U.S. stock markets experienced a severe selloff on November 5, 2026, with the Dow Jones Industrial Average falling 1,833 points (-3.44%), the S&P 500 declining 334.89 points (-4.36%), and the Nasdaq Composite plunging 1,421.52 points (-5.52%) in its worst session since April 2025 [1]. The Nasdaq also entered correction territory, closing at least 10% below its October 6 record high [1]. The selloff was led by chipmakers and AI infrastructure stocks, with specific declines including Micron down 17%, SanDisk down 22%, and GE Vernova down 7% [2][3]. The iShares Semiconductor ETF (SOXX) fell 3% on Wednesday and another 10% on Thursday [3].
Wall Street's concerns extend beyond a single trading day and center on three interconnected issues. First, investors are questioning the massive debt obligations being issued by tech companies to build data centers, particularly given rising interest rates and political uncertainty [1]. Second, there are doubts about return on investment, as investors question how long hyperscalers can sustain current spending levels while free cash flow turns negative [1]. Third, semiconductors represent the most crowded trade in the market, and crowded trades "unwind violently" according to market analysts [1]. Additionally, power capacity has emerged as "the real bottleneck in AI today, more than chips," raising concerns about grid capacity, utility costs, and facility deployment timelines [1].
The immediate catalyst for the selloff appears to be a combination of midterm election results and disappointing tech earnings [1][2]. Political uncertainty surrounding data center construction—including regulatory concerns, slower permitting, and opposition to power bills—has raised fears about the pace of infrastructure buildout [1][2]. Jim Cramer warned that the AI stock selloff likely has further to run, though he distinguished this downturn from the dot-com crash by noting that today's hardest-hit companies have profitable businesses [2]. Cramer specifically highlighted concerns about smaller developers such as CoreWeave, Iren, and Nebius, which borrowed money to finance projects that could face regulatory delays [2].
Valuation uncertainty has become a critical issue across the sector. Cramer noted that AI stocks are "increasingly difficult to value" given data center build-out uncertainty [3]. He pointed to Nvidia trading at roughly 12 times next year's earnings with an expected 70% growth rate—appearing cheap only if demand remains strong [3]. Similarly, AMD's stock price rose from $188 in March 2026 to $404 currently, raising questions about whether valuations reflect realistic growth scenarios [3]. Cramer urged investors to maintain cash reserves and avoid panic, noting that falling interest rates and signs of easing inflation could provide support, though he acknowledged that semiconductor stocks could face further declines [2].
CONCLUSION
The November 5, 2026 AI stock selloff reflects a fundamental reassessment of data center economics and regulatory feasibility rather than a collapse of underlying business fundamentals. While analysts expect further downside in the near term, the market is distinguishing between profitable AI companies and speculative ventures, with particular vulnerability concentrated in smaller developers dependent on project financing. The key determinant of market recovery will be clarity on whether regulatory hurdles merely delay data center construction or significantly constrain long-term growth assumptions.