By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel discusses Druckenmiller's rotation from Alphabet to Sandisk, with most agreeing that while Alphabet's growth may be priced in, Sandisk's extreme valuation and potential margin compression pose significant risks.
Risk: Margin compression and supply normalization for Sandisk
Opportunity: Potential long-term growth in AI for Alphabet
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 →
Though we're entering the heart of earnings season, don't overlook how valuable the filing of Form 13Fs can be for investors. These quarterly filings allow investors to track which stocks Wall Street's smartest and most successful money managers have been buying and selling.
Few, if any, billionaire investors have proven more successful over multiple decades than Duquesne Family Office's Stanley Druckenmiller. Known for his relatively active trading style and his love for innovative tech stocks, Druckenmiller dumped shares of Google parent Alphabet (NASDAQ: GOOGL)(NASDAQ: GOOG) in the first quarter, and piled into the hottest member of the S&P 500 (SNPINDEX: ^GSPC): memory titan Sandisk (NASDAQ: SNDK).
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 »
According to Duquesne Family Office's mid-May-filed 13F, billionaire Stanley Druckenmiller sent all 385,000 shares of Alphabet (the Class A shares, GOOGL) to the chopping block.
Profit-taking is certainly a logical explanation for Duquesne's boss to ring the register. Alphabet stock practically doubled over the two quarters (the third and fourth quarters of 2025) during which Druckenmiller was a buyer. Alphabet's integration of artificial intelligence (AI) solutions into Google Cloud has helped reaccelerate sales growth for this high-margin platform and sent its stock catapulting higher.
"Google Cloud revenues grew 63% with backlog nearly doubling quarter on quarter to over $460 billion."
-- Qualtrim (@qualtrim) April 29, 2026
Analysts Projection: +52% YoY
Google Results:
Cloud Revenue: +63% YoY
Cloud Backlog: +300% YoY$GOOGL $GOOG pic.twitter.com/zNkiP1vcd1
But this may not be the complete story.
Druckenmiller has also gone on record as stating that "AI may be a little overhyped now, but underhyped long term." He likely recognizes that every hyped technological innovation over the last three decades has succumbed to a bubble-bursting event early in its expansion. Though Alphabet is well-diversified, with a virtual monopoly in internet search, it wouldn't be immune if the AI bubble bursts.
Alphabet also isn't the screaming bargain it once was. Its forward price-to-earnings (P/E) ratio of 24 represents a 15% premium to its average forward P/E over the trailing five years.
While Alphabet was given the boot, Duquesne's billionaire chief purchased 38,500 shares of memory storage solutions provider Sandisk. Over the trailing year (as of July 16), Sandisk shares have rallied more than 3,200%, making it Wall Street's top-performing S&P 500 component by more than 2,500 percentage points!
Druckenmiller's optimism likely stems from Sandisk's critical role in AI data centers. The company's NAND flash memory, solid-state drives, and high-bandwidth flash are designed to support high data transfer rates in AI-accelerated data centers.
2027 EPS estimates for SanDisk $SNDK are about to bend backwards in time while still moving parabolically higher: pic.twitter.com/qRsPDqR5KA
-- Consensus Media (@ConsensusGurus) May 3, 2026
The "issue" is that enterprise demand for these solutions far exceeds their supply. This chokepoint is stymying an even faster data center build-out, but also affording Sandisk a level of pricing power and gross margin it's never seen before.
In September 2025, Wall Street was forecasting less than $10 in full-year earnings per share (EPS) for Sandisk in 2027. Today, this EPS forecast is approaching $208. If memory supply shortages persist, Sandisk (and Stanley Druckenmiller) will be sitting pretty.
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 $371,842! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,244,783!
Now, it’s worth noting Stock Advisor’s total average return is 900% — 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 July 20, 2026. *
Sean Williams has positions in Alphabet. The Motley Fool has positions in and recommends Alphabet. 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
"Sandisk's 3,200% rally and sky-high 2027 EPS forecasts reflect temporary supply shortages more than sustainable pricing power, making Druckenmiller's pivot riskier than the article portrays."
Druckenmiller's exit from Alphabet (GOOGL) after a near-doubling and entry into Sandisk (SNDK) looks like classic rotation from a now 24x forward P/E name trading at a 15% premium to its 5-yr average into the S&P 500's hottest performer (+3,200% in a year). The article correctly flags AI tailwinds for both but underplays SNDK's extreme valuation: 2027 EPS forecasts have exploded from <$10 to ~$208 on supply shortages that are unlikely to persist. Memory cycles are notoriously volatile; NAND pricing power can evaporate quickly once new capacity comes online. Alphabet's Google Cloud 63% growth and $460B backlog remain compelling long-term AI proxies at a more reasonable multiple than SNDK's parabolic run.
The strongest case against is that persistent AI data-center chokepoints could keep NAND/SSD margins elevated far longer than history suggests, allowing SNDK's EPS to keep compounding and justifying its massive re-rating while Alphabet faces regulatory and search-margin pressure.
"Druckenmiller is trading Alphabet's long-term platform stability for the high-beta, cyclical volatility of the AI hardware supply chain."
