Wall Street ends lower on semiconductor selloff amid AI spending concerns

The semiconductor trade that powered markets for the better part of two years just hit a wall. On June 23, the Nasdaq Composite fell roughly 2.2% while the S&P 500 dropped 1.4% to approximately 7,364, as investors staged a broad retreat from chip stocks amid mounting anxiety over the sustainability of AI infrastructure spending.

The chip wreck

Nvidia, the poster child of the AI boom, saw its shares slide more than 4%. Intel, AMD, and Marvell Technology all suffered losses exceeding 5.8%, dragging the broader semiconductor sector into a significant decline.

The catalyst was a collective realization that hyperscalers, the Microsofts, Googles, and Amazons of the world, have been piling on debt to fund AI infrastructure at a pace that may not be matched by revenue on the other side.

Earlier in June, Broadcom’s Q3 AI revenue guidance came in at $16 billion, undershooting the $17.2 billion analysts had expected. That miss alone triggered a staggering $1.3 trillion wipeout in semiconductor market value in a single trading session. The June 23 selloff was, in many ways, the aftershock of that earthquake.

Asian memory chip stocks amplified the pain. Samsung and SK Hynix fell by as much as 12%, feeding back into US trading sentiment and creating a feedback loop of declining confidence across the global chip supply chain.

Why this time feels different

The Philadelphia Semiconductor Index had posted significant gains earlier in 2026, riding a wave of seemingly insatiable demand for AI-related hardware. But the June 23 decline carried a different flavor, representing a fundamental questioning of the return-on-investment math underpinning the entire AI infrastructure buildout.

Adding to the pressure: anticipation of a more hawkish stance from the Federal Reserve. If the Fed signals it intends to keep rates elevated or push them higher, the cost of servicing all that AI-related debt goes up.

What this means for investors

Analysts who track the semiconductor industry have characterized the current volatility as a potential market recalibration rather than a fundamental shift against AI investments.

For crypto-adjacent investors, semiconductor capacity constraints and pricing directly affect the cost of mining hardware and the economics of proof-of-work networks. A sustained downturn in chip demand from AI could, paradoxically, benefit crypto miners by easing competition for fabrication capacity and lowering hardware costs.

The semiconductor sector’s next major test will be earnings season. If Nvidia, AMD, and their peers can post results that show resilient demand despite the macro headwinds, the selloff may prove to be a healthy correction. If they can’t, the $1.3 trillion that evaporated in June might be just the beginning of a longer repricing.