Billionaire investor Stanley Druckenmiller has exited his positions in Broadcom, Intel, and Micron while maintaining a single AI chip stock position. According to The Globe and Mail, Druckenmiller took a sizable position in Nvidia shortly after ChatGPT's release, though he sold out of his Nvidia position in 2024. The investment manager's top-down approach typically involves buying groups of stocks poised to benefit from trends such as artificial intelligence, having previously owned stakes in Nvidia, Broadcom, Micron, and Intel at various points during recent years. Only one big AI chipmaker remains in his portfolio, though he has kept stakes in a few smaller ancillary AI-related semiconductor companies. This portfolio shift reflects Druckenmiller's confidence in the AI infrastructure value chain, as he opened his current AI chip position during the same period he sold his other major semiconductor stocks.
U.S. stocks reopened after the long holiday with mixed performance, but the semiconductor sector showed remarkable resilience. According to BiyaPay market data, the Dow Jones fell 0.77%, the S&P 500 dropped 0.48%, and the Nasdaq 100 declined 0.29% on September 9, yet Intel surged approximately 9%, AMD rose nearly 6%, and Qualcomm, Broadcom, and ASML posted varying gains. In contrast, NVIDIA retreated about 0.91%, while Micron and SanDisk, among other storage-focused companies, underperformed. This divergence reflects a structural revaluation rather than an emotional rebound, as capital seeks technology assets that can better prove their value in a high-interest-rate environment. The market isn't ignoring interest rates—it's seeking technology assets that can effectively connect customers, orders, pricing, and profit margins within the AI infrastructure value chain.
AMD shares have risen 7.7% in the past month, outperforming the Zacks Computer and Technology sector's return of 0.7%. The company's Data Center revenues jumped 107% year over year to $6.7 billion in Q2 2026, driven by strong EPYC server CPU demand and continued ramp of Instinct GPUs. AMD expects the Data Center business to more than double in 2027, with server CPU revenues projected to increase more than 80% year over year in the second half of 2026 and more than 70% in 2027. The company acknowledged that inference is overtaking training as a major source of AI compute demand, while agentic AI is creating incremental demand for both GPUs and CPUs. However, AMD's Data Center AI and MI450 gross margins remain somewhat below the company average, with the mix shift toward AI accelerators potentially constraining percentage-margin expansion even while gross profit dollars and earnings increase substantially.
Intel shares are down 6% to $99.67 in early Thursday trading, giving back part of a parabolic run that had lifted the stock 188% year-to-date through Wednesday's close. According to reports, the decline comes without a fresh company-specific catalyst, with the selloff appearing mechanical as investors rotate out of the name that had run up sharply. The iShares Semiconductor ETF (SOXX) is down 3%, sitting below the sector's late-summer highs, while the Invesco QQQ Trust (QQQ) is off 1.14%, indicating chips are falling harder than large-cap technology stocks generally. The move reflects profit-taking after the stock's dramatic surge, with the sell-side coverage remaining at 32 Holds against 13 Buys and an average target of $115.88 that still sits above current trading levels. However, Intel closed at $106.24, up another 1.69% following a sharp 9.05% surge the previous day, indicating the market is at least temporarily willing to absorb shares at higher prices.
AI chip stocks faced pressure as oil prices rose above $100 per barrel for the first time since July, adding to inflation concerns and raising worries about Federal Reserve policy. According to reports, Brent crude traded near $101.50 per barrel while WTI crude was near $95.26 per barrel, with the move intensifying tensions between the U.S. and Iran. The rise in energy prices could keep inflation elevated, making it harder for the Fed to cut rates and potentially leading to another rate hike. The 10-year Treasury yield has climbed to around 4.84%, a near three-year high that compressed chip valuations and contributed to the current selloff. Higher yields can reduce the appeal of earnings expected further in the future, particularly affecting companies whose valuations depend on expectations for strong AI and data-center growth.
The latest market conditions most clearly demonstrate that AI trading is shifting from a single龙头 narrative to a more granular industry chain pricing model. In the early stage, the market favored companies with the highest certainty, making NVIDIA, TSMC, memory chips, and AI servers the main themes. However, as valuations rise, capital begins seeking new narrative frameworks—for example, Intel is trading on price and earnings recovery, Qualcomm on AI inference and new entry points in data centers, AMD on alternative elasticity beyond NVIDIA, and ASML and semiconductor equipment on the expansion cycle of advanced process technologies. This explains why semiconductors can rise even when the broader market is falling. The market isn't ignoring interest rates—it's seeking technology assets that can better prove their value in a high-interest-rate environment. As long as AI capital expenditures aren't disproven, chip stocks won't simply exit the stage; they'll merely rotate among different segments. The following variables are crucial: first is the U.S. August CPI release on September 11, with stronger-than-expected inflation data potentially increasing Fed pressure ahead of its September meeting. Second is the FOMC meeting on September 15–16, with the market currently pricing in a high likelihood of a rate hike. Third is the follow-up orders in the AI supply chain—collaborations between Qualcomm and Amazon, expectations of Intel price increases, and AMD's server demand—all require further validation in upcoming earnings reports and guidance.
Despite the recent decline, analysts maintain strong conviction in the AI chip sector. According to the TipRanks Stock Comparison Tool, Micron has the highest analyst upside at 55.43%, based on an average price target of $1,554.67. Nvidia follows with 44.08% upside and an average price target of $325.23, while AMD has 27.97% upside with an average target of $647.19. Intel has 11.33% upside with an average price target of $116.31. The Strong Buy consensus ratings for Micron, AMD, and Nvidia suggest limited downside risk despite current market volatility, though the analyst community's skepticism toward Intel reflects concerns about its foundry business and competitive positioning. Companies that will continue to outperform in the future may not be the ones with the most flashy narratives, but rather those that can effectively connect customers, orders, pricing, and profit margins within the AI infrastructure value chain.