Druckenmiller’s rotation from Alphabet to Sandisk reflects a classic shift from 'platform' exposure to 'infrastructure bottleneck' plays. While Alphabet remains a dominant AI player, its valuation has arguably reached a point where the growth-to-multiple expansion trade is exhausted. Conversely, the move into Sandisk is a pure-play bet on the supply-demand imbalance in high-bandwidth memory (HBM) and NAND flash. However, investors should be wary: the article cites a 2027 EPS forecast of $208 for Sandisk, which implies a parabolic, potentially unsustainable margin expansion. This suggests the market is pricing in a permanent supply shortage, which rarely lasts as capacity catches up.
The thesis assumes Sandisk maintains its pricing power, but memory is historically a commoditized, cyclical industry that inevitably suffers from massive oversupply once capital expenditure cycles peak.
"A 20x EPS forecast revision in months driven by supply constraints, not demand growth, is a crowded trade waiting for a margin reset, not a generational opportunity."
This article conflates two separate signals into one narrative. Yes, Druckenmiller sold GOOGL after a 100% run—textbook profit-taking on a 24x forward P/E (15% premium to 5-yr avg). But the SNDK purchase is the real story, and it's also the riskiest. SNDK's 2027 EPS forecast jumped from <$10 to ~$208 in months—a 20x revision driven entirely by supply constraints, not demand fundamentals. That's not a signal of genius; that's a bet on a bottleneck persisting. Bottlenecks break. When NAND supply normalizes (and it will), SNDK's margin expansion evaporates faster than it appeared.
Druckenmiller's track record deserves deference—he may see structural NAND scarcity lasting longer than consensus expects, or he may have conviction that AI capex will sustain pricing power even post-normalization. The article doesn't provide his actual thesis.
"13F moves are not predictive; Sandisk’s upside hinges on a durable AI memory upcycle that may not materialize if NAND pricing or AI capex slows."
The article leans on a mid‑May 13F to imply a conviction trade from Alphabet to Sandisk, but 13Fs are backward-looking and can reflect timing, tax, or window-dressing rather than durable theses. The Sandisk bull case rests on an apparently tight AI data-center memory cycle, which is highly cyclical and prone to pricing/margin compression if supply catches up. Alphabet, by contrast, remains a diversified AI leader with multiple engines (Search, Cloud, YouTube) that historically sustains margins even as AI hype fluctuates. The piece also seems to conflate branding with standalone AI hardware exposure, risking overstatement of Sandisk’s AI leverage.
The rotation could reflect a legitimate, longer‑term AI data‑center upcycle for memory suppliers; if AI capex accelerates and NAND pricing stabilizes, Sandisk could outperform even as Alphabet stays solid.
"AI server unit growth and rising memory content per rack could extend Sandisk's margin tail beyond historical cycle norms, while Alphabet faces distinct regulatory pressure the panel underweights."
Gemini and Claude both treat the 2027 EPS jump to $208 as purely unsustainable supply-driven hype. Yet Druckenmiller's move coincides with AI server shipments still tracking +80% YoY into 2026; if HBM/NAND content per rack keeps rising, pricing power could endure longer than the classic memory-cycle playbook. The real unmentioned risk is regulatory decapitation at Alphabet, not Sandisk mean-reversion.
"The cyclical nature of memory manufacturing poses a greater risk to Sandisk's valuation than regulatory headwinds do to Alphabet's diversified business model."
Grok, you are ignoring the capital intensity of the memory cycle. Even if AI server shipments remain elevated, Sandisk’s margin profile is inherently fragile compared to Alphabet’s software-defined moat. You cite regulatory risk, but that is a known quantity for Alphabet, whereas the risk of a supply-side glut in NAND is an existential threat to Sandisk's current valuation. Betting on 'this time is different' for memory pricing is a classic trap for institutional investors.
"Alphabet's Search moat is eroding faster than the article acknowledges, making the rotation less about SNDK hype and more about exiting a deteriorating core business."
Gemini's 'software moat' framing misses that Alphabet's Search margin is already under pressure from AI cannibalization—Claude's own search results are eating into Google's query volume. SNDK's cyclicality is real, but so is Alphabet's structural headwind. Neither is a 'known quantity.' The real question: which margin compression happens first? If Search deteriorates faster than NAND normalizes, Druckenmiller's rotation timing looks prescient, not reckless.
"Sandisk's 2027 EPS of $208 is highly contingent on an unsustainable bottleneck; expect margin reversion as capacity adds, limiting upside."
Responding to Grok: The 'bottleneck lasts' premise is the weakest link. Even with AI-driven capex, NAND supply shines a light on the capacity sprint: fabs come online on a 12–24 month cycle, and price/margins compress quickly once supply catches up. The $208 2027 EPS claim for SNDK hinges on an enduring shortage; the more likely path is reversion to mean margins and capex discipline. Therefore, Sandisk's upside is more limited than the headline suggests.
The panel discusses Druckenmiller's rotation from Alphabet to Sandisk, with most agreeing that while Alphabet's growth may be priced in, Sandisk's extreme valuation and potential margin compression pose significant risks.
Potential long-term growth in AI for Alphabet
Margin compression and supply normalization for Sandisk