Why Sandisk Stock Crashed on Friday

Sandisk stock fell 8% in early trading after gaining 22% the prior day, as investors reacted to reports that OpenAI may delay its IPO until 2027. A later OpenAI listing could postpone spending on AI infrastructure and memory chips, which would delay a potential demand tailwind for Sandisk. The move also reflects profit-taking after Micron's strong Q3 report lifted the stock sharply on Thursday.

Analysis

The market is treating SNDK as a near-term AI capacity derivative, but that framing is too linear. If OpenAI pushes out its financing event, the pain is not just a timing delay in memory demand; it also gives hyperscalers more room to optimize capex mix toward higher-ROI compute before locking in incremental storage purchases, which can compress the implied upside for flash suppliers even after the next funding milestone.

The bigger second-order effect is relative positioning. MU looks better insulated because the market already rewards its earnings power and balance-sheet discipline, while SNDK is being traded more as a beta expression of AI spend optionality. That makes SNDK more vulnerable to flow-driven unwinds after a sharp one-day move, especially if investor attention shifts from narrative acceleration to actual order visibility over the next 1-2 quarters.

Near term, this is a classic “good news gap” setup: a stock that rerated on an AI headline can give back a meaningful portion of the move if the catalyst slips even modestly. The contrarian view is that a 2027 IPO delay may not materially reduce total OpenAI-related demand, only shove it rightward; if investors are willing to look past the calendar, the current selloff could overstate the long-term impact. But in the next 30-90 days, timing matters more than terminal size, and that favors fading the most crowded expression of the trade.

Key Decisions for Investors

• Fade strength in SNDK via a short or put spread on any rebound over the next 1-2 weeks; thesis is timing compression, not demand destruction, with upside limited unless another AI-capex headline lands quickly.
• Prefer MU over SNDK on a relative-value basis for the next quarter: long MU / short SNDK captures the difference between durable earnings delivery and narrative-dependent multiple expansion.
• If maintaining AI memory exposure, use call spreads rather than outright longs in SNDK to cap downside from a 30-50% retracement of the recent spike while preserving upside if the IPO timeline is reaffirmed.
• Monitor hyperscaler capex commentary over the next earnings cycle; a shift toward efficiency or deferred storage deployment would extend the underperformance window for SNDK by 1-2 quarters.