Palantir and Sandisk Stocks Are Down 35% and 25%, but Only One Is a Buy Now

The article contrasts Sandisk (SNDK) and Palantir (PLTR): SNDK is down ~25% from its recent highs amid an ongoing memory chip shortage, while PLTR is down ~35% from its last October peak. It argues Sandisk has near-term tailwinds from NAND tightness (Micron expects tightness to continue beyond 2027) but warns its stock could sharply decline if memory supply expands or AI/storage demand stalls; Palantir is cited for 85% quarterly revenue growth driven by commercial and government demand, but the stock is viewed as more fully valued (forward P/E ~8.4x for Sandisk vs “nearly all upside” priced into Palantir). Net takeaway: Sandisk’s risk is primarily memory pricing/supply dynamics, while Palantir’s risk is valuation and limited upside.

Analysis

The cleaner relative value is long memory-cycle exposure over AI-software duration risk. SNDK is effectively a leveraged bet on tight NAND pricing: if spot pricing stays firm, incremental gross margin can expand faster than revenue, and the stock’s low multiple leaves room for rerating as earnings visibility improves. The second-order winner is MU, which benefits from the same supply discipline with a larger balance sheet and better survivability if pricing softens; the losers are downstream storage buyers and any data-center capex names that were assuming cheaper memory in 2025-26 budgets.

PLTR’s problem is not business quality, it’s expectation density. At this valuation, even strong execution can underperform if growth merely normalizes from hyper-growth to high growth; the stock is more exposed to multiple compression than to absolute revenue misses. The market is underweight the possibility that government demand remains resilient while commercial growth decelerates, which would still be “good business, bad stock” because the market is paying for persistent >70% growth.

The key timing issue is that SNDK’s upside is tied to a 1-3 quarter pricing tape, while PLTR’s downside can unfold over the next 1-2 earnings prints if the market starts marking down terminal growth assumptions. The main falsifier for the SNDK bull case is evidence of NAND oversupply: falling spot prices, rising inventory days, or capex announcements that push supply ahead of demand. For PLTR, the thesis breaks if commercial growth re-accelerates without margin giveback, or if a major government award extends the runway enough to justify the multiple.

Key Decisions for Investors

• Put on a dollar-neutral long SNDK / short PLTR pair trade into the next 1-2 earnings cycles; target 15-25% relative upside if NAND pricing stays firm and PLTR multiple compresses. Risk: if PLTR re-accelerates and SNDK spot pricing rolls over, the pair can move sharply against you.
• Overweight MU versus the broad semis complex as the higher-quality expression of the memory-tightness thesis for the next 6-12 months; MU is better positioned if the cycle persists longer than expected and offers a cleaner hedge if SNDK proves too single-name volatile.
• For tactical downside in PLTR, buy 3-6 month put spreads rather than outright shorts; the risk/reward is better because implied volatility is likely elevated, and the thesis only needs one earnings report showing growth deceleration for the spread to work.
• Set an alert on NAND spot prices and SNDK inventory commentary; if pricing drops or inventories build for two consecutive months, cut SNDK exposure immediately because the valuation support disappears quickly.
• If you want to stay agnostic, rotate away from high-multiple AI software baskets into cash-generative semis and memory suppliers; the next 1-3 months should favor earnings power over narrative premium.