US tech stocks stumble. Is the AI rally losing momentum?

• US tech stocks sold off over two days, with hyperscalers leading Monday's decline and semiconductor stocks driving Tuesday's weakness.

• The selloff was likely triggered by rising rate expectations, AI CAPEX concerns, and a sharp correction in Asian semiconductor stocks, which spilled over into global chip markets.

• We view the pullback as a healthy correction rather than a deterioration in fundamentals, with strong AI demand, robust earnings growth, and continued hyperscaler spending supporting both semiconductor and internet companies.

• Investors should stay invested and consider buying the dip or dollar-cost averaging. We remain neutral on US semiconductors, prefer Asian semiconductors due to more attractive valuations, and favour financially strong hyperscalers such as Alphabet, Microsoft, and Amazon.

US tech stocks suffered a sharp pullback on Tuesday, 23 June 2026, with the tech-heavy Nasdaq-100 falling 3.3%. Semiconductor stocks were hit particularly hard, with the MVIS US Listed Semiconductor 25 Index plunging 7.1%. Micron Technology led the decline, falling 13.2%, while NVIDIA and Broadcom lost 4.2% and 3.1%, respectively. The broader technology sector had already come under pressure on Monday, with Alphabet dropping 5.0%, Amazon falling 4.7%, Microsoft declining 3.2%, and Meta Platforms losing 2.3%.

What triggered the recent selloff in tech stocks?

1.Rising interest rates concern

One of the potential catalysts behind the selloff was a research note from Bank of America released on Monday, which suggested the possibility of up to three Federal Reserve rate hikes this year. The report added to existing investor concerns that persistent inflation could force the Fed to keep monetary policy tighter for longer than previously expected.

The Federal Reserve signalled last week that higher borrowing costs in 2026 remain a possibility as policymakers seek to contain inflationary pressures stemming from months of elevated oil prices following the Iran conflict. Reflecting this shift in sentiment, CME FedWatch data shows that the probability of at least one rate hike by year-end has surged to 85.3%, up from 59.5% just a week earlier.

Higher interest rates tend to weigh disproportionately on technology stocks. Not only do they increase financing costs for companies pursuing aggressive growth initiatives, but they also reduce the present value of future earnings, a key component of valuations for high-growth businesses whose profits are expected further into the future.

This dynamic has become increasingly relevant as investors scrutinise the enormous capital commitments being made to develop AI infrastructure. Technology companies are investing unprecedented amounts in data centres, advanced semiconductors, networking equipment, and power infrastructure to support the next phase of AI adoption. While the long-term growth opportunity remains compelling, investors are increasingly questioning the timing and magnitude of the returns these investments will ultimately generate.

Adding to these concerns, SpaceX's bond issuance on Monday may have fuelled speculation that other large technology companies could increasingly turn to debt markets to finance their growing AI-related capital expenditure requirements. While we believe this may have contributed to negative sentiment, it is worth noting that SpaceX stated the proceeds would primarily be used to repay borrowings under its existing bridge loan facility rather than fund new AI infrastructure investments.

2. Asia-led chip selloff sparks global semiconductor retreat

The weakness in US chip stocks on Tuesday could also be due to developments in Asia.

South Korea's KOSPI Index plunged 10%, after semiconductor shares sold off amid concerns that the sector's rally had become overheated. This came after Lee Chan-jin, ‌governor of the Financial Supervisory Service said the regulator ​had been too hasty in approving leveraged funds tied to some of the country's best-known chip stocks like Samsung Electronic and SK Hynix, and are cautiously monitoring and seriously looking into introducing stabilising measures. Industry heavyweights Samsung Electronics and SK Hynix each fell more than 10%, triggering an automatic trading halt and erasing billions of dollars in market value.

Elevated retail leverage may also have contributed to the sharp selloff. Margin-related liquidations and selling pressure from leveraged exchange-traded products linked to Samsung and SK Hynix could have intensified the downward move.

Another possible contributing factor was local media reports suggesting that SK Hynix may slow the expansion of certain AI memory chip production plans while placing greater emphasis on conventional DRAM products. The reports sparked concerns about whether demand growth from AI data centres could eventually moderate.

The negative sentiment quickly spilled over into US markets. Semiconductor stocks have been among the strongest beneficiaries of the AI boom, making them particularly vulnerable to any shift in investor sentiment.

Positioning also likely played an important role. Semiconductor stocks had become one of the most crowded trades globally, supported by strong earnings momentum, aggressive AI spending, and substantial investor inflows. When positioning becomes crowded, even modest disappointments or uncertainties can trigger outsized price movements as investors rush to reduce exposure simultaneously. In such environments, profit-taking can quickly snowball into broader market weakness.

AI fundamentals remain strong despite recent selloff

Despite the sharp correction, we view the recent pullback as a healthy consolidation following an exceptional rally rather than the beginning of a structural downturn.

Prior to Tuesday's decline, our recommended semiconductor ETF, VanEck Semiconductor ETF (SMH), had risen approximately 86.0% year-to-date. Given the magnitude of those gains, some degree of profit-taking was both natural and arguably necessary.

Micron's quarterly results, due tonight, will provide important clues regarding AI-related demand trends. However, the broader industry backdrop remains supportive. Earlier this year, Micron reported revenue growth from USD 8.1 billion in 1Q25 to USD 23.9 billion in 1Q26, driven by robust demand for AI memory products. High-bandwidth memory (HBM) demand continued to exceed supply, supporting favourable pricing conditions. Management also guided for 2Q26 revenue of USD 33.5 billion and indicated that tight supply conditions across both DRAM and NAND markets could persist beyond 2026.

