Artificial intelligence has pivoted from a "theme-based speculation" back to a phase of "earnings validation" in the recent period. An overseas AI juggernaut posted quarterly revenue and next-quarter guidance that both surpassed consensus expectations, with data center demand continuing to scale higher. More significantly, the company delivered an initial growth outlook for the coming fiscal year that was substantially above market forecasts, shifting the narrative focus from whether AI demand is overestimated to whether supply constraints can be further alleviated.

For A-shares, this suggests the AI hardware boom may be far from over—it is transitioning from first-order aggregate expansion to second-order differentiation in pricing. This dynamic tends to favor segments capable of delivering technological iteration and enhanced value-per-unit. Concurrently, advancements in domestic large language models have enabled the "domestic model + domestic chip" pairing to achieve its first robust instance of scaled validation. Zhipu this week unveiled an open-source model release, with all public test traffic carried by domestic chip clusters, achieving end-to-end performance and unit costs comparable to mainstream overseas solutions. Meanwhile, MiniMax continues to post strong growth in commercialization metrics, with explicit progress on domestic chip adaptation that is expected to lift its market share going forward.

Alibaba and other players are also persistently open-sourcing cost-effective, high-performance models, while industry-wide token call volumes are still accelerating—clear evidence that inference demand is being unleashed steadily. The central challenge for compute infrastructure is shifting from "whether we have the chips" to "how to efficiently organize systemic computing power." (Mentions of specific equities are purely illustrative and do not constitute investment advice; markets carry risk, and investment decisions require caution.)

The implications for A-shares are clear: overseas mapping remains essential, yet domestic self-reliance is no longer merely a policy narrative—it is beginning to demonstrate commercial viability and cost closure. The domestic chip, switching and interconnect, server ODM, advanced packaging, and materials and equipment supply chain deserve active monitoring. Strategically, we continue to hold the view of short-term consolidation with accumulation, and the technology sector remains the mid-term main line. The index has policy and valuation support on the downside, while the upside is constrained by external rates, geopolitical disruptions, and the intensity of incremental capital inflows—resulting in a landscape closer to range-bound trading with structural upside.

On style, technology is still set to remain the mid-term core theme, but the framework needs to evolve from pure overseas mapping to one built on "overseas AI earnings validation + domestic self-reliance commercialization + interim report delivery." Correspondingly, attention should pivot toward a balanced strategy of "growth as the spear, dividends and the price-hike chain as the shield." Across asset classes, A-shares may still outperform high-multiple overseas growth assets, though careful attention to timing management is necessary. Gold retains medium-term allocation value, but short-term volatility is likely elevated on U.S. Treasury and oil price fluctuations. For overseas equities, the focus remains on whether AI earnings momentum can broaden into the wider mid- and downstream segments.

Moving ahead, three sets of signals warrant close observation: first, whether two-market turnover stays active and whether tech repair transitions from impulses to sustained momentum; second, the impact of the Jackson Hole symposium and Federal Reserve Chair Warsh's tone on long-end yields; third, whether Middle East tensions and oil prices reconfigure once more, and whether post-interim earnings delivery continues to broaden into equipment, materials, and domestic supply chain segments.

Disclaimer: This communication draws on sources the company deems reliable and the individual judgment of research personnel, but the company offers no express or implied warranty as to its accuracy or completeness. This communication is not a full account or summary of any related securities or markets, and any opinions expressed are subject to change without notice. It should not be relied upon by recipients as a substitute for independent judgment or as a basis for investment decisions. The company, its affiliates, employees, or agents accept no liability for any actions taken or losses incurred from the use of this content, in whole or in part. Without prior written permission from Great Wall Fund Management Co., Ltd., no one may distribute, reproduce, copy, or publish this report in any form, in whole or in part, or make any deletions or alterations contrary to its original intent. Fund managers remind citizens that everyone has the duty and right to report money laundering crimes, and all citizens should strictly comply with anti-money laundering laws and regulations. Investing involves risk and demands caution.