By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel consensus is overwhelmingly bearish on the article's thesis that Sandisk (now a segment of Western Digital) offers superior growth and upside versus Nvidia amid Mag7 capex. Key risks include cyclical NAND pricing and margins, potential oversupply, and margin compression from competition. The 'millionaire-maker' storage play narrative is considered fragile and high-risk.
Risk: Cyclical NAND pricing and margins, potential oversupply, and margin compression from competition
Opportunity: None identified as the panel was largely critical of the article's thesis
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 →
Nvidia and Sandisk are in the middle of the AI infrastructure boom, driven by the terrific demand for their chips deployed in AI data centers.
Both companies are on track to deliver phenomenal earnings growth, but investors looking to buy one of these two stocks have an easy decision to make.
The "Magnificent Seven" are a group of seven mega-cap technology companies with enormous valuations that have a sizable influence on the stock market.
Alphabet, Amazon, Apple, Meta Platforms, Microsoft, Nvidia (NASDAQ: NVDA), and Tesla make up the Magnificent Seven, and their earnings growth is expected to be significantly higher than the broader market over the next couple of years. It is worth noting that this group has been spending heavily to build artificial intelligence (AI) infrastructure.
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 »
Alphabet, Amazon, Microsoft, and Meta are poised to spend a collective $725 billion on capital expenses this year, up by 77% from last year's record outlay. Meanwhile, Tesla's capital spending is expected to jump to more than $25 billion this year from $8.5 billion last year, driven by investments in humanoid robots and robotaxis. Though Apple has been conservative with capital spending, the iPhone maker has still spent a notable $4.3 billion on capex over the past couple of quarters.
So, the total capex of the Magnificent Seven could exceed $750 billion in 2026. Investors looking to build a million-dollar portfolio can capitalize on this spending spree by buying top tech stocks poised to grow earnings faster than the broader market. Nvidia and Sandisk (NASDAQ: SNDK) are two stocks with massive upside potential that are benefiting from this enormous AI infrastructure spending.
Let's see why these stocks are ideal fits for a potential million-dollar portfolio and check which one is the pick of the lot.
While Nvidia is part of the Magnificent Seven, it is one of the biggest beneficiaries of this group's massive capital spending. Nvidia's chip systems are the cornerstone of the AI infrastructure that's being built by the major hyperscalers in the U.S. When Nvidia revealed its next-generation Vera Rubin AI processors earlier this year, the company noted that this platform will be deployed widely by the top four hyperscalers in the U.S.
Additionally, there is strong demand for Nvidia's Vera Rubin processors from pure-play AI companies such as OpenAI and Anthropic. At the same time, AI start-ups, neocloud providers, and server makers are also expected to deploy these chips on a big scale. The good part is that the strong demand for Nvidia's AI chips is poised to translate into stronger growth for the company.
Nvidia reported 85% year-over-year revenue growth in the first quarter of fiscal 2027 (which ended on April 26). Its $91 billion revenue estimate for the current quarter points toward a stronger jump of 95% from the year-ago period. So, it is easy to see why analysts are predicting a larger increase of 88% in its earnings per share this fiscal year to $8.94 per share. That's well above the 60% growth it delivered in the previous fiscal year.
With the stock trading at just 25 times forward earnings, a discount to the tech-laden Nasdaq-100 index's forward earnings multiple of 27, investors are getting a great deal on Nvidia stock right now. After all, it should ideally trade at a premium after a year, given its market-beating earnings growth potential.
Similarly, Sandisk is also trading at an attractive 22 times forward earnings. The storage specialist's bottom line has been growing exponentially, as the demand for its NAND flash storage solutions is exceeding supply. Importantly, the AI boom should ensure that storage demand continues to grow at an exceptional pace.
According to a research report, the AI-focused storage market's revenue could increase from $36 billion last year to almost $322 billion in 2035. That's not surprising, as supporting AI workloads such as training models and running inference applications requires large datasets that must be quickly fed to AI accelerators in data centers and at the edge.
