JPMorgan: Asian Tech Stock Pullback Does Not Signal End of AI Investment Cycle

Recent corrections of approximately 25% to 30% in Asian technology stocks and the Philadelphia Semiconductor Index have raised market concerns about whether the artificial intelligence (AI) boom is cooling down. However, according to a new report released by JPMorgan, the firm believes this market volatility does not signify the end of the AI investment cycle, and emphasizes that industry fundamentals remain strong.

Solid Industry Fundamentals

JPMorgan's analysis indicates that this correction marks the third major pullback since the AI-driven rally began at the end of 2022. Analysts stress that there is almost no evidence suggesting a slowdown in AI industry fundamentals, and they do not expect any significant weakening over the next 6 to 12 months.

JPMorgan anticipates that hyperscale cloud service providers will not easily cut back on AI capital expenditures and may continue investing in AI computing infrastructure through at least 2027. These companies could raise funds through equity or debt markets to support ongoing infrastructure expansion.

Equipment and Advanced Packaging Lead the Way

Within the semiconductor supply chain, JPMorgan is most bullish on semiconductor equipment manufacturers. With wafer fabrication equipment (WFE) investments accelerating, this segment is expected to be best positioned over the next 12 months. Additionally, as 2.5D packaging technology becomes more widespread and TSMC initiates its 3D packaging investment cycle, the packaging and testing sector is poised for strong growth. Among components, IC substrates are viewed most favorably.

In the memory market, while fundamentals remain robust and supply growth is expected to lag demand over the next two to three years, JPMorgan warns that memory stocks may struggle to revisit their May highs due to NVIDIA (NVDA-US) and AMD (AMD-US) adjusting memory configuration ratios in their products.

Potential Risks and Future Constraints

Despite optimism about technological prospects, JPMorgan cautions in another report that current AI momentum bears similarities to the dot-com bubble era, with profits overly concentrated in a few stocks and valuations increasingly decoupled from fundamentals. The firm advises investors to moderately shift toward value stocks and adopt geographically diversified portfolios.

Looking ahead, JPMorgan forecasts that power supply could become the primary bottleneck for AI infrastructure development within the next 18 to 24 months, surpassing chip availability. Simultaneously, as computational efficiency becomes critical, interconnect technology will emerge as the next major challenge.