Wall Street awaits strong earnings to sustain AI rally after recent selloff

Wall Street heads into earnings season facing a critical test: can major tech companies prove their enormous AI infrastructure spending is translating into profits, or has the market overestimated the near-term payoff of the artificial intelligence boom? This season is expected to mark the second consecutive quarter of 20%-plus earnings growth for the S&P 500, with the AI trade driving much of that momentum. Yet over the past week, the sector has taken significant hits. Memory chip highfliers Micron Technology and SanDisk have dropped more than 10% and 25% respectively over five sessions, with semiconductor stocks falling further into bear-market territory after their parabolic rally earlier this year. Dan Ives, Yorkville Ives partner, noted that memory stocks have been "the golden child of this AI revolution" over recent months and predicted continued rotation as the AI story evolves. "It comes down to earnings season," Ives told Yahoo Finance. "You have to show the monetization piece, because right now it's just capex." The scale of that capex is staggering. Hyperscalers are set to spend a collective $650 billion（約100兆円） on AI infrastructure this year, according to data compiled by Bloomberg and Janus Henderson Investors. As earnings reports begin, investors are hunting for evidence of return on investment. Alphabet will be the first of the Big Tech players to report, on Wednesday. Seema Shah, Principal Asset Management chief global strategist, underscored the stakes: "Their earnings need to be strong. They are really the foundation for the entire AI ecosystem." Recent corporate moves have sent mixed signals. Taiwan Semiconductor Manufacturing, the world's largest contract chipmaker, raised its capital spending plans to expand capacity, partly due to rising equipment costs—a move that spooked the market. However, TSM also hiked its revenue guidance, citing strong demand. Equipment supplier ASML issued a similarly bullish forecast as chipmakers continue purchasing advanced equipment. Sylvia Jablonski, Defiance ETFs chief investment officer, interpreted ASML's strength as "a real vote of confidence for the AI trade" and noted that AI represents a multitrillion-dollar addressable market. Not all technology has benefited equally from the boom. IBM shares suffered their steepest decline in decades after the company warned that enterprise clients are prioritizing spending on memory and AI infrastructure over mainframe computers and traditional software. Conversely, cybersecurity stocks received a boost after IBM noted that clients are increasingly focused on security concerns amid surging use of agentic AI. The verdict for the sector will hinge on whether companies can demonstrate that the $650 billion（約100兆円） being poured into AI infrastructure generates proportional business value—proof that has, so far, proven elusive enough to rattle investors.

The AI trade entered earnings season under pressure after a sharp selloff in semiconductor stocks—the very companies that powered the sector's gains earlier this year. Memory chip makers like Micron and SanDisk, described as "the golden child of this AI revolution" by analysts, have tumbled into bear-market territory, signaling investor concern that enthusiasm has outpaced fundamentals. The market's anxiety centers on a fundamental mismatch: hyperscalers have committed to spend $650 billion（約100兆円） on AI infrastructure this year, a massive capital commitment driven partly by competition and rising equipment costs (as evidenced by Taiwan Semiconductor's capex hike), yet investors lack clear evidence that these investments will generate proportional returns. This earnings season will test whether that narrative changes. Alphabet's Wednesday report arrives as the opening bell for Big Tech's earnings, carrying outsized significance because these hyperscalers form the foundation of the entire AI ecosystem. If their results demonstrate strong monetization of AI spending rather than just capex growth, it may restore confidence and halt the recent rotation out of the sector. If earnings disappoint or merely show rising costs without commensurate revenue gains, the market may view this as confirmation that the AI capex cycle has run ahead of real business returns.