flow is essentially betting on the loss of the investment ability of the best capital operation giants in the past two decades.

In fact, negative free cash flow only means that the company is in the stage of large-scale investment, and the success or failure of the investment cannot be directly determined. The final result needs to be verified based on subsequent revenue data.

The industry pointed out that this round of AI computing power investment is fundamentally different from traditional infrastructure. The iteration speed of computing hardware such as GPUs is much faster than that of traditional infrastructure such as railways.

If the actual effective commercial life of the GPU is lower than the company's management expects, depreciation pressure will be released early, and the recovery window period in which the asset can be used to generate revenue will be compressed.

This is the core issue in this wave of AI capital expenditures, and each company needs to be evaluated individually and cannot be generalized. Four companies, four transcripts The company will report earnings on Wednesday, with market consensus forecasting earnings of $4.22 per share.

The company's full-year capital expenditure plan for 2026 will reach US$190 billion, and investment pressure has gradually emerged. Capital expenditures in the last quarter were US$31.9 billion, and free cash flow fell back to US$15.8 billion, a sharp decline from the US$25.7 billion two quarters ago.

The core support of its huge investment is the Azure cloud business, and management guided that the revenue growth rate of this business should be maintained at 39%-40%. Whether the growth rate of cloud business can be maintained will directly determine the rationality of large AI investment.

Once the growth rate is lower than expected, the cash flow balance of enterprises will quickly come under pressure. It will announce earnings on Thursday, with market expectations of earnings per share of $1.82.

Its capital expenditure logic is consistent with Microsoft but larger, with a capital expenditure plan of nearly US$200 billion, resulting in the company's free cash flow of only US$1.2 billion in the past twelve months.

However, there are clear positive signals for the company: the growth rate of AWS cloud business rebounded to 28% in the first quarter, setting a new high in more than three years, and the scale of orders to be executed exceeds 360 billion US dollars.

Whether the growth rate of cloud business can be sustained is the key to verifying the value of its infrastructure expansion. When the previous capital expansion plan was implemented, the company's stock price fell 8% in a single day, which showed the market's cautious attitude.

It will release earnings on Wednesday, with market expectations for earnings of $7.23 per share. The company's 2026 capital expenditure guidance has been raised to $125 billion to $145 billion. This also makes Meta the purest AI investment stress test target among the four companies.

As investment continues to increase, the company's operating profit margin has fallen from its high point. If capital expenditures continue to rise, free cash flow is likely to turn negative. It will release earnings on Thursday, with consensus expectations of earnings per share of $1.89.

It is almost outside the main line of the story of this AI capital expenditure. Its capital expenditure scale is only a fraction of its own operating cash income generation scale, and the company can still generate huge free cash flow every quarter.

Apple's real risks lie elsewhere: the lengthening iPhone upgrade cycle, slowing service business growth, intensifying competition in the Chinese market, and the outside world's perception that it "has fallen behind in AI." In addition, this will also be Tim Cook's last financial report meeting as CEO, and he will be succeeded by John Ternus, which may also add variables other than financial report data to the market.

The industry pointed out that the revenue growth rate used by the five major cloud vendors to justify capital expenditures is rising, not slowing down. Google Cloud revenue growth reached 82%, Azure was close to 40%, and AWS accelerated again to 28%.

The contract backlog behind them is staggering: Google Cloud reaches US$514 billion, Microsoft exceeds US$600 billion, and Amazon exceeds US$360 billion. These data are not a reflection of corporate business decline, but a signal that corporate capacity expansion cannot keep up with market demand.

Note: The horizontal axis is the year-on-year growth rate of cloud business revenue of the three giants in the latest quarter; the vertical axis is the scale of pending orders for cloud business This month, the market seems to have doubts about the prospects of this round of AI expansion.

Alphabet's revenue exceeded expectations, and Google Cloud's second-quarter revenue soared 82% year-on-year. The stock price still fell about 5% after the market closed that day, and the trigger was only the company's further increase in capital expenditures.

According to data from FactSet, a global financial data service provider, second-quarter earnings of S&P 500 constituent stocks increased by approximately 24.7% year-on-year, marking the second consecutive quarter of growth exceeding 20%.

Continuously rising performance continues to raise the threshold of market expectations, and investors' tolerance for "substandard" financial reports continues to decrease.

According to data from Bespoke Investment Group, a US stock market research institution, Among companies that have announced earnings this season, those whose performance fell short of expectations fell an average of 4.2%, while the historical average was about 2.9%.

The market's callback reaction to performance that fell short of expectations was significantly larger than usual.

#Stocks #Nvidia #Apple #Microsoft #Meta