If the AI Bubble in U.S. Stocks Bursts, Can the Chinese Market Stand Strong on Its Own?

The Crisis Lurking Beneath the Frenzy in the U.S. AI Market: Can the Chinese Market Stand Strong If the AI Bubble Bursts?With sky-high orders, massive debts, and circular trading, the AI market frenzy in the U.S. stock market has raised alarms among global investors. After a series of soaring days, Oracle saw a sharp drop of over 7% on October 7, after media reported its cloud profit margins were underwhelming, reigniting the debate about the AI bubble in U.S. stocks. Reflecting on the bursting of the dot-com bubble at the beginning of this century, it revealed a harsh reality: Most internet companies couldn’t justify their valuations with real business performance. Back then, valuation metrics shifted from traditional factors like cash flow and profitability to website traffic and growth data. Today, AI companies are facing a similar test – despite historic levels of AI investment in the U.S., the revenue gap remains vast. Tech writer Ed Zitron recently pointed out that Microsoft, Meta, Tesla, Amazon, and Google have collectively invested around $560 billion in AI infrastructure over the past two years, but their total AI-related revenue is only $35 billion.It’s easy to imagine that if OpenAI’s capital investment returns fall short of expectations, the high valuations of U.S. tech giants like Oracle may face major revisions.

As AI develops independently in both China and the U.S., if the AI bubble bursts in the U.S., can the Chinese market navigate a different course?U.S. AI Sector Exposes “Shared Risk” Pitfalls.In September this year, Oracle’s stock surged over 40% in a single day, the largest single-day increase since 1992, making its founder, Larry Ellison, briefly the world’s richest person, surpassing Tesla’s Elon Musk. Once seen as a “dying” database company, Oracle seemingly ascended overnight to become an AI powerhouse. This incredible growth was driven by a $300 billion OpenAI computing order and the resulting massive debt. Oracle’s debt-to-equity ratio now stands at 427%, far exceeding the 100% threshold typical for most tech companies. The core issue in the U.S. AI sector is the creation of a closed loop of “high investment – high valuation – higher investment,” which is inflating the AI bubble. After Oracle announced its deal with OpenAI, Nvidia also revealed plans to invest up to $100 billion in OpenAI, which, in turn, began purchasing Nvidia GPUs. This “invest before purchasing” model has become the new norm in the AI industry. From chip giants to cloud computing service providers, every participant is both a benefactor and a beneficiary, forming a unique “shared risk” ecosystem.

This deep entanglement has begun raising red flags among some observers, who compare it to the “vendor financing” seen before the 2000 internet bubble burst — upstream vendors provide funding to downstream customers, who then generate revenue through purchases. Alibaba’s Chairman, Jack Ma, recently pointed out that the U.S. AI data center market is showing signs of a bubble, noting that “many projects are raising funds before even securing clear customers.” Barclays Capital’s research report issued in September also sent a warning, stating that AI is still a powerful investment theme but is “in a bubble that has not yet fully materialized,” and if capital expenditures decline, valuations are likely to be more impacted than profits.China’s AI Path: A Different Approach Amid U.S. Bubble Risks.Despite the potential bursting of the AI bubble in the U.S., China’s AI market faces distinct challenges and opportunities. The key difference in the AI development paths of China and the U.S. lies in their approaches to technology and commercialization. Chinese AI companies are more focused on cost control. For instance, DeepSeek’s DeepSeek-R1 model offers comparable performance to leading models at a lower cost, sparking concerns over potentially reduced demand for computational power.In terms of business models, U.S.

AI giants mainly focus on direct charging, relying on subscriptions and hardware sales for monetization, while Chinese AI companies focus on deeply integrating AI with existing businesses. For example, iFlytek’s AI solutions business saw revenue of 439 million RMB in the first half of 2025, a nearly 3.5x increase year-on-year, showing strong commercialization potential. Additionally, the Chinese government’s support for the AI industry has been more systematic and long-term. The Chinese State Council recently issued guidelines to promote the “AI+” strategy, aiming to integrate AI with six major sectors by 2027. This clear top-down design provides a clearer direction for China’s AI industry development.China’s AI industry also benefits from a unique regional synergy effect. Due to the lower costs of building and operating data centers in the western region (50%-70% lower than in the east), the “East Data, West Computing” project has established a national integrated computing network, now one of the world’s largest AI computing clusters. The synergy between data, computing, electricity, and networks in this project is remarkable. Each billion yuan of computing power investment can generate an additional 1.8 billion yuan in other industries. The project spans multiple Five-Year Plans, with clear stages of development. The “14th Five-Year Plan” is focused on building computing infrastructure and coordinating electricity, while the “15th Five-Year Plan” (2026-2030) will be a period of significant growth in computing power.

China’s Unique Edge in AI Commercialization.Unlike the U.S., which focuses heavily on hardware, China’s AI industry emphasizes practical application scenarios and commercialization. According to an MIT report from August, although 90% of enterprise employees use general large models for their work, only 5% of companies achieve quantifiable commercial returns from large model applications. In contrast, many Chinese companies are focusing on “AI application technology,” developing intelligent systems that can solve real business problems.While China’s AI market has unique characteristics, if the U.S. AI bubble bursts, China may not be completely immune. Global capital sentiment is interconnected. Barclays Capital has noted that if data center capital expenditure declines by 20% in the next two years, rather than growing as expected, the S&P 500 could see a 10-13% drop by FY2026. Such volatility would inevitably affect global investors’ risk tolerance for AI, including Chinese companies that heavily rely on U.S. AI giants’ orders, such as the optical module leader “Yi Zhongtian,” which recently shone in A-shares.Chinese Tech Giants’ Valuations as a Buffer Against U.S. AI Bubble.Despite the potential impact of the U.S. AI bubble burst, China’s tech giants are better positioned to withstand the shock, mainly due to their more reasonable valuations compared to U.S.

AI giants. Morgan Stanley’s chief China equity strategist, Wang Ying, pointed out in an early October interview that Chinese internet giants transforming into AI leaders are still significantly behind their U.S. counterparts in terms of total market value, freely tradable market cap, and valuation levels. These giants are not “storytellers” but have fertile grounds for commercializing AI: massive data, rich application scenarios (e-commerce, social media, cloud computing), and strong engineering teams. For them, AI is not just a burning money “concept” but a “super amplifier” that empowers their existing businesses and creates a second growth curve.Wang Ying also noted that market concerns about regulation and competition have led to these companies being undervalued, creating a huge potential upside once their AI business starts contributing significant profits. However, she cautioned that this feast is not for everyone — smaller companies focused on niche sectors with sky-high valuations require caution. The real opportunity lies in leaders with “stable profit delivery tracks and market share expansion potential.” Without a doubt, as the U.S. AI market’s enthusiasm wanes, China’s AI industry is transitioning from “hardware-driven” to “hardware-software synergy.” Those Chinese tech giants that establish global competitiveness in computing infrastructure, large model development, and vertical applications will not only benefit from domestic policy and market dividends but also expect sustained capital flow from the global revaluation of the AI industry chain.