June 24 was one of the most volatile days in tech markets this year. Semiconductor and AI stocks led a sharp selloff that erased hundreds of billions in market value, as a wave of regulatory, financial, and geopolitical news converged on the sector.
The Nasdaq Composite fell more than 2%, extending a two-day losing streak. The Philadelphia Semiconductor Index (SOX) dropped roughly 4.8%, its steepest single-day decline in months. Nvidia closed near $200, down 4.1% on the day. AMD and Intel posted losses of 7% or more. Broadcom slipped about 3%, and Micron, slated to report earnings after the close, came under heavy pressure as traders braced for a read on AI-driven memory demand.
Why This Matters
This is not just a technical pullback. It reflects a structural shift in how investors are pricing the AI trade. After more than two years of near-unconditional enthusiasm for AI-related capital expenditure, the market is now demanding evidence of returns. Reports from Reuters and Bloomberg described a growing sense of "AI fatigue" among institutional investors, who are increasingly skeptical of hyperscalers' debt-funded spending sprees.
The conversation among AI insiders is no less dramatic. Two weeks ago, Anthropic launched Claude Fable 5 and Claude Mythos 5, its first "Mythos-class" models, positioned above the Opus tier and marketed as the company's most capable publicly available systems. Within days, the U.S. Commerce Department issued an export-control directive ordering Anthropic to suspend global access to both models, citing national security risks tied to their advanced coding and vulnerability-discovery capabilities. Anthropic complied, disabling the models for all users, though it disputed parts of the government's assessment.
The Bigger Picture
OpenAI filed a confidential S-1 with the SEC on June 8, with Goldman Sachs and Morgan Stanley underwriters. The company's most recent funding round, closed in March, valued it at $852 billion. An IPO could come as early as fall 2026 or slip into 2027, depending on market conditions, regulatory review, and internal readiness. The filing joins a historic wave of tech listings: SpaceX went public this week, and Anthropic itself submitted a confidential S-1 in early June.
Meanwhile, the Wall Street Journal published a deep investigation into Sam Altman's personal investments, revealing that much of his wealth is tied to a network of companies benefiting from their connections to OpenAI. The House Oversight Committee opened an investigation into his portfolio in May, which reportedly exceeds $2 billion in overlapping stakes. Altman's stake in nuclear fusion startup Helion Energy alone, which he allegedly proposed OpenAI commit approximately $500 million to, more than doubled to at least $4.1 billion after a recent funding round.
Andrej Karpathy, the former OpenAI co-founder who joined Anthropic's pre-training team in May, described the pace of change in the industry as "extremely large and disruptive." In a recent interview with Fortune, he said he hasn't typed a line of code manually since December 2025, nearly all of his work is now done through AI agents. He called it a "state of psychosis" exploring new possibilities.
The Bottom Line
The market is sending a clear signal: the era of blind AI spending is ending. Valuations for names like Nvidia will now be measured not just by adoption rates, but by demonstrated return on the enormous capital being raised. Regulators in Washington and Brussels are tightening oversight, both on the models themselves and on how AI companies operate. The message for anyone watching from the sidelines: Big Tech's grass is still green, but the pruning has begun.
