DeepSeek emerges, as the seven giants transform from "bull market engines" to "bear market triggers" in the US stock market?

The instability of some of the largest market capitalization companies in the U.S. stock market, especially the so-called "Magnificent Seven," has significantly increased, inevitably leading to record levels of statistical indicators measuring the "vulnerability" of individual stocks. Since the arrival of DeepSeek's "extremely low training/inference cost AI large model" from China, which has shocked Silicon Valley and Wall Street, the "grand narrative logic" of these seven giants, which hold significant weight in the S&P 500 and Nasdaq 100 indices, seems to be shifting from leading the entire U.S. stock market "bull market engine" to becoming the heaviest burden dragging down the U.S. stock market's upward momentum, potentially even dragging it into a bear market.

Looking ahead, the U.S. stock market seems increasingly likely to experience scenarios of severe price volatility similar to those seen during the late 1990s internet bubble and the bear market triggered by the bursting of the internet bubble in 2000. Statistical data shows that the instability of large-cap companies in the S&P 500 index has soared to its highest level in over 30 years. The "DeepSeek shockwave" has triggered a wave of severe sell-offs in U.S. tech stocks, especially among the seven giants, reminiscent of a series of fluctuations seen during the internet era.

The so-called "Magnificent Seven" in the U.S. stock market, which holds significant weight in the S&P 500 and Nasdaq 100 indices, includes: Apple, Microsoft, Google, Tesla, NVIDIA, Amazon, and Facebook's parent company Meta Platforms. They are the core driving force behind the S&P 500 index's repeated record highs.

Across the entire U.S. stock market, the seven tech giants have been the core driving force leading the market since 2023. With their incredibly strong revenue driven by AI initiatives, solid fundamentals, years of sustained strong free cash flow reserves, and continuously expanding stock buyback programs, they have attracted a flood of global capital. However, since the beginning of this year, the logic behind the seven giants leading the U.S. stock market has fundamentally changed. Except for Meta, the stock price performance of other tech giants has significantly underperformed the S&P 500 index, becoming the core negative catalyst dragging down the entire U.S. stock market's rise.

According to top strategists at Barclays Bank and Morgan Stanley, the core origin of the significant cracks in the logic of the U.S. stock market bull run lies in the Federal Reserve's pause on interest rate cuts and the "hawkish monetary policy stance," as well as the ongoing valuation pressure on large tech stocks brought about by the "low-cost AI shockwave" from DeepSeek. The strategy analysis team from Barclays Bank recently stated that now is the best time to turn to other stock markets outside the U.S., especially to short U.S. stocks and buy European stocks.

With an investment cost of less than $6 million and under the conditions of the H800 chip, which has performance far below that of the H100 and Blackwell chips, the DeepSeek team has created an open-source AI model with performance comparable to OpenAI's o1. In contrast, the training costs for Anthropic and OpenAI reach as high as $1 billion, while DeepSeek's pricing for inference input and output tokens can be described as "fracture-level" promotional pricing compared to OpenAI's pricing DeepSeek continuously reduces the "ineffective consumption" of general computing power through "extreme engineering, parallel optimization, and precise data screening," concentrating resources on the core modules that can most enhance model performance (attention heads, key operators, RL/distillation fine-tuning, etc.). It demonstrates how the new paradigm of "extreme engineering + post-training distillation + professional data integration + focus on reinforcement training" can approach or even surpass the performance of mainstream industry large models under limited GPU resources, posing a strong challenge to the traditional "massive cash-burning" model. Therefore, DeepSeek maximizes the potential of hardware and algorithms—this stands in stark contrast to the "extensive cash-burning" approach of American tech giants for a long time.

As the "DeepSeek low computing cost storm" sweeps the globe, investors have begun to strongly question whether the American tech giants' fervent AI cash-burning plans are reasonable. After all, expenditures reaching hundreds of billions of dollars, compared to DeepSeek's mere million-dollar costs, leave these American tech stock investors both shocked and furious. They believe that shareholder profits are being continuously eroded by unreasonable expenditures, and the so-called "AI monetization" prospects are becoming increasingly unclear.

Due to concerns that the "low-cost AI large model computing paradigm" led by DeepSeek will prompt tech giants to significantly reduce AI GPU orders in the short to medium term, NVIDIA's stock price plummeted nearly 17% on January 27, with a single-day market value evaporation of $589 billion, marking the largest market value loss in U.S. stock market history. Although NVIDIA's stock price rebounded afterward, the volatility was much greater than before the "DeepSeek shockwave." The three giants, Microsoft, Amazon, and Google, have been significantly affected since the DeepSeek impact, continuously dragging down the U.S. stock market.

