Semiconductor ETFs Attract $46 Billion! Is the Next Surge in the AI Boom Coming?

According to Benzinga, while the semiconductor sector faces intensifying sell-offs, investors continue to pour money into the industry, pushing semiconductor ETFs toward a record-breaking year for capital inflows.

The financial market newsletter cites data from BofA Global Research and Bloomberg, indicating that semiconductor ETFs have attracted $46 billion in net inflows by 2026—about 31% of the category’s total assets under management (AUM) and more than double the cumulative inflows from 2017 to 2025. Since 2017, total inflows have reached a historic high of $68 billion, with $2.3 billion added just last week.

Despite the iShares Semiconductor ETF falling 20.3% from its June 2 peak—officially entering bear market territory—capital inflows continue to rise. The Roundhill Memory ETF, launched in April, surged to an all-time high of $80.72 on June 22 before retreating about 35%.

Record ETF inflows reflect investor belief that semiconductors remain the most critical investment channel for capturing the AI boom, driven by massive investments in AI data centers, advanced chips, and cloud infrastructure.

Top Beneficiaries

The VanEck Semiconductor ETF (SMH-US) remains the largest and most closely watched fund in the sector, primarily holding AI leaders such as NVIDIA (NVDA-US), TSMC (TSM-US) (2330-TW), and Broadcom (AVGO-US).

The iShares Semiconductor ETF offers broader exposure to U.S. semiconductor firms, while the SPDR S&P Semiconductor ETF (XSD-US) uses an equal-weight strategy to reduce reliance on mega-cap stocks.

In addition to the Roundhill Memory ETF, notable funds include the First Trust Nasdaq Semiconductor ETF (FTXL-US) and the Invesco Dynamic Semiconductors ETF (PSI-US).

While semiconductor ETFs continue to attract record inflows, technical indicators suggest the market is still awaiting clearer signals to confirm whether the recent correction has ended.

Among them, the largest fund, VanEck Semiconductor ETF, shows a relatively stronger technical pattern. Its Relative Strength Index (RSI) stands at 45.15—above the oversold zone but below the 70 overbought threshold—indicating neutral-to-slightly bullish momentum.

More notably, SMH’s MACD has crossed above the signal line, and the histogram has turned positive, preliminarily suggesting that selling pressure is easing. However, both MACD and the signal line remain below the zero axis, meaning a clear bullish trend has not yet been established.

The iShares Semiconductor ETF (SOXX-US) shows a similar but slightly weaker pattern. Its RSI is 39.64—momentum remains weak but not yet oversold. It also shows a MACD golden cross with a positive histogram, indicating that downward momentum may be weakening. However, both MACD and signal line remain in negative territory, suggesting the market is still in a bottoming phase, not yet launching a new upward rally.

For investors, this divergence is worth noting. With continued investment in AI infrastructure and strong earnings from chip leaders, capital flows and industry fundamentals remain optimistic.

However, technically, semiconductor ETFs are still in the early stages of technical recovery. While momentum is improving, the next upward leg has not yet been officially confirmed.

This implies that after a significant rally, the sector is now entering a consolidation phase, while investors continue to bet on the long-term growth potential of AI.

AI Investment Remains the Primary Driver

Unlike past semiconductor cycles driven by smartphones and PCs, this growth wave is primarily fueled by massive corporate investments in AI infrastructure.

Companies including NVIDIA, Broadcom, AMD, Micron, TSMC, and SK Hynix continue to benefit from strong demand for AI accelerators and memory chips.

Meanwhile, major cloud providers such as Microsoft, Amazon, Alphabet, and Meta have committed hundreds of billions of dollars to expand AI infrastructure.

Strong corporate earnings, combined with record ETF inflows, continue to bolster investor confidence in the semiconductor industry.

However, after historic gains, further price increases will depend less on valuation expansion and more on companies’ ability to deliver consistent revenue and profit growth.

For investors seeking exposure to AI’s long-term growth while diversifying individual stock risk, semiconductor ETFs remain one of the most popular investment tools.

Yet, with technical indicators suggesting valuations are becoming stretched and record inflows raising new questions, the market is asking: Are investors chasing short-term momentum, or are they positioning early for the next phase of the AI boom?