AI Spending Surge Meets a Market Reckoning as Big Tech’s Cash Burn Draws Scrutiny

Big Tech’s AI buildout is still accelerating, but investors are no longer cheering the spending the way they once did. Alphabet, Amazon, Meta and Microsoft have laid out more than $740 billion in combined capital expenditure plans for this year, and the market is increasingly fixated on whether that money will generate enough revenue — and fast enough — to justify the cash burn.

That shift in mood was visible in a sharp selloff in tech shares, with the Nasdaq down 2.9% on Friday, Nvidia falling 4%, Broadcom dropping more than 5% and Qualcomm sliding 11%, according to TechCrunch. The broader message: the market is moving from applauding AI infrastructure spending to questioning who pays for it, how long the spending can continue, and when the returns arrive.

Why investors are getting uneasy

The new worry is not whether demand for cloud and AI infrastructure exists. It does. The concern is that the cost of supplying that demand is rising faster than the cash being produced, even by the largest hyperscalers. TechCrunch reported that investors have started to focus more on negative free cash flow, heavier debt loads and uncertain payback periods.

That matters because the current phase of the AI race is capital intensive by design. Data centers, chips, power, networking gear and the land to house them are expensive, and those costs arrive long before product revenue does. Free cash flow — the money left after operating costs and capital spending — is the clearest line in the sand for whether a company can fund that buildout from its own business. When that figure turns negative, as it has for several of the biggest players, the market starts asking harder questions.

How the mood changed

For much of the AI boom, markets rewarded companies that spent aggressively because spending signaled ambition, scale and competitive strength. That playbook is now under pressure. TechCrunch said the change in sentiment is showing up in stock prices and in the way investors parse earnings calls: growth in AI infrastructure still matters, but profitability and cash generation matter more than before.

The concern is not abstract. S&P Global Market Intelligence said Microsoft, Alphabet, Amazon, Meta and Oracle raised $326 billion in debt in 2026, about triple the amount raised last year, according to TechCrunch. That figure underlines how much external financing is becoming part of the AI buildout story, even for companies with giant balance sheets and strong core businesses.

What the big spenders are actually saying

Alphabet, Amazon, Meta and Microsoft have each signaled huge spending plans, but the details differ sharply company by company. The table below summarizes the key numbers reported by TechCrunch.

Alphabet is the clearest example of how quickly the market can turn. TechCrunch reported the company’s shares fell 12% on Thursday, the worst day in its history, wiping out about $500 billion in market value. The company also reported Q3 free cash flow of negative $9.1 billion, its second straight negative quarter, alongside Q3 capex of $51.4 billion.

Alphabet’s spending outlook is still climbing. TechCrunch said the company sees capex reaching up to $205 billion this year and even more in 2027. At the same time, Google Search ad growth slowed to 12%, the weakest pace in five quarters, even as AI Mode gains traction. That combination — slower growth in the core ad business and heavier infrastructure spending — helps explain why investors reacted so sharply.

Amazon’s numbers show a different strain. The company’s trailing-12-month free cash flow came in at negative $42.4 billion, compared with positive $14.8 billion a year earlier, according to TechCrunch. Yet AWS revenue still grew 39%, showing that demand for cloud compute remains strong even as the cost of meeting it weighs on cash generation. Amazon expects $220 billion of capex this year.

Meta, meanwhile, reported Q3 free cash flow of negative $2.1 billion and is guiding to as much as $145 billion in capex this year, TechCrunch said. Microsoft has so far been the relative outlier: its latest-quarter free cash flow was $18.6 billion and capex came in at $45.2 billion, below a $50 billion projection. Even there, however, the company is still spending at a scale that would have looked extraordinary only a few years ago.

Why it matters

This matters because the AI boom is no longer just a story about product launches or model performance. It is also a financing story, and financing stories tend to matter most when markets start asking who shoulders the risk if revenue growth slows. The current AI buildout depends on a small number of giant companies committing extraordinary sums to chips, data centers and networking. If the payback period stretches out, the pressure can spill from stock prices into borrowing costs, capital allocation and hiring plans.

For readers, the implications go beyond Silicon Valley. A sustained shift in market sentiment can affect the cost of capital for companies across the AI supply chain, from chipmakers to cloud vendors and infrastructure providers. It can also shape how quickly new AI services reach consumers and businesses. If the biggest platforms keep spending, AI capacity expands. If investors push back harder, companies may have to slow, sequence or finance that expansion differently.

The market reaction also matters because it changes the scoreboard. For years, investors often treated heavy AI investment as proof that a company was serious about the future. Now they are asking whether the spending is self-funding. That is a harder test, especially for businesses still trying to show that AI can become a durable profit engine rather than just a cost center.

Background: how we got here

The present debate began with a simple premise: the companies that control the biggest cloud platforms would need to build out enough capacity to train and run AI models at scale. That meant spending heavily on data centers, chips and networking. Hyperscalers — the largest cloud providers — became the financial backbone of the AI race because they could pool demand from many customers while building the infrastructure needed for model training and inference.

At first, markets largely welcomed that spending. The logic was straightforward: if cloud providers invested early and aggressively, they would lock in demand, win developers and serve as the default infrastructure layer for generative AI. But as the amounts grew, so did the questions. Capital expenditure became so large that it stopped looking like ordinary growth investment and started looking like a test of balance-sheet endurance.

TechCrunch framed the current moment as a transition from enthusiasm to skepticism. The combined capex plans from Alphabet, Amazon, Meta and Microsoft — more than $740 billion this year — show how concentrated the AI infrastructure race has become. But the same figures also make it clear why investors are uncomfortable: the industry is asking for huge upfront spending before the market has seen equivalent revenue payback.

