M7 Lags Market by 13 Percentage Points Since Start of Year

Rise of AI Superstars, Semiconductor, and Space Competitors

Performance Remains Solid, But Rankings Slip

"Re-rating Will Come...A Time for Patience"

For Korean retail investors focused on U.S. stocks, big tech companies have long been considered "safe assets." There was a firm belief that buying on dips and holding would be rewarded with gains. This year, however, that belief is being shaken. While big tech still holds its status as the dominant force in the AI era, the so-called Magnificent 7 (M7: Nvidia, Apple, Microsoft, Alphabet, Amazon, Meta, and Tesla) have underperformed the broader market by 13 percentage points since the beginning of the year. Considering that over the past decade, the M7 outperformed the market by an average of 23 percentage points, the atmosphere has changed dramatically.


"I Bought the 'Bargain Sale,' But Why Is It Falling Further?" Big Tech Worries Among Korean Retail Investors [Weekend Money] View original image

It's not because of expensive valuations either. The price-to-earnings ratio (PER) is actually at the lower end of its historical range. The average PER for the M7 (excluding Tesla) is 18.8 times, significantly below the three-year average of 24.5 times. The valuations of individual stocks have fallen 40% to 46% from their one-year highs for Meta, Nvidia, and Microsoft. Despite becoming cheaper, the stocks are not rising. What is happening?


Jung Han Kim, a researcher at Samsung Securities, points to the "relative valuation" environment as the answer. While big tech's capabilities remain solid, the emergence of more attractive competitors has led to a diversion of both capital and investor attention.


Big Tech Spends, Semiconductors Earn

Screenshot from Samsung Securities report.

Screenshot from Samsung Securities report.

View original image

The first competitor is the semiconductor sector. The average operating profit of the three major memory companies—Samsung Electronics, SK hynix, and Micron—is expected to surpass the M7 average in 2026. Yet, their stock prices are trading at lower valuations than big tech. Big tech companies are pouring vast amounts of capital into AI infrastructure, while semiconductor companies collect that money as revenue. This has created a situation where the companies earning more are actually trading at cheaper valuations.


The scale of investment is also a burden. Hyper-scalers are continually raising their capital expenditure estimates, and some are financing these investments through bond issuance and paid-in capital increases. Kim notes, "While the risks in the current situation are limited, unexpected external variables—such as memory shortages, rising interest rates, or a recession—could make the sector vulnerable."


This means that growth stocks investing with borrowed funds in a high-interest rate environment are starting to be treated like cyclical stocks sensitive to the economy. Apple's higher valuation compared to other big tech companies, as it stands apart from the infrastructure investment race, supports this trend.


‘AI Prodigies’ Catching Up to Big Tech in Seven Years... The Space Sector Remains Out of Reach

Capture from Samsung Securities report.

Capture from Samsung Securities report.

View original image

The second group of competitors is companies like OpenAI and Anthropic. While it took big tech an average of 27 years to reach $100 billion in revenue, Anthropic is projected to reach this milestone in just about seven years. Anthropic's annual recurring revenue (ARR) jumped from $9 billion at the end of 2025 to $47 billion as of May this year.


Capital flows are also becoming a concern. Both companies could go public as early as this fall, with expectations to raise at least $60 billion at valuations near $1 trillion. If these IPOs go ahead, some of the funds that would have flowed into big tech may be diverted to these new entrants. Since big tech has traditionally served as an indirect way to invest in frontier AI, the possibility of direct investment could erode the premium that big tech has enjoyed.


The third competitor operates in an entirely different domain. SpaceX, with its reusable launch vehicles, captured 86% of the U.S. launch vehicle market in 2025. This is a market where big tech cannot easily enter simply by deploying capital. During the listing process on June 11, some hedge funds reportedly sold big tech stocks to secure funds to include SpaceX in their portfolios. Facing competitors that are more profitable, growing faster, and even enjoying monopolistic status, big tech’s relative allure has diminished.


Big Tech’s Lament: “Disappointed by Speed, But Direction Remains Intact”

Despite stock price trends, big tech’s fundamentals are generally considered stable.


The enterprise trend of companies adopting AI into their work is spreading from coding to sectors like finance and healthcare. Google’s monthly token processing volume has jumped nearly sevenfold over the past year, and its cloud backlog is also growing rapidly. The availability of financial resources is also cited.


The four major big tech companies are expected to secure funds exceeding their annual capital expenditures through 2030, and Alphabet has announced a $85 billion capital increase plan, including a $10 billion investment from Berkshire. In Kim’s words, “There’s no concern about direction, but speed is disappointing.”


Paradoxically, entry barriers may work in big tech’s favor. Kim said, “The hurdles of building and upgrading large-scale infrastructure in the AI era are certainly not low. Considering the accelerating improvement cycles of AI models, the strategic importance of big tech companies that secure key infrastructure will only increase in the future.”


The higher the degree to which AI models self-improve, the more important it becomes for big tech companies to secure large-scale infrastructure to run those models. Regarding the agentic enterprise trend, he noted, “The trend, already demonstrated by strong first-quarter performance, is likely to strengthen over time. At least from a fundamentals perspective, concerns are limited.”


Screenshot from Samsung Securities report.

Screenshot from Samsung Securities report.

View original image

Of course, there are still variables. These include concerns over the soundness of the circular investment structure in which big tech invests in AI companies and that capital returns as computing contracts; the emergence of “AI cost control” as companies begin managing token usage; the departure of key research talent; and the pursuit of Chinese AI firms.



Nevertheless, Kim’s conclusion is broadly positive. He stated, “Although things are confusing right now, once the period arrives when big tech’s infrastructure investments are justified, there’s a high likelihood of a sharp re-rating. Rather than being swept up by the market’s fickleness, this is a time for patience and for capturing that timing.”


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing