The Morning in Review

The morning session delivered exactly what pre-market futures warned. Nasdaq Composite futures had pointed to losses of as much as 2.6% ahead of Tuesday's open, and the index tracked that projection closely, falling 2.4% as selling pressure in technology and semiconductor names showed no sign of abating. The S&P 500 and Dow Jones Industrial Average also declined, though losses remained concentrated in the tech-heavy corners of the market.

The session's defining feature was not a single catalyst but a convergence of pressures that had been building since Monday. Investors are demanding evidence that the enormous capital flowing into AI infrastructure will generate returns on a timeline that justifies current valuations. So far, no such evidence has arrived — and the market is repricing accordingly.

The phrase analysts have reached for is "rotation not liquidation" — a framing that suggests institutional money is shifting out of high-multiple tech names into other sectors, rather than exiting equities wholesale. The scale of the two-day move complicates that reassurance. The Nasdaq 100 is on track to shed over $1 trillion in market capitalization across the episode.

Where the Pain Is Concentrated

Memory-chip stocks absorbed the sharpest losses. Micron Technology (MU) and Western Digital (WDC) each slid approximately 10%, while SanDisk (SNDK) fell 11%. The moves tracked a near-10% crash in South Korea's Kospi index, whose heavy weighting in memory-chip producers — including Samsung and SK Hynix — makes it a sensitive real-time gauge of global semiconductor demand.

The Micron story carries an additional layer of complexity. The company had reported exceptional earnings driven by surging demand for high-bandwidth memory used in AI accelerators, and its stock had risen nearly ninefold over the prior year heading into that report. A contrarian reading is now gaining traction: in the memory-chip industry, peak earnings have historically preceded sharp cyclical downturns as supply catches up with a demand surge. Tuesday's 11% decline suggests some investors are already positioning for that turn.

Nvidia (NVDA) also featured among the declining names, reinforcing that this is a broad repricing of AI-linked hardware rather than a company-specific event.

Alphabet (GOOGL) is carrying a heavier burden than most. The stock has fallen 6% across two sessions, with the latest leg down tied to a high-profile talent drain: prominent AI researcher John Jumper departed for Anthropic, joining a pattern of exits to Anthropic and OpenAI that is drawing scrutiny from investors. In AI development, small teams of elite researchers can determine whether a model leads or lags the field. Alphabet's DeepMind and Google Brain divisions have long been considered premier talent pools, and losing senior figures to well-funded competitors introduces a company-specific headwind that sits on top of the sector-wide pressure.

Tesla (TSLA) added a separate negative catalyst to an already difficult tape. A fatal crash triggered a federal investigation into the company's autonomous driving systems, renewing questions about the timeline and liability framework for full autonomy. Tesla's long-term valuation argument rests heavily on its self-driving ambitions, and federal scrutiny of that technology carries weight with investors regardless of the broader market environment.

A Countercurrent Worth Watching

Not every chip-sector move this session points in the same direction. Qualcomm (QCOM) is reportedly in advanced discussions to acquire Modular, an AI infrastructure company, in a deal valued at approximately $4 billion. The timing is striking: Qualcomm is moving toward a significant AI acquisition precisely as the sector is selling off, suggesting its management sees the dislocation as a strategic opening rather than a warning sign.

Modular's technology focuses on running AI models efficiently across different hardware — a capability that could help Qualcomm offer enterprise customers an alternative to Nvidia's tightly integrated software-hardware ecosystem. If completed, the deal would be one of the larger AI-focused acquisitions of the current cycle and would mark a meaningful expansion of Qualcomm's ambitions beyond its traditional mobile-chip franchise.

The Afternoon Setup

The central question for the afternoon session is whether the rotation thesis holds. If institutional selling is repositioning rather than risk-off, there should be visible buying in sectors that benefit from a shift away from high-multiple tech — financials, energy, industrials. If those flows are absent, the liquidation interpretation becomes harder to dismiss.

Federal Reserve policy is the macro variable that could determine the session's direction. Rate-cut expectations have been reset, and higher-for-longer rates weigh disproportionately on growth stocks whose valuations depend on discounting future earnings at low rates. Any Fed commentary this afternoon that signals further hawkishness could extend the selloff into a third session. Conversely, any language that softens the rate path could provide the technical bounce that oversold chip names might otherwise attract.

The Korean market's trajectory overnight will set the tone for Wednesday's open in U.S. semiconductors. A stabilization in the Kospi would offer some relief to MU, WDC, and SNDK; a continued decline would likely extend the pressure.

For Alphabet specifically, the talent story is unlikely to resolve quickly. Investors will be watching whether the company makes any public response to the researcher departures and whether additional exits are reported. The federal probe into Tesla's self-driving systems will also develop on its own timeline — the scope and focus of the investigation are still emerging, and any indication that it targets a specific software version or triggers a mandatory recall would carry meaningful financial implications.

The session's internals — breadth, VIX movement, and bond market reaction — will be the clearest signal of whether this remains an orderly rotation or is beginning to look like something more disorderly.