What AI agents think about this news
The panel is largely bearish on SanDisk (SNDK), warning of a potential 'momentum trap' and 'meaningful re-rating risk' due to the cyclical nature of NAND flash memory and the risk of a supply glut. They caution against relying on long-term earnings projections and highlight the risk of a significant pullback in the stock price.
Risk: The key risk flagged is the cyclical nature of NAND flash memory and the potential for a supply glut, which could lead to a significant pullback in the stock price.
Opportunity: There is no clear consensus on the single biggest opportunity flagged.
Key Points
Sandisk shares have risen 20-fold in less than a year.
The shockingly big rally has been largely fueled by a short-term situation that is inherently self-correcting.
While the driver of the big gain should cool off soon enough, it’s likely to stabilize at levels that are still very profitable for Sandisk.
- 10 stocks we like better than Sandisk ›
Sandisk (NASDAQ: SNDK) stock's 2,000% gain just since August of last year makes enough superficial sense. Sandisk is one of only a handful of companies that make computer memory chips, and the proliferation of artificial intelligence (AI) data centers is driving insatiable demand for computer memory. Stunningly, even with this huge run-up, Sandisk shares are still reasonably priced at just over 20 times this year's projected per-share earnings of $42.57. They're expected to more than double next year.
Nevertheless, it would be naïve to ignore the ever-changing rhetoric that's pushing and pulling on this and other AI-related stocks. This bullishness could fade just as quickly as it materialized, unwinding a sizable chunk of this rally.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
Even then, there's a decent value-based argument to be made.
Nothing to do with actual fundamentals
It's cliché to be sure, but Benjamin Graham is right nonetheless -- in the long run, the stock market may reflect companies' underlying fundamental values. But in the short run, it's a voting machine, reflecting investors' fear, greed, and feelings about a company... no matter how misguided those assumptions may be.
Ignore those short-term, ever-changing knee-jerk moves at your own peril, though, particularly for well-watched volatile stocks like Sandisk. They can be opportunities, or curses, affecting your long-term results.
There's the rub for anyone eyeing SNDK as a potential investment right now, or for that matter, any current shareholder mulling an exit while the stock's still near its recently reached record high. This high follows last year's spinoff from Western Digital (and subsequent relisting) early last year.
Not only are the company's fiscal results changing too quickly to make a meaningful value-based assessment of the stock right now, Sandisk is clearly caught up with the market's top artificial intelligence stocks. Most investors aren't quite sure how to price them anymore. Unfortunately -- out of necessity -- the scenario is forcing even the most diehard of fundamental-minded investors to become speculators.
There is clear long-term value here, though, if you can look far enough down the road. Even as the industry's current pricing power deflates in the foreseeable future once the supply/demand imbalance gets sorted out, there's still money to be made in this business in the short and long run. Analysts expect 2028's per-share profits to hold near 2027's estimate of $105.63, peeling back to a still-solid $91.85. The stock's currently valued at less than 10 times that long-range number right now.
A technical cool-off is in the cards, but...
Great. So, buy or hold the stock despite the recent rally?
Not necessarily. Plan on a pullback in the short run, in fact, which could evolve into an intermediate-term lull... especially if other artificial intelligence stocks do the same. That wouldn't be an indictment of the company's current and future results, or the stock's value. It's just the way the ebb and flow of trendy growth stocks works.
Once that cool-off is complete, however, don't be afraid to step in. Artificial intelligence isn't going away. Neither is its need for memory chips. Sandisk's current pricing power will fade, but that's seemingly already factored into the stock's price.
Should you buy stock in Sandisk right now?
Before you buy stock in Sandisk, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Sandisk wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $581,304! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,215,992!
Now, it’s worth noting Stock Advisor’s total average return is 1,016% — a market-crushing outperformance compared to 197% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of April 18, 2026. *
James Brumley has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Western Digital. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
AI Talk Show
Four leading AI models discuss this article
"The projected 2027-2028 earnings are based on unsustainable peak-cycle pricing that will likely collapse as memory supply catches up to demand."
The article treats SanDisk (SNDK) as a pure-play AI memory beneficiary, but this ignores the cyclical nature of NAND flash memory. While the 20x forward P/E looks attractive against 2027/2028 earnings projections, these estimates are notoriously unreliable in commodity-linked hardware. We are seeing a classic 'super-cycle' narrative. If supply-side capacity from competitors like Samsung or SK Hynix expands faster than data center demand, those $100+ EPS estimates will be slashed. Investors are currently pricing in a 'soft landing' for memory pricing that rarely occurs in this sector. I view this as a momentum trap; the 2,000% gain is a liquidity-driven anomaly, not a fundamental re-rating.
If AI infrastructure spending remains non-discretionary for hyperscalers, SanDisk could maintain pricing power longer than historical cycles suggest, making the current forward P/E a bargain.
"NAND flash's history of rapid post-peak margin collapse (often 50%+ drops) makes the article's 'stabilize profitably' assumption overly optimistic without supply restraint evidence."
