October 9 Financial Highlights and In-Depth Analysis

OpenAI Annual Revenue Disappoints, AI Sector Tumbles, Nvidia’s Market Value Plunges $168.4 Billion Overnight

OpenAI has informed investors that its annualized revenue as of the end of September stood at approximately $50 billion, roughly $20 billion lower than the previously estimated market figure of $70 billion. This discrepancy largely stems from differences in statistical methodologies. The overnight news dealt a severe blow to sentiment in AI-related trading. Nvidia’s stock dropped nearly 3%, resulting in a market value loss exceeding $168.4 billion. Oracle’s shares fell more than 5%, Intel’s stock declined by over 5%, the Philadelphia Semiconductor Index dropped by 3.39%, and the Nasdaq Composite Index fell by 1.25%, marking its largest decline in over seven weeks.

Chief Interpretation: The gap between the $50 billion and $70 billion figures reflects not only variations in statistical methodologies but also market expectations regarding the pace of AI commercialization. The current AI trading environment has entered a phase characterized by high expectations and heightened volatility, where even minor deviations from expectations can trigger sharp revaluations, as evidenced by Nvidia’s staggering single-day loss of $168.4 billion. However, a calm and measured analysis is warranted: firstly, differences in statistical methodologies are inherent, and OpenAI’s revenue is still experiencing rapid growth, with an absolute scale that is by no means insignificant; secondly, the market’s genuine concern is that subpar monetization at the AI application level could backfire on upstream computing power procurement, creating a negative feedback loop. This is consistent with the simultaneous declines observed in Broadcom, Oracle, and CoreWeave. For A-shares, this directly dampens sentiment in the optical module, computing hardware, and semiconductor sectors, as evidenced by today’s collective decline in tech stocks. Nevertheless, the rationale for domestic computing power remains relatively independent, supported by DeepSeek’s $100 billion financing and the Ascend ecosystem, which is underpinned by its own industrial cycle. Therefore, a differentiated perspective is warranted. Investors should distinguish between trading volatility and underlying industrial trends, focusing on the alignment between computing power orders and real demand, and awaiting configuration opportunities following sentiment-driven sell-offs.

DeepSeek Secures At Least 80 Billion Yuan in New Financing Round, Led by CATL and Tencent

According to multiple media reports, DeepSeek’s latest financing round has nearly secured at least 80 billion yuan, with the final amount potentially approaching 100 billion yuan under the investment terms, significantly surpassing the initial fundraising target of around 50 billion yuan. CATL and Tencent Holdings are the lead investors in this round, with the financing process nearing completion. This round of financing will pave the way for DeepSeek’s planned initial public offering in early 2027. The company is also advancing the construction of a large-scale data center in Inner Mongolia, which will be equipped with at least 160,000 Huawei Ascend AI acceleration chips.

Chief Interpretation: DeepSeek’s successful securing of nearly 100 billion yuan in financing, nearly double the initial target, marks a historic milestone in the capital narrative of domestic large-scale AI models. The financing logic warrants deeper scrutiny—DeepSeek has never been short of capital but has been constrained by a shortage of high-end chips. Given these computing power limitations, leveraging capital to buy time and build an ecosystem is an inevitable strategic choice. The data center in Inner Mongolia, equipped with 160,000 Ascend chips, represents one of the largest single projects in domestic computing power, with its significance in binding Huawei’s industrial chain far exceeding mere financial financing. At the industrial level, the collective increase in financing scales among leading large-scale AI model companies (e.g., Yuezhiyuan’s 25 billion yuan, Zhipu’s two rounds totaling 70 billion Hong Kong dollars) signals that the AI industry’s capital expenditure cycle remains on an upward trajectory, with hot money in the primary market ultimately translating into real demand for domestic computing power, servers, liquid cooling systems, IDCs, and other infrastructure. For A-shares, this bodes well for the Huawei Ascend ecosystem, AI chips, domestic servers, and computing power infrastructure. However, caution is warranted against the transmission of primary market valuation inflation to secondary market expectations, with a preference for segments with real order backlogs and earnings delivery capabilities.

Institutional View: AI Primarily Drives China's Economic Growth Through Exports, Not Investment

Institutional reports indicate that AI has become a critical pillar of China's economic growth, with its driving force primarily stemming from exports rather than domestic investment. AI is expected to contribute approximately 0.8 percentage points to GDP growth in 2026 and 0.5 percentage points in 2027. Without AI-related export and investment demand, China's GDP growth rate could fall below 4% this year. China is emerging as the manufacturing hub for the global AI boom, supplying hardware such as servers, data processing equipment, and AI consumer electronics.

Chief Interpretation: This perspective sheds light on the unique structure of China's AI dividends: global AI prosperity’s demand is being met by Chinese manufacturing, with servers, data processing equipment, and AI consumer electronics driving export growth rather than domestic investment. This carries two key implications: firstly, the sustainability of growth contributions depends on the stability of the global AI capital expenditure cycle and the trade environment, with redistributed dividends rapidly diminishing if overseas demand slows or trade frictions escalate; secondly, the direction of policy countermeasures becomes clearer, with the necessity for policies to stabilize domestic demand rising amid a strong export and weak domestic demand landscape. For A-shares, the AI hardware export chain (optical modules, PCBs, server OEMs, consumer electronics) represents the highest-momentum direction currently, but it is already fully priced for high growth, with valuations highly sensitive to order fulfillment timelines. Simultaneously, attention should be paid to policy rebalancing toward domestic demand, with sectors like consumption and infrastructure potentially becoming the next focus for capital rotation if incremental policies are introduced. The interplay between export-oriented and domestic demand-oriented sectors will be the most critical structural theme to grasp in the coming quarter.

Click "Recommend" for good luck!

- END -