By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
Druckenmiller's Q1 purchases suggest a measured exposure to AI's supply chain rather than a wholesale shift, with tiny position sizes and no significant change in stance towards NVDA. The panel is neutral on his AI conviction, with concerns raised about execution hurdles, capacity constraints, and capex risks.
Risk: Capex risks if AI ROI stalls, and geopolitical risks such as export controls tightening on advanced nodes to China, which could hit TSM and AVGO harder than INTC.
Opportunity: Potential policy tailwinds for Intel's foundry bet, such as U.S. subsidies under the CHIPS Act, which could reduce execution risk relative to TSM's China exposure.
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 was Stanley Druckenmiller’s biggest holding just a few years ago.
The billionaire’s recent moves suggest he is still positive about the overall AI growth story.
It's that time of year again. Billionaire investors have offered us a glimpse into their latest trades -- in this case, during the first quarter of the year -- and we may use this precious information for investing inspiration. How does this happen? Managers of more than $100 million in securities must file their latest moves on a quarterly basis with the Securities and Exchange Commission, and these 13F filings are publicly available.
Today, I'm focusing on the latest moves of Stanley Druckenmiller, an investor with an impressive track record over time. At the helm of Duquesne Capital Management, he delivered an average annual return of 30% over three decades -- without losing money during any of those years. Druckenmiller has been retired for many years but today leads investing at the Duquesne Family Office.
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 »
And here, the billionaire recently made a few key moves concerning artificial intelligence (AI) stocks. Druckenmiller still isn't buying Nvidia (NASDAQ: NVDA), a stock he sold a couple of years ago and said he would consider returning to, but he just loaded up on three stocks driving the next stages of the AI boom. Let's check out this top investor's latest moves.
So, first, it's important to note Druckenmiller's past experience with Nvidia. The stock was the billionaire's top position for part of 2023 -- but he started cutting that position later in the year. A year later, Druckenmiller would go on to sell all of his Nvidia shares and express regret about doing so too soon. In an interview with Bloomberg in the fall of 2024, Druckenmiller said selling Nvidia when he did was one of the mistakes he's made during his investing career. And he said he would consider buying Nvidia shares again in the future at a lower valuation.
Nvidia's valuation has dropped since that time, but so far, Druckenmiller hasn't returned to the AI chip leader.
That said, the billionaire has opened positions in three other companies that are also playing key roles in this technology boom. In the first quarter of this year, he did the following:
Meanwhile, Taiwan Semiconductor Manufacturing remains one of his biggest holdings -- the third-largest to be exact -- and makes up nearly 5% of the portfolio.
Though Druckenmiller hasn't offered reasons behind his moves, all of this suggests that he remains optimistic about the next stages of the AI story. Broadcom is making great strides in the AI chip market with its custom AI chips delivering massive growth. While Nvidia offers a chip that's highly versatile across a broad range of uses, Broadcom differentiates itself by designing chips tailored to specific tasks.
Intel, which had fallen behind in the earlier stages of the AI boom, may find itself in a better position these days as AI workloads also need central processing units (CPUs), which is Intel's strength. The company has also partnered with Nvidia to bring Intel's CPUs to Nvidia's AI platforms and Nvidia GPU chiplets into Intel's systems for personal computing.
Finally, memory and storage giant Micron is seeing soaring demand as AI workloads are memory-intensive. In fact, its biggest challenge is keeping up with this demand. In the latest quarter, Micron delivered record revenue, gross margin, and earnings per share -- and expects more records in the current quarter.
So, should you seek inspiration from Druckenmiller -- and favor these stocks over Nvidia? This depends on your investment strategy. Over the past five years, these stocks have underperformed Nvidia, so if investors rotate out of the biggest winner, that shift could favor these other players. Aggressive investors may consider this route.
But it's important to keep in mind that, today, these stocks don't necessarily offer you a bargain from a valuation perspective. In fact, Nvidia looks like the best deal. Though Micron is cheaper, its valuation has increased, while Nvidia's has declined.
So value-conscious investors might favor Nvidia at the moment. Meanwhile, the good news is that all of these companies have shown growth in recent times, thanks to the AI boom, and they could continue to benefit well into the future. This means that by picking up any of these Druckenmiller buys -- or even Nvidia -- you might score an AI win down the road.
Before you buy stock in Nvidia, 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 Nvidia 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 $469,293! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,381,332!
Now, it’s worth noting Stock Advisor’s total average return is 993% — 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 18, 2026. *
Adria Cimino has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Broadcom, Intel, Micron Technology, Nvidia, and Taiwan Semiconductor Manufacturing. 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
"Druckenmiller's modest new stakes signal diversification inside AI infrastructure, not conviction that AVGO, MU, or INTC will displace Nvidia's lead."
Druckenmiller's Q1 purchases of 195k AVGO shares, 23k MU, and 411k INTC, alongside a still-large TSM position, point to measured exposure across AI's supply chain rather than a wholesale shift. These holdings remain tiny at 1.8%, 0.2%, and 0.5% respectively, while he has stayed out of NVDA despite its valuation reset. The article underplays Intel's multi-year execution hurdles and Micron's capacity constraints that could cap near-term upside even if AI memory demand stays elevated. Broadcom's custom ASIC traction is real but competes directly with Nvidia's software moat.
The article's own data shows Micron already posting record revenue and margins from AI workloads, so these names could compound faster than NVDA if hyperscalers accelerate custom silicon adoption and memory intensity rises.
