By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel consensus is bearish on Western Digital (WDC), previously SanDisk, due to cyclical and structural risks in the NAND memory market. Key concerns include margin compression, customer concentration, and monopsony power of hyperscalers.
Risk: Monopsony power of hyperscalers leading to margin compression and unsustainable pricing.
Opportunity: Potential near-term mitigation of margin compression through Western Digital's enterprise SSD/data-center mix.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Shares in Sandisk have exploded upward because of rising AI-driven demand for memory and storage.
Investors should remember that memory is a cyclical industry.
Sandisk (NASDAQ: SNDK) is the type of stock that growth-focused investors dream about. If you had bet $10,000 on the company just 12 months ago, you would have $327,200 today -- a return of roughly $3,272% that has been driven by optimism about the surging demand for memory and data storage hardware for artificial intelligence (AI).
But past performance doesn't guarantee future results. And investors who missed the big rally will be curious to know if Sandisk can maintain its explosive momentum. Let's dig deeper into the pros and cons of the company to decide what the next five years might have in store.
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 »
For regular consumers, Sandisk is probably most recognizable for its USB flash drives, memory cards, and portable SSDs, which help people store things like documents and photos outside of their computers. But in the enterprise space, the company has become a leader in high-performance memory and storage solutions, which are vital for data centers that need to serve surging demand for generative AI-related workloads.
Memory is so important because it stores the vast amount of training and inference data that large language models (LLMs) need to operate. And Sandisk stands out because of its focus on NAND flash technology.
Unlike traditional hard disks, NAND flash is designed to store data electronically, with no moving parts, making it faster and more energy-efficient. While upfront costs can be greater, this can often make better economic sense for data center clients that need to process and store massive volumes of information around the clock.
Right now, the memory industry is caught in a perfect storm of rising demand and short supply, driving explosive growth and margins for the major players. Sandisk is no exception. Third-quarter revenue surged 251% year over year to $5.95 billion, driven mainly by data center demand and edge computing, which often refers to data solutions embedded within devices themselves instead of at a data center.
Perhaps most importantly, the company's gross margin is also exploding -- rising from just 22.7% in the third quarter of 2025 to a whopping 78.4% today. This number puts Sandisk ahead of the AI industry leader Nvidia, which reported a gross margin of 75% in its most recent quarter.
Investors love improving margins because they translate into better operating leverage, a metric that measures how efficiently a company converts top-line growth into profits. And Sandisk's third-quarter earnings per share skyrocketed from a loss of $0.30 to a gain of $23.41, easily explaining the stock's rapid rise as investors quickly reevaluate its intrinsic value. This trend looks set to continue in the near term as AI-related demand remains high and memory shortages keep prices elevated.
Despite its surging stock price, Sandisk's shares remain quite cheap at a forward price-to-earnings (P/E) ratio of 24. For context, the S&P 500 averages 22, but Sandisk is growing significantly faster than the average company and would typically command a much higher premium.
Investors may still be cautious because the memory hardware industry is notoriously cyclical. Periods of booming demand are typically followed by a rapid increase in production capacity, leading to a supply glut and a decline in prices. There is also a risk that the current AI data center demand boom is itself a bubble that won't last much longer.
Over the next five years, these will be real risks for Sandisk. And it might make more sense to look for the next undiscovered growth phenom rather than betting on this company after its price has already risen so rapidly.
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 $483,476! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,362,941!
Now, it’s worth noting Stock Advisor’s total average return is 998% — a market-crushing outperformance compared to 207% 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 May 19, 2026. *
Will Ebiefung has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Nvidia. 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.
Four leading AI models discuss this article
"Cyclical capacity additions will likely erode Sandisk's margins within 18 months despite current AI tailwinds."
The article touts Sandisk's 251% revenue surge and 78.4% gross margins from AI NAND demand but glosses over the sector's structural cycles. Memory makers historically ramp capacity aggressively once prices rise, flooding supply within 12-18 months and crushing ASPs. No mention is made of Sandisk's 2016 acquisition by Western Digital, which makes the standalone SNDK ticker and valuation narrative historically inaccurate. Second-order risk is hyperscaler capex digestion that could hit before new fabs come online.
Persistent AI training and inference workloads may structurally lift baseline demand above prior cycles, delaying any glut and supporting elevated margins for leaders like Sandisk.
"Sandisk's 78% gross margin is a cyclical peak, not a new normal, and the stock's valuation assumes margin stability that memory economics cannot support."