This explains why McKinsey is estimating 35% annual growth in sales of enterprise solid-state drives (SSDs) that Sandisk manufactures through 2030. The consulting firm notes that AI training needs will drive an 11x increase in SSD content per server between 2024 and 2030, while AI inference will drive a 7x jump in the same.
Given that the memory chip shortage could continue until 2030, it is easy to see why Sandisk's earnings are set to rise at a parabolic pace from fiscal 2025 levels of $2.99 per share.
It isn't too difficult to see which one of these two AI stocks is worth buying hand over fist right away. While both of them are growing at solid rates, Sandisk's growth is outpacing Nvidia's significantly. Of course, Nvidia is anticipated to deliver healthy double-digit earnings growth over the next three years, but its average growth rate isn't expected to be anywhere close to Sandisk's.
Additionally, Sandisk has a slightly cheaper forward earnings multiple. If Nvidia and Sandisk were to trade at an identical 30 times forward earnings after three years, the latter's stock price upside will be much higher. That's why Sandisk looks like an ideal investment for anyone looking to build a million-dollar portfolio. The key role that memory chips are playing in AI infrastructure should ensure that Sandisk's red-hot earnings growth continues beyond the next two to three years, paving the way for multibagger gains in the long run.
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 $443,191! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,258,838!
Now, it’s worth noting Stock Advisor’s total average return is 941% — 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 June 6, 2026. *
Harsh Chauhan has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Alphabet, Amazon, Apple, Meta Platforms, Microsoft, Nvidia, and Tesla. 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
"Capex scale alone won’t reliably translate into outsized stock gains without durable end-demand and cyclical memory-market stability; the article’s thesis rests on fragile, optimistic assumptions."
The article leans on a megacap capex boom to justify bets on NVDA and what it calls Sandisk, but Sandisk isn’t a standalone public name anymore (it's a Western Digital segment). The thesis rests on an AI capex surge that’s highly cyclical; memory markets price in volatility and NAND pricing can derail earnings, while Nvidia’s growth depends on end-demand durability and data-center capex more than raw spending scales. Claims of attractive multiples and outsized earnings assume a sustained, disorderly AI uplift; geopolitical, supply-chain, and policy headwinds could throttle upside. A capex-driven narrative risks mispricing risk if AI demand cools or realization lags.
Even if the capex boom persists, the mislabeling of Sandisk undermines credibility, and memory cycles could reverse faster than expected; in that case, the purported upside collapses and multiples compress sharply.
"The article is factually compromised by recommending a company that no longer exists as an independent entity, rendering its valuation and growth analysis void."
The article's premise is fundamentally flawed: it recommends 'Sandisk' (SNDK), a company acquired by Western Digital in 2016. Relying on outdated data to build a 'millionaire-maker' thesis is a massive red flag. While the AI infrastructure capex boom is real, the article ignores the cyclical nature of the NAND flash memory market, which is prone to volatile boom-bust cycles that often erode margins. Nvidia remains the clear leader, but at a $3T+ market cap, the law of large numbers makes 88% EPS growth extremely difficult to sustain. Investors should focus on the underlying hardware demand but be wary of analysts citing defunct tickers as growth plays.
If memory supply remains constrained through 2030 as the article claims, the resulting pricing power could lead to unprecedented margin expansion for actual NAND producers like Western Digital or Micron, regardless of the article's ticker error.
"Sandisk's valuation already reflects the AI storage thesis; the article mistakes a crowded trade for an undiscovered opportunity and ignores that commodity NAND producers face structural margin pressure from competition."
The article conflates two separate theses without rigorous support. Yes, Mag7 capex is real—$725B collective spend is documented. But the leap to Sandisk (SNDK) as a 'millionaire-maker' rests on: (1) assumed 35% annual SSD growth through 2030 (McKinsey estimate, not guarantee), (2) a memory shortage lasting until 2030 (speculative), and (3) forward multiples re-rating identically for both stocks. The article ignores that SNDK trades at 22x forward P/E precisely because the market already prices in AI tailwinds. Nvidia's 25x multiple despite 88% EPS growth suggests either Nvidia is undervalued OR the market is skeptical about sustaining that growth—not that SNDK is cheaper. The article also omits competitive dynamics: Samsung, SK Hynix, and Kioxia all produce NAND; SNDK has no moat here.