Record Stock "Vulnerability" Issues a Warning to the Broader U.S. Stock Market

The so-called "stock market vulnerability" is an indicator that measures the daily price fluctuations of listed companies relative to their recent volatility. According to the latest analysis report from strategists at Bank of America Corp., by 2025, this measure for the 50 largest stocks in the S&P 500 index is heading towards its highest level in over 30 years. Data tracking shows that significant volatility is closely related to the overall volatility of the "seven giants," which can be considered the source of this extreme volatility.

This analysis is based on the average volatility and frequency of individual stock shocks so far in 2025, showing that the instability of some of the largest stocks is pushing the measure of "vulnerability" for individual stocks to record levels.

The research by Bank of America covers periods including the dot-com bubble, the Russian debt default, and the early 1990s Asian financial crisis. The emergence of DeepSeek has triggered severe fluctuations among the seven giants, leading to increased instability of individual stocks, sending warning signals to the overall market. Even as stock indices hover near historical highs, combined with tariffs and concerns over the Federal Reserve's prolonged high interest rates, this situation may bring more risks

"The U.S. stock market has been in a sideways consolidation for nearly the past three months, masking a significant increase in the volatility of individual stocks," said Matt Maley, chief market strategist at Miller Tabak + Co., in an email. "When you combine this with higher long-term U.S. Treasury yields and concerns over Trump tariffs, it creates an atmosphere that is more uncertain and tense than usual when the market is close to historical highs."

Is a "brutal bear market" similar to the bursting of the internet bubble about to descend on U.S. stocks?

Last month, the stunning emergence of China's low-cost AI model DeepSeek triggered a wave of severe sell-offs in large tech stocks led by NVIDIA, with 70 stocks in the S&P 500 experiencing "vulnerable price volatility events" where price fluctuations reached three standard deviations or more, indicating that these volatilities could cluster and trigger broader reactions in the U.S. stock market.

"The DeepSeek shockwave demonstrated how a catalyst can cause severe fluctuations in the seven giants and how these giants can drive severe fluctuations in a broader range of stocks. We also saw a similar clustering effect during the bursting of the internet bubble," said Nitin Saxena, head of U.S. equity derivatives research at Bank of America, in a phone interview.

The vulnerability of the U.S. stock market began to surge last year when stocks of mega-cap companies like NVIDIA and Tesla experienced severe fluctuations due to earnings reports, leading to a pullback in the entire U.S. stock market. This trend has continued into 2025, but has now become more intense and widespread with the arrival of DeepSeek.

This year, traditional tech giant IBM experienced severe fluctuations of about 14 standard deviations due to disappointing earnings results, leading to significant volatility in the S&P 500 on the same day, marking the largest fluctuation among the top 50 companies in the S&P 500 index so far this year.

The stock prices of pharmaceutical giant Merck and tobacco giant Philip Morris International have also seen significant fluctuations—representing the second-largest negative and positive fluctuations this year, respectively—indicating that vulnerability is spreading to other corners of the market.

Extremely high vulnerability will ultimately affect the entire market, as the largest outliers, including tech giants, hold a significant weight in benchmark indices. For investors, especially those looking to continue holding tech stocks long-term, this means they may need to incorporate more low-volatility stocks into their portfolios, potentially leading to continued sell-offs of high-valuation tech stocks like NVIDIA. This sell-off will undoubtedly push the heavily weighted seven giants and other high-valuation large tech stocks further down, dragging the S&P 500 index into a bear market According to the latest statistics from Bloomberg Intelligence, the technology stocks with the lowest volatility in the broader U.S. stock index, the Russell 1000, have outperformed those with higher volatility.

Since the beginning of 2024, the so-called low volatility factor tracked by Bloomberg Intelligence, which includes technology stocks showing the lowest standard deviation of daily investment returns over the past 6 and 12 months, has surged by 22%, while the high volatility group has only increased by 10%.

Scott Rubner, a Goldman Sachs strategist who successfully predicted last summer's sell-off wave, recently warned that the U.S. stock market is about to face a new round of selling pressure. Rubner emphasized that current market participants have become saturated, and the motivation for buying on dips is weakening. "Everyone has entered the market, including retail traders, 401(k) fund inflows, early-year asset allocation, and corporate funds," he stated. "The dynamics of fund demand are changing rapidly, and we are approaching a seasonally negative phase."

"What I am most certain about is that this strong ability to buy on dips is weakening." Among the "list" of demand slowdown that Rubner outlined, the downside risk for trend followers is particularly prominent. "If the market declines, these CTA strategies are expected to sell about $61 billion of U.S. stocks over the next month, while in a bullish scenario, the buying scale is only about $10 billion."