Investors are nervous about spending on AI infrastructure as big tech companies double down.
Over the past two weeks, Alphabet, Amazon, Meta and Microsoft have outlined plans to spend more than $740 billion in capital spending this year, with Wall Street seeing this as spending on AI infrastructure, including data centers and chips.
In the early days of the AI boom, investors saw such large spending as a sign of confidence in the technology’s growth potential and rewarded it primarily. But in the fourth quarter, that enthusiasm appears to be giving way to negative free cash flow (a company spending more cash than it brings in), rising debt, and concerns about whether these large investments will generate enough returns.
Uncertainty spread across the market on Friday. The Nasdaq fell 2.9%, and semiconductor stocks also fell sharply. Nvidia fell 4%, Broadcom more than 5% and Qualcomm 11%. Alphabet, which had its worst day ever on Thursday, plunged 12% and wiped out $500 billion in market capitalization, is back down again.
A big part of the concern is that AI threatens the economics of these businesses, hurting companies that traditionally had free cash flow from their core businesses. This week, Alphabet provided the clearest example of that. Google’s parent company reported negative free cash flow of $9.1 billion in the third quarter, its second consecutive quarter of a loss, even as capital spending increased to $51.4 billion in the quarter. Alphabet expects to spend up to $205 billion this year and expects spending to continue rising in 2027.
Meanwhile, the huge success of AI Mode, Google’s AI-powered chatbot feature built directly into search, is eating into the company’s revenue stream. Growth in Google’s search advertising business, which accounts for about half of its revenue, slowed to 12%, the slowest pace in five quarters. This leaves Google walking a tightrope. The company is spending billions of dollars on AI technology, which could undermine the search business that pays for the technology.
Amazon is in a similar situation, reporting negative free cash flow of $42.4 billion in the past 12 months, compared to positive free cash flow of $14.8 billion a year ago. The company said it expects to spend $220 billion in capital spending this year. Amazon’s stock price fell 7% after Thursday’s earnings report, even though revenue at its cloud business, AWS, rose 39%.
Meta was also burning cash, reporting negative free cash flow of $2.1 billion in the third quarter as capital expenditures exceeded operating cash. Capital investment this year is expected to reach up to $145 billion. The company noted that its new free personal AI agent, Muse, is a bit of a cash cow (CEO Mark Zuckerberg said it’s “very computationally intensive to run”). However, advertising revenue increased by 26%. Still, no matter how good the news was, it was inevitable that Meta would take a 10% hit to its stock price last week.
Microsoft was the only exception when it came to free cash flow. The company had positive free cash flow of $18.6 billion in its most recent quarter, but spent $45.2 billion on capital expenditures, less than the $50 billion it had expected. Microsoft said this was not a deliberate setback in AI spending, but instead was due to delays in processing power from third-party providers. As a result, the stock prices of small and medium-sized companies that lend out AI computing power have fallen.
It’s worth noting that the money to pay for all this AI infrastructure isn’t coming solely from the profits of big tech companies. This is also reflected in the sale of stocks and bonds. Microsoft, Alphabet, Amazon, Meta and Oracle have collectively raised $326 billion in debt in 2026, triple the amount they raised last year, according to S&P Global Market Intelligence.
Demand for AI computing remains strong. Google, Amazon, and Microsoft all made money from their cloud businesses. But investors are spooked by the cost of meeting that demand.
These murky waters could complicate Anthropic’s long-awaited IPO. The AI company filed to go public earlier this week, and its future growth will depend on securing massive amounts of compute. The Financial Times reported Friday that Blackstone warned investors that the company may not qualify for an investment-grade rating after its IPO. That means the company could become more reliant on hyperscalers and third parties to fund its infrastructure ambitions.
If you buy through links in our articles, we may earn a small commission. This does not affect editorial independence.