Google’s slowdown is an early warning sign

Alphabet’s results are particularly important because they show that even a dominant business can feel the strain of the AI cycle. Google Search advertising growth slowed to 12%, the weakest pace in five quarters, while AI Mode gains traction. That does not mean the search business is weakening outright, but it does suggest that the core revenue engine is not accelerating fast enough to offset the infrastructure bill without pressure.

The market’s response was severe because Alphabet’s cash profile changed so quickly. Negative free cash flow in two straight quarters, a huge capex number and an expanding investment outlook changed the way the company is being priced. Investors are no longer simply valuing the company on growth potential; they are weighing the timing of returns.

Amazon shows scale, but also strain

Amazon’s case is different because AWS continues to grow briskly. A 39% increase in AWS revenue is not a sign of weakness. But a trailing-12-month free cash flow figure of negative $42.4 billion shows how fast the infrastructure bill can overwhelm operating strength when investment is heavy. That gap between top-line demand and bottom-line cash generation is the central tension in the current market.

In other words, the demand side of the AI story is intact. The financing side is what is being re-priced.

How are markets reacting right now?

The first answer is that they are selling AI-linked stocks more aggressively than before. Friday’s 2.9% drop in the Nasdaq, together with declines of 4% in Nvidia, more than 5% in Broadcom and 11% in Qualcomm, points to broad anxiety around the chip and infrastructure trade, according to TechCrunch. Alphabet’s 12% drop on Thursday was the clearest single-company warning sign.

The second answer is that investors are not rejecting AI. They are rejecting open-ended funding assumptions. That distinction matters. Cloud demand remains strong across Google, Amazon and Microsoft, TechCrunch reported, but investors are more worried about the cost to supply it. The market is not saying AI is a dead end. It is saying the economics need to improve.

What CNBC-TV18’s Anand Shah said about the broader setup

CNBC-TV18’s interview with Anand Shah of ICICI Prudential AMC adds a useful contrast to the U.S. mega-cap picture. Shah argued that the AI cycle is being driven by investment rather than a broad revenue boom. He said,

“Today, the boom is in AI investment. We are not seeing a big boom in terms of revenue”

That view lines up with the market’s current skepticism. It also helps explain why asset managers are being cautious even as they acknowledge the strategic importance of AI. CNBCTV18 reported that ICICI Prudential AMC has started adding to IT stocks after a long underweight stance, while remaining overweight manufacturing across market caps. The firm also sounded constructive on Indian telecom, pointing to low ARPU, consolidation and tariff upside.

Shah’s remarks on India’s broader market setup also matter because they highlight a different kind of dependence: capital flows. He said,

“We need to be mindful that, unlike many other economies that run surpluses, exporting more than they consume or producing more than they consume, India is much more dependent on foreign flows”

That comment underscores why global risk sentiment can have ripple effects well beyond the U.S. tech sector.

What the debt data says about the next phase

One of the most telling details in TechCrunch’s reporting is the jump in debt issuance. S&P Global Market Intelligence said Microsoft, Alphabet, Amazon, Meta and Oracle raised $326 billion in debt in 2026, about triple last year. That does not mean these companies are in financial distress. It does mean the AI buildout is increasingly being financed through a mix of operating cash flow and borrowing.

That shift changes the risk profile in three ways:

• It raises the bar for revenue growth needed to justify spending.
• It makes markets more sensitive to any slowdown in cloud demand or ad growth.
• It increases the importance of interest rates, credit conditions and investor appetite for large issuers.

The debt picture also makes the AI story less like a short burst of exuberance and more like a multi-year infrastructure cycle. Once companies begin funding a major buildout with debt at scale, the market starts looking for operating leverage, not just expansion.

Why investors are watching cash flow so closely

Free cash flow is the cleanest measure of whether a company can fund growth without constantly leaning on the market. For AI hyperscalers, that metric has become a battleground because capital expenditure can temporarily overwhelm even healthy earnings. Negative free cash flow does not automatically mean weakness. It can reflect a deliberate choice to invest ahead of demand. But when the investment phase lasts long enough, investors start to wonder whether the future payoff is arriving at the pace they were promised.

That is why the market has become less forgiving. Alphabet’s negative free cash flow, Amazon’s steep swing from positive to negative, and Meta’s own cash burn all suggest that the AI arms race is still very much in the spending phase. Microsoft’s relatively stronger cash generation only sharpens the comparison. It shows there is still room for discipline inside the sector — and that investors will notice when one firm seems to manage the balance better than others.

What happens next

The next phase of this story will depend on three questions. First, whether AI infrastructure spending starts producing enough revenue to improve cash flow at scale. Second, whether the biggest companies keep pushing capex even higher or begin to moderate spending as investors get more vocal. Third, whether the recent selloff in AI-related stocks continues or stabilizes once markets absorb the latest results and financing data.

TechCrunch pointed to open questions around whether Alphabet, Amazon, Meta and Microsoft will raise capex or debt further if demand stays strong. It also flagged whether Anthropic’s IPO or financing plans could be affected by higher infrastructure-funding costs. Those questions matter because the financing environment around AI is no longer confined to the largest public companies. It is starting to shape the broader ecosystem around them.

For now, the market is drawing a clearer line than it did earlier in the AI cycle. Spending alone is no longer enough to impress. Investors want evidence that the infrastructure buildout can translate into durable revenue, healthy cash generation and a more predictable return on capital. Until then, the sector is likely to remain caught between the promise of AI demand and the reality of how expensive it is to supply.

By the numbers

The core question now is not whether AI spending will continue. It almost certainly will. The question is whether the companies funding it can prove that the returns will be large enough — and arrive soon enough — to satisfy investors who are increasingly less willing to wait.