During the National Day Golden Week holiday, news at home and abroad was generally quiet, yet global capital markets still displayed a pattern of divergence with an upward bias, driven by the AI narrative. Powered by tech stocks, the Nasdaq index rose 2.52% and the S&P 500 index gained 1.96%. (Data source: Wind)

But upon closer examination, this rally was not simply a broad rise across all tech stocks. The standout performers remained AI infrastructure names, with companies such as Nvidia, Broadcom and AMD continuing to attract capital flows. (The individual stocks mentioned are for illustrative purposes only and do not represent any specific investment advice. Stock markets carry risk, and investment requires caution.)

Turning to the domestic market, the first trading day for A-shares after the holiday failed to deliver a positive open, and the technology sector also saw a certain degree of correction. How will the AI sector play out from here? Liu Lisi, fund manager of the Invesco Great Wall technology team, offered her outlook for the market ahead.

The Major AI Industry Trend Remains Positive

Liu Lisi believes the AI industry is advancing from an earlier phase of conceptual exploration toward commercialization and validation of returns. Whether it is the monetization of B2B enterprise service tools or the reshaping of traffic by consumer-facing applications, the establishment of a commercial closed loop will continue to feed back into upstream infrastructure. As the foundational support of the AI industry, computing power enjoys relatively strong demand support.

In her view, the major AI industry trend remains positive, though various fluctuations are inevitable along the way, such as a vacuum period for new commercialization scenarios, trade wars, and Middle East conflicts. Looking back at past major corrections, US stocks have generally stabilized and recovered earlier than A-shares. One important reason is that investors positioned closest to the frontier of AI innovation are better able to grasp the latest positive changes in the industry amid short-term volatility, thereby returning to the investment certainty of a medium- to long-term perspective.

Liu Lisi concluded: although new commercialization scenarios still require monitoring and waiting, and some segments of the industrial chain face policy interference risks, the strong performance of overseas markets once again confirms that the major trend has not changed materially. At the current stage, she maintains a neutral-to-optimistic view, keeps tracking developments, deepens her understanding, and optimizes the portfolio.

The Strategic Value of Domestic Computing Power Deserves Attention

Among the many high-growth segments of the AI industrial chain, why does domestic computing power deserve special attention? This can be understood from three dimensions.

First, demand is rigid. Computing power refers to the underlying computational capacity required for AI to operate, and it is the foundational base that supports every model query, inference and training session. Simply put, the stronger the computing power, the faster and smarter AI becomes. In Liu Lisi's view, one important characteristic of AI is that its marginal cost is not zero, since every query, inference and training session consumes real computing resources, making continuous investment in computing power a necessary prerequisite for delivering AI services.

The data is even more compelling. According to National Bureau of Statistics data, China's average daily Token call volume has grown from 100 billion in early 2024 to 140 trillion by March this year. (Data source: National Bureau of Statistics, March 2026) In addition, as AI applications move from single-turn dialogue to composite tasks, computing power demand is expanding exponentially.

Second, there are threefold constraints on the supply side. Overseas computing power faces supply and cost pressures, and the threefold constraints of not being sufficient, not being cheap enough and not being safe enough have raised concerns, thereby elevating the strategic value of domestic computing power.

Third, domestic computing power is in an acceleration phase. In terms of the pace of investment, overseas computing power investment began to accelerate markedly in 2023 and has continued to this day, while domestic AI investment started roughly two years later. This time gap means that domestic computing power may be expected to enter a period of rapid development, with the industrial chain continuing to iterate and more companies entering the supply chain. Building in-depth research and a long-term tracking system for the domestic computing power industrial chain helps to grasp industry development trends more proactively and uncover new opportunities.

Against the backdrop of the rising strategic value of domestic computing power, the Invesco Great Wall Digital Economy Ruixuan (Class A 029141/Class C 029142), to be managed by Liu Lisi, will be launched on October 12. The fund focuses on the digital economy sector, with domestic computing power as its core direction. The fund manager will focus on the entire industrial chain, including chips, servers, networking, data centers and computing power services, striving to capture structural opportunities in the gradual implementation of domestic substitution.

Liu Lisi is a new-generation fund manager in the Invesco Great Wall technology team, with 10 years of experience in the securities and fund industry and 1.8 years of investment experience. Her research experience is relatively extensive, covering industries spanning electronics in the technology sector, machinery in the manufacturing sector, and transportation in the cyclical sector. (Data source: Invesco Great Wall, as of August 2026)

In addition, Liu Lisi upholds an investment philosophy that respects industrial laws and plain facts, and excels at starting from in-depth research while focusing on identifying certain targets within medium- to long-term industry trends. This is reflected in portfolio allocation. Taking the Invesco Great Wall Growth Trend fund she manages as an example, second-quarter report data show that the top ten holdings together accounted for a weight close to 70% of the equity position, relatively concentrated in semiconductor packaging, chips, PCB and other segments of the AI industrial chain.

In terms of performance, as of September 30, the fund's Class A shares achieved a net value growth rate of 17.65% over the past six months. (Data source: Wind; the benchmark rose 8.87% over the same period; Liu Lisi began participating in managing the fund on September 16, 2025; holdings data come from periodic reports)

Beyond deep exploration within her circle of competence, Liu Lisi also emphasizes continuous high-quality expansion rather than pursuing short-term expansion on an excessive scale. Leveraging her investment research foundation in technology and cyclical manufacturing, she currently focuses on expanding into AI, automobiles, resources and other related fields, building sustainable and replicable investment research competitiveness.

Risk disclosure: The above views represent opinions at the time and may change in the future. They are for reference only, do not constitute investment advice or guarantees, and should not serve as any legal document. A MACD golden cross signal has formed, and these stocks are rising well!