Chinese Sell-Side Analysis: AI Hardware Faces High Volatility Risks

In mid to late September, the AI hardware market quickly transitioned into a downward wave, with China and South Korea adjusting first, experiencing deeper declines and longer durations. Subsequently, Japan, Taiwan, and the semiconductor ETF SOXX also weakened. There is significant divergence in the market regarding whether AI has entered a mid-cycle collapse, but there is less controversy over the industry's advancement. Analysts point out that the main reasons for this round of decline are the high long-term interest rates in the U.S., capital expenditures relying on external financing, multi-layered leverage, and tight funding. The financing structure and cost burden of companies like CoreWeave have significantly worsened. Although the cost of model invocation has decreased, capital expenditures and supply chain costs are difficult to reduce quickly, making short-term investment returns hard to realize. ASML and TSMC have provided strong guidance this quarter, but near-term pressures cannot be ruled out. The bullwhip effect and differences in value chain levels may lead to greater volatility than expected. Additionally, use cases beyond coding are gradually emerging in scenarios such as SaaS and AI drug development, which may alleviate investment return concerns and expand demand. Analysts judge that, until more evidence emerges, the current situation resembles a high volatility phase rather than a collapse; the decline may not yet be over, and hardware leadership may yield to applications. Even if both houses change hands in the midterm elections, the impact is mainly limited to constraints on certain data center projects, and attention should be paid to the performance of large tech companies during the U.S. earnings season.