Similarly, reports that SK Hynix may adjust its product mix should not necessarily be interpreted as evidence of weakening AI demand. Rather, such decisions may reflect management's efforts to optimise profitability across different memory categories as market conditions evolve.

More importantly, earnings growth across AI-exposed semiconductor companies should continue to be supported by sustained investment spending. Capital expenditure on AI infrastructure remains exceptionally strong. Collectively, hyperscalers including Amazon, Meta Platforms, Microsoft, and Alphabet are expected to spend approximately USD 725 billion on capital expenditures this year, representing a 77% increase from 2025 levels.

This spending should continue to support demand throughout the semiconductor ecosystem, from AI accelerators and memory chips to networking and data-centre infrastructure.

That said, while we remain constructive on the long-term outlook of the semiconductor industry, valuations had already reflected much of the near-term optimism. As a result, we believe a neutral stance on US semiconductor equities remains appropriate.

Valuations also appear more attractive in Asia. The MVIS US Listed Semiconductor 25 Index currently trades at approximately 33.7x forward 2026 earnings, compared to just 16.2x for the FactSet Asia Semiconductor Index. This valuation gap provides a larger margin of safety for investors seeking exposure to the AI theme.

Table 1: Projections for the MVIS US Listed Semiconductor 25 Index

MVSMHTR Index	2025	2026E	2027E	2028E
Earnings Per Share (EPS)	446.8	755.0	980.0	1220.0
Earnings Growth YoY	30.9%	69.0%	29.8%	24.5%
PE Ratio (X)	33.0	33.7	26.0	20.9
Target Price (based on a fair PE of 25X)				29,280
Upside Potential				15.1%
Target Price for ETF (USD)				717
Source: Bloomberg Finance L.P., iFAST estimates. Data as of 23 June 2026

At the same time, we continue to see attractive earnings growth prospects among internet and digital platform companies. Hyperscalers remain particularly well positioned, as rising AI adoption drives greater token consumption while declining costs per token improve operating efficiency. This combination could support margin expansion despite elevated capital expenditure requirements.

We also expect cloud growth to accelerate further in the second half of 2026, particularly for AWS and Google Cloud, as additional computing capacity comes online and enterprise AI adoption continues to broaden.

Another important trend is the increasing use of internally developed chips. Companies such as Amazon, Microsoft, and Alphabet are deploying custom-designed processors that improve efficiency and reduce reliance on third-party suppliers. These firms are also increasingly offering cloud services powered by their proprietary chips, creating additional revenue streams while enhancing competitiveness.

Nevertheless, capital expenditure remains a key overhang for the sector. Investors continue to question whether future returns will ultimately justify the enormous investment commitments being made today. We also expect increased debt issuance across the technology sector in the second half of 2026 as companies seek to finance growing infrastructure requirements. As such, selectivity remains important. We favour hyperscalers such as Alphabet, Microsoft, and Amazon, whose strong balance sheets provide greater flexibility to fund AI-related investments and absorb higher borrowing costs. In contrast, we are more cautious on companies with weaker financial profiles, such as Oracle.

In recognition of this structural shift towards more capital-intensive business models, we previously lowered our fair price-to-earnings multiple for the digital economy sector to reflect both higher infrastructure spending requirements and our base-case expectation of higher-for-longer inflation. Even after this adjustment, however, valuations remain attractive. We estimate an upside potential of approximately 54.5% from closing levels on 23 June 2026, supporting our positive outlook on the digital economy sector.

Table 2: Projections for the Nasdaq CTA Internet Total Return Index

NETX Index	2025	2026E	2027E	2028E
Earnings Per Share (EPS)	61.4	67.7	77.3	87.2
Earnings Growth YoY	26.5%	10.2%	14.1%	12.9%
PE Ratio (X)	27.6	20.8	18.3	16.2
Target Price (based on a fair PE of 25X)				2180
Upside Potential				54.5%
Target Price for ETF (USD)				69
Source: Bloomberg Finance L.P., iFAST estimates. Data as of 23 June 2026

Investment implications

The recent selloff has understandably unsettled investors. However, the decline appears to have been driven more by shifting sentiment than by any meaningful deterioration in the underlying AI investment thesis. While higher interest rates and elevated capital expenditure commitments warrant close monitoring, the long-term drivers supporting AI adoption remain firmly intact.

Periods of heightened market volatility can be uncomfortable, but they also create opportunities for long-term investors. Rather than reacting to short-term sentiment swings, investors should remain focused on the fundamental drivers of value creation, including earnings growth, technological innovation, and the continued adoption of AI across industries.

For investors with a long-term horizon, the recent pullback may present an opportunity to gradually add exposure to high-quality technology companies at more attractive valuations. Alternatively, investors can stay disciplined by continuing their regular investment or dollar-cost averaging plans, avoiding the temptation to time short-term market movements.

Table 3: Recommended products

Sector/Style	Recommended Products
Digital Economy	•       Fidelity Global Technology A-ACC-USD •       Eastspring Investments Unit Trusts – Global Technology SGD •       Invesco NASDAQ Internet ETF (NASDAQ: PNQI) •       VanEck Semiconductor ETF (NASDAQ: SMH) •       Global X Asia Semiconductor ETF (HKEX:3119)

Declaration:

This research report was prepared with the assistance of artificial intelligence (AI) tools. iFAST Financial Pte Ltd does not rely exclusively on AI for content generation; the content of this report – including all investment theses, ratings, price targets and conclusions – has been independently reviewed and verified by the research analyst(s) to ensure accuracy and professional integrity.

For specific disclosure, at the time of publication of this report, IFPL (via its connected and associated entities) and the analyst who produced this report hold a NIL position in the abovementioned securities.