Here's How Micron Technology, Nvidia, and AMD Could Help This Unstoppable ETF Turn $250,000 Into $1 Million in 10 Years
Summarize this article with:
Most artificial intelligence (AI) development happens inside large data centers, where vast numbers of advanced chips and networking components process mountains of information at a lightning-fast pace to train models. Nvidia (NVDA 0.33%), Advanced Micro Devices (AMD +1.23%), and Micron Technology (MU 2.28%) are three of the most important suppliers of AI data center hardware. There's more demand for their wares than they can supply now, a condition that has been driving their revenues and their stock prices higher. Those companies are three of the five largest holdings in the iShares Semiconductor ETF (SOXX +1.32%), which is an exchange-traded fund (ETF) that invests in 30 of the chip industry's top companies. It has delivered blistering returns over the last 10 years. Here's how it could turn an investment of $250,000 into $1 million over the next 10 years. Image source: Getty Images. Hardware is at the center of the AI boom The iShares Semiconductor ETF exclusively invests in U.S.-listed companies that design, manufacture, and distribute chips and tech components, and many of those stand to benefit from the ongoing AI revolution. Graphics processing units (GPUs) are the primary parallel processors used to power AI development. Nvidia's Blackwell GPUs have led the industry in processing power, but the company's next-generation Vera Rubin architecture, which will ship to customers in the second half of 2026, will deliver another leap forward in performance. AMD is one of Nvidia's closest competitors, and although its chips aren't as popular, it is slowly gaining market share. Micron is one of the world's top suppliers of high-bandwidth memory (HBM), which needs to sit alongside GPUs in data centers to facilitate the rapid flow of information. Without access to enough rapidly accessible memory, GPUs handling data-heavy analysis workloads would have to repeatedly pause while waiting to receive more data. This would create a poor experience for anyone using AI chatbots or agents, which is why Micron's memory products are being deployed alongside both Nvidia and AMD chips. Nvidia, AMD, and Micron have delivered astonishing returns for their shareholders over the last 10 years. Micron was the worst performer of the three, yet investors who bought its stock a decade ago still would have booked a gain of 3,570% -- multiplying their original investment by almost 37 times. (And that doesn't even factor in the stock's dividend payouts, which admittedly, were fairly small.) NVDA data by YCharts. The three stocks have a combined weighting of 23.2% in the iShares Semiconductor ETF: Stock iShares ETF Portfolio Weighting 1. Micron Technology 8.7% 2. Nvidia 8.2% 3. Broadcom 8.2% 4.
Advanced Micro Devices 6.3% 5. Applied Materials 6.0% Data source: iShares. Portfolio weightings are accurate as of March 20, 2026, and are subject to change. Turning $250,000 into $1 million over the next decade The iShares Semiconductor ETF has generated a compound annual return of 12.8% since its inception in 2001 and an accelerated annualized return of 30.1% over the last 10 years. Semiconductor demand from companies in the cloud computing and AI industries has fueled a lot of its recent growth. Here's how long it could take for the iShares ETF to turn an investment of $250,000 into $1 million, based on three different average annual returns: Starting Balance Compound Annual Return Time to Reach $1 Million $250,000 12.8% 12 Years $250,000 21.4% (midpoint) 8 Years $250,000 30.1% 6 Years Calculations by author. ExpandNASDAQ: SOXXiShares Trust - iShares Semiconductor ETFToday's Change(1.32%) $4.43Current Price$341.02Key Data PointsDay's Range$333.62 - $343.2252wk Range$148.31 - $368.82Volume79K It's unrealistic to expect any ETF to produce returns of over 30% per year indefinitely. Semiconductor and infrastructure spending is already at record levels because of AI, and even some of the industry's biggest customers are quickly burning through their cash flows and turning to debt financing to fund more hardware purchases -- which simply isn't sustainable over the long term. With that said, Nvidia CEO Jensen Huang believes there is so much demand for computing capacity from AI developers that data center operators could be spending up to $4 trillion per year on infrastructure by 2030. If he's right, then it's certainly possible for the iShares ETF to continue delivering above-average returns for the foreseeable future. The iShares Semiconductor ETF can still turn $250,000 into $1 million in under a decade if its average annual return falls to 21.4%. And it can help investors join the million-dollar club in around 12 years, even if its yearly return reverts to the long-term average of 12.8% -- but I think that type of deceleration is unlikely, because even after the AI infrastructure spending boom inevitably winds down, emerging industries like autonomous vehicles, robotics, and quantum computing could fuel a fresh wave of semiconductor spending.