As AI rally wobbles, stock market's spotlight turns to earnings

A wobbling AI rally means the stocks that powered much of the market have lost momentum. Semiconductor companies and hardware makers such as Dell Technologies fell sharply. That matters because AI-linked shares helped lead a four-year bull market and attracted enormous investment.

Earnings reports offer a practical test. Meta, Lam Research and Texas Instruments are among the companies reporting results. Investors want evidence that demand remains strong and that AI infrastructure spending is creating profits, rather than only higher costs and debt.

The market has not collapsed. The S&P 500 was about 1 per cent below its record this month, while other sectors began catching up. Third-quarter earnings for S&P 500 companies are expected to rise more than 30 per cent, but investors will closely examine whether AI-related growth can continue.

An AI hyperscaler is a company large enough to operate and finance enormous computing networks, including data centers built for AI workloads. Meta is one example identified in the article. These companies are central to the AI build-out because they provide the infrastructure behind AI assistants and related services.

Data centers contain powerful computing equipment and supporting systems. Building them requires huge capital outlays. Meta is among five AI hyperscalers expected to spend about $1.2 trillion collectively next year. The aim is to expand capacity as companies develop and offer more AI products.

This spending supports economic growth and profits across many businesses, including chip companies and hardware makers. However, investors want proof that returns will follow. The key question is whether hyperscaler profitability can grow faster than capital spending, especially as higher financing costs raise the required returns.

The AI investment boom is measured in extraordinary sums. Goldman Sachs expects five AI hyperscalers to increase combined capital spending to about $1.2 trillion next year. That spending covers data centers and related infrastructure needed to expand AI capacity.

The boom is also affecting company profits. LSEG IBES expects S&P 500 companies to report earnings growth of more than 30 per cent in the third quarter. The article links that profit jump partly to massive outlays on AI data centers and infrastructure. Chip companies have been among the biggest beneficiaries.

The figures show both the opportunity and the risk. Strong earnings can support the market, but investors still need to know whether the profits will keep pace with investment. Semiconductor shares remained up more than 60 per cent in 2026, despite falling 8 per cent during October.

Investors are testing whether AI spending creates enough profit to justify its scale. The central measure is the relationship between profitability and capital expenditure, or capex. If profits expand faster than capex, companies may be showing that their investments are earning attractive returns.

Meta provides an immediate test because it is a major hyperscaler and has recently launched its Muse AI assistant. The company is part of a group expected to spend about $1.2 trillion collectively next year. Its results and management commentary may indicate whether new AI services can support that spending.

The standard becomes more demanding when financing costs rise. Chuck Carlson said higher capital costs make the capex blowout more expensive and raise the bar for returns. Strong earnings would support confidence that the spending is sustainable; weaker profitability could intensify doubts about the AI trade.

CoreWeave's debt mattered because the AI build-out depends on enormous financing as well as technology. The company, an AI cloud provider, had about $50 billion in debt. That level made investors focus on the risks tied to funding data centers and other infrastructure.

When a heavily indebted company comes under pressure, investors may question its ability to keep financing expansion and meeting obligations. CoreWeave's shares tumbled during the week. The article presents this move as part of wider worries about how the AI boom is being funded, not simply about one company's operations.

Those concerns can spread through the market because AI spending has supported profits across many companies. Investors therefore want hyperscalers to show that earnings can grow faster than capex. If financing looks less secure or more expensive, confidence in the sustainability of the build-out can weaken, even while AI demand remains strong.

Treasury yields influence the financing environment for companies building AI infrastructure. When yields rise, borrowing generally becomes more expensive. Companies must then spend more to fund data centers, equipment and related expansion, which can reduce the profits left from each investment.

The article connects this pressure directly to AI capex. Chuck Carlson said that higher capital costs make the spending blowout more expensive and raise the bar for returns. This matters for Meta and other hyperscalers planning enormous investments. Investors want their profits to grow faster than those expenditures.

Inflation data increased bets on another Federal Reserve rate increase at the end of October. Futures pricing still showed most investors expecting the central bank to wait until December for a second increase this year. Moderately higher rates may be absorbed if earnings remain strong, UBS analysts said, but financing risks remain important.

A stock price reflects what investors believe a company can earn in the future, not only what it earns today. Expectations of stronger profits can therefore lift shares. Conversely, if future earnings look less valuable or less certain, investors may pay less for the stock.

Financing costs affect that calculation. A company borrowing to build AI data centers must earn enough from the investment to cover operating costs and interest. When Treasury yields and other borrowing costs rise, the project becomes more expensive. The company needs stronger profits for the same investment to make sense.

The article applies this logic to hyperscalers such as Meta. Investors are asking whether profitability can grow faster than capex. Possible Federal Reserve rate increases add pressure, while strong earnings could help equities withstand moderately higher rates, according to UBS Global Wealth Management analysts.