If the memory shortage extends and AI training truly demands 11x SSD content per server by 2030, Sandisk's earnings could compound faster than modeled, justifying multiple expansion beyond 30x. The article may be early, not wrong.
"Sandisk has not existed as a standalone public company with ticker SNDK since its 2016 acquisition, rendering the buy recommendation invalid on its face."
The article's core thesis—that Sandisk offers superior growth and upside versus Nvidia amid Mag7 capex—rests on shaky ground. SNDK has not traded since Western Digital's 2016 acquisition, so any forward P/E or EPS projections for a nonexistent ticker are meaningless. Even treating this as WDC, NAND cyclicality, potential oversupply post-2025, and margin compression from competition undermine the parabolic-growth narrative. Nvidia's 88% EPS growth and 25x multiple already price in AI dominance; Sandisk's claimed edge ignores execution and balance-sheet realities. Investors chasing 'millionaire-maker' storage plays face higher risk of mean reversion than the piece admits.
If the piece intends WDC and AI-driven SSD demand truly hits McKinsey's 35% CAGR through 2030, valuation compression could still deliver outsized returns despite the ticker error.
"NAND pricing cycles and capacity dynamics threaten durable margin gains, so the Sandisk-based upside is fragile."
Overlooked risk: NAND pricing and margins are highly cyclical, and capacity from Samsung, SK Hynix, and Micron could erode pricing power even if AI capex stays elevated. The article’s Sandisk narrative rests on a perpetual scarcity story (2030) that’s unlikely; a reversal could crush SNDK/WDC multiples far faster than Nvidia’s upside, making the 'millionaire-maker' thesis fragile. Even a modest upside for Nvidia risks leaving the memory-cap axis far behind.
"The shift toward AI inference will force a permanent re-rating of storage infrastructure providers, regardless of the article's ticker blunder."
Claude is right to highlight the 11x SSD content requirement, but everyone is missing the 'data gravity' shift. As AI models move from training to inference, the bottleneck moves from HBM (High Bandwidth Memory) to high-capacity NAND storage. While the Sandisk ticker error is fatal to the article's credibility, the underlying thesis on storage scarcity is structurally sound. We are looking at a permanent shift in storage architecture, not a temporary cyclical NAND boom.
"Inference workloads compress data, not expand it; Gemini's 'data gravity' argument lacks empirical grounding and doesn't rescue the Sandisk thesis."
Gemini's 'data gravity' pivot is clever but unsubstantiated. Training-to-inference shift is real, but inference workloads typically compress data footprints via quantization and pruning—they don't inherently demand 11x NAND expansion. The article never claimed this architectural shift; Gemini is retrofitting a narrative to salvage a defunct ticker. If inference truly drives storage demand, that thesis applies to WDC or Micron directly, not to a ghost company. The scarcity story still needs a 2030 timeline nobody can verify.
"Hyperscaler custom storage and inference compression together erode any durable NAND scarcity for actual producers like WDC."
Gemini's inference pivot overlooks that hyperscalers are already shifting toward custom ASICs and object storage tiers that slash raw NAND requirements by 40-60% versus legacy setups. Even if training-to-inference migration occurs, this vertical integration directly undercuts the scarcity premium for WDC or Micron. Claude's compression point compounds the issue: without verifiable 11x content growth materializing, the 2030 shortage narrative collapses faster than cyclical oversupply risks alone would imply.
The panel consensus is overwhelmingly bearish on the article's thesis that Sandisk (now a segment of Western Digital) offers superior growth and upside versus Nvidia amid Mag7 capex. Key risks include cyclical NAND pricing and margins, potential oversupply, and margin compression from competition. The 'millionaire-maker' storage play narrative is considered fragile and high-risk.
None identified as the panel was largely critical of the article's thesis
Cyclical NAND pricing and margins, potential oversupply, and margin compression from competition