U.S. stocks sold off sharply on Oct. 22 as investors reassessed the expense, financing demands and eventual returns of the technology industry’s AI buildout. The Nasdaq Composite fell 2.6% to 26,008.86, its steepest decline since June, while the S&P 500 lost 1.2% to 7,663.43. The Dow Jones Industrial Average was effectively unchanged, rising 11.89 points to 52,867.98.
The split among the major averages captured the concentration of the move. Nvidia dropped 4.4%, Meta Platforms fell 7%, Tesla declined 3.1% and CoreWeave lost 10.1%, according to an Associated Press market report. Yet two of every three S&P 500 constituents rose. A relatively narrow group of heavily weighted technology shares was enough to pull the capitalization-weighted index lower.
A concentrated selloff masked broad gains
The market action put renewed attention on a basic feature of the AI trade: the companies most exposed to data centers, advanced chips and computing capacity also carry outsized influence in major equity benchmarks. Their declines overwhelmed gains across most of the index, producing a materially weaker S&P 500 despite positive breadth.
That is not evidence that every AI-related business faces the same commercial outlook. Semiconductor designers, cloud-computing providers, data-center operators and platform companies have different revenue models and spending profiles. But their shares have increasingly moved together when investors question the scale or timing of returns from infrastructure investment. A Reuters report republished by The Standard described the session as a revival of worries over the AI spending race and its financing needs.
The closing data also require a small distinction. Reuters’ initial figures were labeled preliminary and put the S&P 500 and Dow marginally higher than later end-of-day reports. The final figures reported by AP and Yahoo Finance place the S&P 500 at 7,663.43 and the Dow at 52,867.98. All of the accounts agree on the larger picture: a roughly 2.6% Nasdaq drop, a 1.2% S&P 500 decline and a flat Dow.
Meta disclosure highlights the financing question
Meta became a focal point after projecting capital expenditures of $135 billion to $145 billion, compared with prior guidance that had a $130 billion lower bound, according to Yahoo Finance. The company’s shares fell 7% as investors weighed a higher spending plan against the timetable for converting AI investment into revenue and cash flow.
Separately, Meta has set out a corporate mechanism that could expand its financial flexibility without changing its voting structure. In an information statement filed with the Securities and Exchange Commission, the company said its board approved the creation of non-voting Class C common stock. Subject to the amended charter taking effect Jan. 19, 2027, holders of Class A and Class B shares would receive one Class C share for each share held.
The filing does not announce an immediate stock sale. It says Class C shares could be used in future capital raising, employee compensation, acquisitions and other corporate purposes, while allowing Meta to issue stock without diluting the voting power attached to its existing voting classes. Shareholders affiliated with Chief Executive Mark Zuckerberg approved the recapitalization and related equity-plan amendment by written consent on Oct. 20, the filing says.
Those corporate actions should not be read as a proven cause of the one-day selloff. The SEC document establishes the structure and potential uses of the shares, not the market’s reasons for trading Meta lower. The immediate market narrative, as described in news reports and strategist commentary, centered on the broader cost of AI infrastructure and the financing burden across the industry.
Bonds rally as selling spreads across markets
The equity retreat was accompanied by a pronounced move into U.S. government debt. The benchmark 10-year Treasury yield fell to 4.97% from 5.13% late on Oct. 21, a 16-basis-point move in one session, AP reported. Bond yields fall as prices rise, so the move signaled demand for Treasuries during the equity retreat rather than a broad increase in risk appetite.
By the following day, the yield had fallen more than 19 basis points overnight and was trading near 4.9591%, according to a Reuters follow-up carried by The Straits Times. The report said the decline was the largest daily move in the 10-year yield since March 2023. Nomura Securities International strategist Charlie McElligott characterized the rally as a repricing of the risk that an AI-market reset could become a broader U.S. growth concern.
Oil moved in the opposite direction. Brent crude rose 4% to $113.17 a barrel after reaching a five-month high, AP reported, adding an inflation-sensitive cross-current to the drop in stocks and yields. The next trading session brought further weakness in several Asian markets, including Japan’s Nikkei, South Korea’s Kospi and mainland Chinese blue chips. The combination left investors confronting two separate pressures: sharply higher capital requirements for AI infrastructure and an energy market that was moving higher at the same time.
