By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel consensus is neutral, with a focus on Palantir's (PLTR) high-margin software growth and the risks associated with Sandisk's (SNDK) cyclical nature. The key risk flagged is the potential for memory cycles to turn, impacting both Sandisk and Palantir's growth. The key opportunity lies in Palantir's ability to sustain operating margins above 20% as it scales its AIP platform.
Risk: Memory cycles turning, impacting both Sandisk and Palantir's growth
Opportunity: Palantir's ability to sustain operating margins above 20% as it scales its AIP platform
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 →
Sandisk (NASDAQ: SNDK) and Palantir (NASDAQ: PLTR) are two of the most popular artificial intelligence (AI) investments on the market, but both have displayed some weakness lately. Sandisk is down around 25% from its all-time highs established just a few weeks ago, while Palantir has slowly declined from its highs set last October. It's down around 35%.
These two may both be popular, but I think only one is worth buying right now.
Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same "Total Conviction" signal is flashing for a company 1/100th the size of Nvidia. Continue »
Sandisk and Palantir aren't competitors in any way; they have two completely different businesses.
Sandisk manufactures NAND memory, which is best used in scenarios that require long-term data storage. Data centers have an immense need for massive storage, which is eating up a lot of NAND production capacity to produce solid-state drives (SSDs). Lack of supply and elevated demand equate to soaring commodity prices, and that's the mechanism that has helped Sandisk's revenue and profits soar over the past few months.
Two catalysts could cause Sandisk's stock to come crashing down. First, the AI build-out could stall or wrap up, and demand for NAND memory would plummet, returning Sandisk to its old demand curve. With the AI build-out expected to last through at least 2030, I don't see that happening anytime soon. Second, increased production capacity from Sandisk and its peers could increase supply, causing prices to fall, which would have a similar effect.
While the second option is realistic, industry peer Micron Technology informed investors that it expects market tightness to continue in the memory chip market beyond 2027. That's great news for Sandisk, and confirms the short-term investment case for the stock.
Palantir operates an AI software business. It uses AI to help businesses automate workflows and develop applications that ingest large amounts of data and generate actionable insights. This software platform is incredibly popular with the U.S. government, as well as commercial clients. During its most recent quarter, Palantir blew expectations out of the water and reported 85% revenue growth -- with demand coming from both commercial and government sectors.
As more businesses and agencies adopt an AI-first mindset, Palantir's software will remain in high demand, so its growth is far from over.
Both Sandisk and Palantir have great core businesses with bright futures, but there is one key factor that separates these two: valuation.
Both companies are fully profitable and growing rapidly, so valuing their stock based on the forward earnings ratio makes the most sense. However, these two are valued at two opposite ends of the spectrum.
At just 8.4 times forward earnings, there isn't a ton of future success priced into Sandisk's stock, even if investors know it's coming. On the flip side, nearly all of Palantir's upside has been priced into the stock, which will make it hard for the stock to rise much more over the next few years. I think that's the ultimate difference maker between these two.
The market is concerned about Sandisk's future, even though the next year and a half looks bright. Meanwhile, Palantir already has multiple years' worth of growth accounted for. As a result, I wouldn't be surprised to see Sandisk's stock rally to close 2026, while Palantir's continues lower. However, investors need to keep an eye on the memory chip market. If something happens and prices plummet, Sandisk's stock could tumble as a result, so investors need to be prepared and monitor it closely.
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 $397,351! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,304,257!
Now, it’s worth noting Stock Advisor’s total average return is 937% — a market-crushing outperformance compared to 211% 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 July 16, 2026. *
Keithen Drury has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Micron Technology and Palantir Technologies. 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
"The article’s Sandisk thesis collapses because SNDK is not a current ticker or standalone public company, rendering the 8.4x valuation comparison meaningless."
The article pushes Sandisk (SNDK) as the buy at 8.4x forward earnings versus Palantir (PLTR) at an implied 40-50x+ given its 85% growth, citing memory tightness into 2027. Yet Sandisk is not an independent public company; Western Digital (WDC) spun off its NAND business years ago and the ticker SNDK has been defunct since 2016. This renders the entire valuation and “25% off highs” narrative factually broken. Palantir’s commercial acceleration (bootcamps, AIP) is real and government tailwinds sticky, but its multiple already prices in years of 30%+ growth. Memory is cyclical; the article downplays how quickly new fab capacity from Samsung, SK Hynix, and Micron can oversupply NAND once AI capex pauses.
If the article’s core error is simply a ticker mistake and the memory business it describes is actually WDC or Micron, then commodity tightness beyond 2027 could still drive outsized earnings beats and multiple re-rating for the correct names, making the cheap-AI-adjacent play the right contrarian call.
"The article's core premise is fundamentally flawed because Sandisk (SNDK) is a defunct ticker, making any comparative valuation between it and Palantir factually impossible."
The article contains a critical factual error: Sandisk was acquired by Western Digital in 2016 and is no longer a publicly traded entity. Relying on 'SNDK' as a current investment vehicle is impossible, rendering the entire thesis moot. Regarding Palantir (PLTR), the author ignores the stock's transition from a speculative AI play to a S&P 500 constituent with improving GAAP profitability. While the 85% revenue growth figure cited is likely an outlier or misinterpretation of specific segments, the company’s pivot toward high-margin commercial software is structurally sound. Investors should ignore the Sandisk comparison entirely and focus on PLTR’s ability to sustain operating margins above 20% as it scales its AIP platform.
