Sandisk: Priced For The AI Apocalypse

Sandisk’s Q1 revenue rose 378% year over year to $11.03B, with 85.1% non-GAAP gross margin and $5.7B in adjusted free cash flow; the company beat estimates, raised guidance and announced an 8-for-1 stock split. Data centers accounted for 43.5% of its business, and long-term contracts were reported to secure $93.9B in minimum revenue, including $16.5B in guarantees. Despite the results and outlook, shares fell 13%.

Analysis

The selloff suggests the market is discounting durability, not the quarter: at these reported margins and cash generation, even modest normalization in NAND pricing, mix, or utilization could overwhelm near-term estimate beats. Data-center demand improves the mix but also ties the thesis more tightly to a small set of hyperscalers’ capex plans and to customers’ ability to qualify alternative suppliers. The second-order risk is that capacity commitments made into strong pricing can become costly if demand or pricing turns before supply can be redirected.

Treat the cited contract floor and guarantees as downside mitigants only after checking the filing: duration, cancellation and delivery conditions, counterparty concentration, and whether guarantees are cash-secured or conditional matter more than headline value. They do not, by themselves, establish margin protection. The stock split changes neither value nor cash flows.

Days: the 13% decline can reflect elevated expectations and positioning; avoid assuming it is a bargain signal without price/valuation context. Over 1–3 months, track backlog conversion, data-center customer concentration, NAND pricing, and guidance revisions. Over 6–18 months, the key question is whether data-center demand supports structurally higher economics or merely amplifies the next memory cycle. The thesis weakens if subsequent guidance or reported margins roll over, contracts prove cancellable/conditional, or customer capex and NAND pricing deteriorate. It strengthens if margins and cash conversion remain resilient through a pricing slowdown.

Key Decisions for Investors

• Do not treat the split or headline contract totals as a catalyst to buy. Before adding SNDK, verify contract terms and concentration in the next filing, and compare the selloff with valuation and implied expectations; those data are absent here.
• For existing SNDK exposure, consider a temporary, defined-risk put spread only if the position is oversized and option pricing is reasonable; avoid naked puts or chasing protection after a sharp drop. Reassess after the next guidance update and NAND-pricing data.
• Set a watch, not an automatic short: sustained margin deterioration or a guidance cut would support reducing/hedging SNDK, while stable margins and firm data-center demand despite weaker pricing would falsify the near-term peak-cycle bear case.
• Monitor memory competitors, including Micron and Samsung, for capacity, pricing, and customer-qualification signals. Evidence of easier substitution or aggressive supply additions would weaken SNDK’s contract-based downside protection.