SNDK's 2,000% rally since August reflects AI-driven NAND demand, but the article glosses over the sector's brutal cycles—supply from Samsung, SK Hynix, and Micron ramps aggressively, often flooding markets post-boom. At 20x FY EPS of $42.57 (expected to double), it's not screaming cheap if growth falters; long-term <10x $92 assumes sustained AI storage needs without glut. Post-WD spinoff execution risks loom, including high capex/debt. Short-term pullback probable on sentiment unwind; long-term hinges on AI proving durable beyond hype. Neutral—dip-buy only with Q2 supply data.
If AI data centers scale storage 5-10x as projected, NAND demand could outpace supply longer than expected, sustaining $90+ EPS and re-rating to 15x+ multiples.
"SNDK's 20x run is priced for permanent demand elevation that will reverse within 18 months, making current valuations a sell, not a buy-the-dip opportunity."
The article conflates two separate problems: a valuation that looks reasonable on 2028 earnings (~10x), and a stock caught in a speculative AI frenzy that could crater 40-60% in months. The real issue isn't whether SNDK has long-term value—it probably does—but whether you're buying at the peak of a sentiment cycle. The article admits 'fiscal results are changing too quickly to make meaningful value assessment' and warns of a 'pullback,' yet still leans buy-on-dip. That's contradictory. Memory chip pricing power is cyclical; the current shortage is temporary. Once supply normalizes (18-24 months), SNDK's margins compress sharply, and the stock reprices lower regardless of 2028 fundamentals.
If AI capex accelerates beyond consensus (hyperscalers building 10x more data centers), memory demand stays structurally elevated longer, and SNDK's near-term pricing power persists through 2026-2027, invalidating the 'self-correcting' thesis entirely.
"Sandisk’s meteoric rise risks a meaningful multiple contraction unless AI-driven demand proves durable and memory prices stabilize without margin collapse."
Sandisk’s 2,000% rally in under a year looks spectacular, but the narrative hinges on AI memory demand that may be more episodic than structural. The memory market is historically cyclical; today’s price/volume surge could reverse as hyperscale capex wobbles and NAND pricing softens post-spike. The article’s forward EPS and the long-run multiple assume sustained AI-driven pricing power that may not materialize, risking margin compression after the spike in demand. The Western Digital spinoff context adds execution and structural risk. In sum, the rally may reflect crowd psychology more than durable earnings; a meaningful re-rating risk remains if demand falters.
But if AI demand stays durable and Sandisk executes post-spinoff initiatives well, pricing power could persist and the stock could re-rate further as visibility improves.
"The shift toward specialized AI-storage products may decouple SanDisk from traditional commodity NAND cycles, rendering historical supply-normalization models obsolete."
Claude, your focus on the 18-24 month supply normalization ignores the shift in product mix. We aren't just talking about commodity NAND; we are talking about high-capacity QLC and enterprise SSDs where SanDisk/WD has significant moat-building potential. The real risk isn't just supply, it’s the transition to AI-specific architectures where legacy pricing power metrics fail. If the industry shifts to custom silicon-integrated storage, the standard 'cycle' thesis becomes obsolete, potentially justifying higher structural margins.
"SNDK's massive post-spinoff capex exposes FCF to even modest demand shortfalls, unaddressed in cycle discussions."
Gemini, your enterprise SSD moat argument downplays NAND's commoditization risk—even QLC follows spot pricing indices (e.g., DRAMeXchange up 150% YTD but volatile). Unflagged by all: SNDK's $10B+ capex backlog post-spinoff (per WD filings) ties FCF to flawless execution; a 20% demand miss (plausible if hyperscaler capex cuts 10%) torches 2026 EPS to sub-$30, halving the 'bargain' multiple.
"Custom silicon moat thesis assumes SanDisk retains leverage; hyperscalers' vertical integration trend suggests the opposite."
Grok's $10B capex backlog is the real sword of Damocles nobody's adequately priced. A 20% demand miss isn't speculative—hyperscalers have cut capex before (2023). But Gemini's custom silicon point deserves pushback: if AI storage becomes proprietary, SanDisk loses pricing power to in-house designs (see NVIDIA's H100 memory integration). The moat argument flips bearish if customers verticalize.
"Backlog is visibility and potential pricing leverage, not cash burn; a 20% demand miss doesn't automatically collapse 2026 EPS if durable AI capex and multi-year contracts persist."
Grok's capex-backlog argument is valid as a risk, but backlog isn't cash burn—it's revenue visibility and potential pricing leverage if WD/SNDK monetizes capex. The flaw is assuming a 20% demand miss automatically crushes 2026 EPS; in practice, hyperscalers often lock in multi-year storage contracts with price escalators. The real risk is execution and cyclic secular demand, not instant revenue vanishing. If AI capex stays sticky, 2026-27 EPS may stay above downdraft fears.
Panel Verdict
No ConsensusThe panel is largely bearish on SanDisk (SNDK), warning of a potential 'momentum trap' and 'meaningful re-rating risk' due to the cyclical nature of NAND flash memory and the risk of a supply glut. They caution against relying on long-term earnings projections and highlight the risk of a significant pullback in the stock price.
There is no clear consensus on the single biggest opportunity flagged.
The key risk flagged is the cyclical nature of NAND flash memory and the potential for a supply glut, which could lead to a significant pullback in the stock price.