The article's 3,272% return claim is almost certainly fabricated or misleading—SNDK doesn't trade under that ticker historically, and no memory stock has delivered those returns in 12 months. Setting that aside: the real risk isn't cyclicality (which the article acknowledges), but *timing*. A 24x forward P/E on a memory company is cheap only if margins stay at 78%—structurally impossible. Memory is commoditizing faster than the article admits. NAND pricing peaked in Q3 2025 and is already rolling over. The article treats current supply tightness as durable; it's temporary. Gross margin compression from 78% to 40-50% over 18-24 months is the base case, not the bear case.
If AI inference workloads genuinely shift to edge computing and embedded memory (as the article hints), Sandisk's NAND focus could be structurally advantaged for a decade, not a cycle.
"SanDisk is no longer a standalone public company, rendering the entire investment thesis in the article factually impossible."
The article's premise is fundamentally flawed because it discusses SanDisk (SNDK) as a standalone public entity, ignoring that Western Digital acquired it in 2016. The financial data cited—such as a 3,272% return and $23.41 EPS—appears to be either a hallucination or a misattribution of Western Digital's (WDC) recent volatility. Investors cannot 'buy' SanDisk stock today. Even if we treat this as a proxy for WDC, the 78.4% gross margin claim is highly suspect for a NAND manufacturer, which typically faces intense commodity pricing pressure. This looks like a 'zombie' article using outdated ticker data to bait retail investors into a non-existent security.
If the author is actually referring to a spin-off or a specific tracking stock I am unaware of, the massive margin expansion could reflect a successful pivot to high-margin enterprise SSDs, but this is highly improbable given the commodity nature of NAND.
"Sandisk's steep rally rests on an AI-driven growth narrative that is unlikely to persist at current margins, making the five-year upside vulnerable to cyclical turnover and multiple compression."
The article markets Sandisk as an AI-memory miracle with 251% YoY revenue to $5.95B and 78.4% gross margin, but memory is highly cyclical and pricing brittle. AI-driven demand may be episodic; NAND pricing tends to fall when capacity expands, squeezing margins. The claimed margins outsized relative to peers, and the forward P/E of 24 may already reflect optimistic assumptions; execution risks include supplier/capacity ramp, customer concentration, and global capex cycles. Without a sustained multi-year AI capex wave, the stock's multiple looks vulnerable to a re-rating if growth cools.
Even if AI demand persists, NAND pricing often collapses during capacity waves, and Sandisk may never sustain 78% gross margins; a normalization in data-center capex would likely crush earnings versus current expectations.
"Western Digital's diversification could extend high margins beyond Claude's 18-24 month forecast."
Claude's margin compression timeline assumes pure-play NAND dynamics, but ignores Western Digital's post-2016 integration allowing hedging via enterprise storage solutions. This could delay the drop from 78% to 40% gross margins by 6-12 months if AI workloads migrate to hybrid flash-HDD setups. Yet this same integration exposes WDC to broader IT spending slowdowns, amplifying the capex digestion risk rather than mitigating it.
"Hyperscaler supply diversification poses a steeper demand cliff than commodity NAND cyclicality."
Grok's hybrid flash-HDD hedge is plausible but unverified—WDC's 10-K doesn't break out margin contribution by storage type. More pressing: nobody's flagged customer concentration risk. If hyperscalers (Meta, MSFT, Google) represent 40%+ of NAND revenue and they're already building in-house memory or diversifying suppliers post-2025, the 251% revenue growth becomes a cliff, not a cycle. That's the real timing bomb.
"Hyperscaler monopsony power will structurally compress NAND margins regardless of AI demand levels."
Claude, your focus on hyperscaler concentration is the missing link. While Grok worries about capacity cycles, the real risk is 'monopsony power.' If three hyperscalers dictate pricing, those 78% margins are not just cyclical—they are structurally unsustainable. Even if AI demand persists, these customers will aggressively leverage their buying power to commoditize NAND, forcing WDC to trade margin for volume. This isn't just a cycle; it's a permanent shift in bargaining power that makes current valuations dangerous.
"Monopsony risk exists, but margins may be cushioned by WD's enterprise mix; 78% GM is unlikely to be sustainable under a prolonged capex cycle."
Gemini correctly flags monopsony risk, but the punchline shouldn't be 'structural margins collapse.' The overlooked point is WD's enterprise SSD/data-center mix that could cushion near-term margins while hyperscaler procurement pressures push ASPs lower. The bigger risk remains a prolonged capex digestion cycle and a delayed re-rating if AI spend accelerates in 2025-26 but slows afterward. A 78% gross margin looks increasingly unsustainable.
The panel consensus is bearish on Western Digital (WDC), previously SanDisk, due to cyclical and structural risks in the NAND memory market. Key concerns include margin compression, customer concentration, and monopsony power of hyperscalers.
Potential near-term mitigation of margin compression through Western Digital's enterprise SSD/data-center mix.
Monopsony power of hyperscalers leading to margin compression and unsustainable pricing.