The strongest case against a bullish stance on Palantir is its persistent valuation premium, which leaves zero room for error if commercial growth decelerates or government contract cycles face federal budget headwinds.
"SNDK's valuation discount reflects real cyclical risk, not opportunity—the article mistakes cheap for undervalued without stress-testing the memory cycle or capacity expansion timelines."
The article's valuation argument is mechanically sound—PLTR at ~32x forward P/E versus SNDK at 8.4x creates asymmetric risk/reward. But the piece conflates 'priced in' with 'overvalued,' ignoring that 85% YoY revenue growth at scale justifies premium multiples. More critically: SNDK's thesis is commodity-dependent. Micron's 2027 tightness claim is unverified here, and memory cycle reversals happen fast. PLTR has secular tailwinds (AI adoption, government lock-in); SNDK has cyclical tailwinds that could evaporate if capacity additions accelerate or AI capex normalizes. The article also omits SNDK's gross margin trajectory—if it's already compressing, the 8.4x multiple reflects hidden deterioration, not opportunity.
PLTR's 32x multiple is genuinely stretched if growth decelerates below 40% within 18 months, and government budgets face pressure; SNDK could outperform simply because mean reversion from 8.4x is mathematically easier than justifying 32x expansion.
"Palantir’s lofty multiple is vulnerable to growth slowdowns and policy-risks, making Sandisk the better risk-adjusted bet in the near term."
Shakeout reading: The piece pits Sandisk's price-to-forward-earnings gap against Palantir's growth glow, but it glosses over cycles and cash flow. Sandisk looks cheap on 8.4x forward earnings, yet NAND demand is cyclic and capex-driven; a sustained downturn or faster capacity adds could pressure prices and margins. Palantir's 85% revenue growth looks strong, but the premium on growth is fragile if government budgets tighten or if commercial deployments stall; the article understates Palantir's dependence on marquee contracts, customer concentration, and how far non-GAAP profitability or FCF can travel. In short, one side may be cheap for a reason, the other richly priced for a reason.
Palantir’s government-contract pipeline and enterprise data-analytics moat could sustain a higher multiple even if growth decelerates; this challenges the bearish take on PLTR.
"NAND tightness and PLTR commercial acceleration are positively correlated, not opposing forces."
Claude's margin compression warning for the memory play is the sharpest observation so far. Yet nobody has flagged that WDC's enterprise SSD exposure now overlaps with PLTR's data-center AI workloads; any NAND tightness into 2027 would directly lift server OEM capex and therefore PLTR's commercial bootcamp traction. The supposed 'cyclical vs secular' divide is more symbiotic than the panel admits.
"Hardware-driven memory cycles and software-driven AI growth represent fundamentally different risk profiles that should not be conflated."
Grok, your 'symbiotic' view ignores the fundamental mismatch in capital intensity. WDC’s enterprise SSD growth requires massive, recurring fab capex—a margin-dilutive drag—whereas Palantir’s AIP is high-margin software with negligible marginal cost per new user. Comparing them as a combined trade is dangerous; if memory cycles turn, WDC’s heavy balance sheet suffers, while Palantir’s software moat remains intact. The 'symbiosis' is a trap for investors who confuse hardware infrastructure with software scalability.
"NAND scarcity is bullish for WDC's pricing but bearish for enterprise IT capex velocity—a headwind Palantir can't fully escape."
Gemini's capital-intensity argument is sound but incomplete. WDC's margin pressure from fab capex is real—but that same capex shortage is precisely what creates NAND pricing power into 2027. Palantir's software moat survives a memory cycle, yes. But if NAND scarcity drives server OEM margins down, enterprise IT budgets contract, and bootcamp adoption slows. The symbiosis Grok flagged cuts both ways: tightness helps WDC's ASPs but potentially starves Palantir's TAM expansion.
"The 'symbiosis' between NAND tightness and Palantir's data-centre demand is not proven; server capex may rise, but software budgets depend on IT strategy, so PLTR's margin upside may not materialize even if WDC benefits."
Grok, the symbiosis claim hinges on a direct, one-to-one lift from NAND tightness to Palantir's data-centre demand and bootcamp traction. That linkage isn't proven: server capex may rise, but budget allocation for software like AIP depends on enterprise IT strategy, not memory cycles. If capex lags or budgets tighten, PLTR's margin upside could disappoint even as WDC benefits. The article's SNDK error aside, the correlation is weaker risk than you imply.
The panel consensus is neutral, with a focus on Palantir's (PLTR) high-margin software growth and the risks associated with Sandisk's (SNDK) cyclical nature. The key risk flagged is the potential for memory cycles to turn, impacting both Sandisk and Palantir's growth. The key opportunity lies in Palantir's ability to sustain operating margins above 20% as it scales its AIP platform.
Palantir's ability to sustain operating margins above 20% as it scales its AIP platform
Memory cycles turning, impacting both Sandisk and Palantir's growth