"Druckenmiller's Q1 purchases are too small and under-explained to constitute a meaningful rotation thesis; they read as portfolio hedging, not conviction."
The article frames Druckenmiller's Q1 buys (AVGO, MU, INTC) as AI conviction plays, but the position sizes tell a different story. AVGO at 1.8%, MU at 0.2%, INTC at 0.5% are portfolio crumbs—collectively under 3% versus TSMC's 5%. This isn't a rotation; it's tactical nibbling. The article omits what he sold to fund these buys, his overall portfolio construction, and whether these are hedges rather than core convictions. Druckenmiller's regret over NVDA exit doesn't validate these new positions—it suggests he's chasing optionality in a crowded trade rather than making a high-conviction bet.
Druckenmiller's 30-year track record earns him the benefit of the doubt; even small positions from him can signal asymmetric risk-reward that the market hasn't priced. The article's framing of 'next stages' may actually be prescient if GPU saturation forces AI capex downstream to memory, networking, and CPUs.
"Druckenmiller is de-risking his AI exposure by moving from the primary chip designer to the essential hardware suppliers and memory providers that face less valuation multiple compression risk."
Druckenmiller’s rotation into Broadcom, Micron, and Intel reflects a shift from 'AI hype' to 'AI infrastructure utility.' By buying AVGO and MU, he is betting on the plumbing of the AI stack—custom ASICs and High Bandwidth Memory (HBM)—which are essential bottlenecks as compute demand scales. Intel remains the outlier; at 0.5% of his portfolio, it’s a speculative 'value' play on the foundry business turnaround. However, the article ignores the massive capital expenditure risks. If hyperscalers like Microsoft or Meta hit a wall in ROI for their AI deployments, the demand for this secondary tier of hardware will evaporate faster than Nvidia’s GPU demand, which benefits from deeper software moats.
Druckenmiller’s small position sizes in these names suggest he is hedging against a broader market drawdown rather than making a high-conviction bet on these specific companies’ fundamentals.
"Druckenmiller’s Q1 buys signal tactical diversification away from Nvidia, but Nvidia’s AI moat could still render this a lagging, catch-up trade rather than a leadership shift."
The article portrays Druckenmiller as still bullish on AI but not back in Nvidia, instead loading into Broadcom (AVGO), Micron (MU), and Intel (INTC) in Q1, while Taiwan Semiconductor remains a large, core exposure. This suggests a belief in AI demand but a hedge against Nvidia’s monopoly by diversifying into more targeted AI-chip suppliers and memory/CPU play. The missing context: 13F data reflects a single quarter and tiny weights, not a durable thesis; it also glosses over Nvidia’s likely ongoing pricing power and the distinct risk profiles of AVGO, MU, and INTC (chip design vs memory vs CPUs). Macro cycles, regulation, and geopolitical risk could upend these bets even if AI spend remains robust.
Strongest counter: 13F moves may be opportunistic rebalancing rather than a true shift in AI leadership; Nvidia could maintain its moat and keep outperforming while these peers catch a later phase-up, not a trend reversal.
"Export controls create correlated downside across TSM and AVGO that applies equally if AI ROI stalls."
Gemini correctly flags capex risks if AI ROI stalls, yet this applies equally to TSM's 5% core holding that nobody has stress-tested. Export controls tightening on advanced nodes to China would hit TSM and AVGO harder than INTC due to their reliance on leading-edge processes, creating correlated downside the 13F filings alone cannot reveal.
"INTC's tiny position size may reflect optionality on geopolitical de-risking, not weakness."
Grok's geopolitical angle on TSM/AVGO export controls is sharp, but misses that INTC's foundry bet actually *benefits* from tightening—domestic capacity becomes strategic asset. Druckenmiller's 0.5% INTC position could be a leveraged bet on policy tailwinds, not just a value play. This flips the risk profile nobody mentioned: if U.S. subsidies accelerate (CHIPS Act), INTC's execution risk shrinks relative to TSM's China exposure.
"Intel's execution risk and cash burn outweigh the potential benefits of domestic policy tailwinds."
Claude, your focus on Intel as a policy-leveraged play ignores the brutal reality of its cash burn. Even with CHIPS Act tailwinds, Intel’s foundry is a multi-year capital sink that dilutes shareholders long before it achieves parity with TSMC. Druckenmiller is likely treating INTC as a distressed option, not a strategic asset. If the AI capex cycle cools, domestic subsidies won't save a company struggling with yield and process node execution against superior competitors.
"Intel's upside depends on near-term ROI and execution, not subsidies; the 0.5% stake reads as hedging, not conviction."
Claude, you argue policy tailwinds could lift Intel, but that assumes subsidies translate into rapid, capital-efficient wins rather than a multi-year, capital-intensive slog. The immediate friction is execution risk and the ROI quality of foundry demand—not just subsidies. A 0.5% position signals hedging, not conviction, and if AI capex stalls, incremental foundry revenue may never justify the burn. INTC's upside hinges on near-term ROI, not policy headlines.
Druckenmiller's Q1 purchases suggest a measured exposure to AI's supply chain rather than a wholesale shift, with tiny position sizes and no significant change in stance towards NVDA. The panel is neutral on his AI conviction, with concerns raised about execution hurdles, capacity constraints, and capex risks.
Potential policy tailwinds for Intel's foundry bet, such as U.S. subsidies under the CHIPS Act, which could reduce execution risk relative to TSM's China exposure.
Capex risks if AI ROI stalls, and geopolitical risks such as export controls tightening on advanced nodes to China, which could hit TSM and AVGO harder than INTC.