During the live broadcast, a review of last week's market highlights was conducted, including the market's expectations for a Fed rate cut, the impact of overseas macro events, and the adjustment of the domestic LPR interest rate. The focus was on the potential for profit expectations to recover for artificial intelligence technology stocks and their possible impact on A-shares, as well as the outlook for loose monetary policy to drive valuation recovery for high-growth technology stocks.

The CES conference displayed the rapid development of AI large models and computing infrastructure, with GPU giants introducing new generation architectures. The computing power of AI chips has reached the exabyte level, and the future of smart wearable devices is promising, with an expected market space of over a billion units in the coming years, driving the growth of the AI industry chain.

The restoration of exports of the H200 chip to China involves considerations of the market share demand of American companies and the potential growth of the Chinese economy. This event has a dual impact on accelerating the localization process of China's AI computing power. On one hand, the restoration of supply can satisfy the short-term computing power needs of enterprises. On the other hand, uncertainty drives domestic chip manufacturers to accelerate technological iteration and ecosystem adaptation, forming a dual-track parallel pattern. This provides a market validation window for the domestic artificial intelligence industry chain and is optimistic about the domestic substitution space for the domestic computing power sector in the medium to long term.

Starting from the fourth quarter of 2025, there has been talk of an AI bubble in the global capital markets, but in early 2026, the strong performance of chip manufacturing giants and the super cycle of storage chips provided a powerful rebuttal. The financial reports of chip giants show double-digit revenue and profit growth, with gross profit margins reaching a ten-year high, and revenue from advanced processes below 7 nanometers accounting for 77% of total revenue, surpassing smartphones for the first time to become the largest source of income for the company. At the same time, the company announced an aggressive capital expenditure plan, locking in production capacity based on real demand prepayments. The price of storage chips has increased by 40%, with HBM chips experiencing even greater increases due to the surge in demand for AI servers, reflecting the true downstream demand. Storage chip manufacturers have increased capital expenditures, collectively validating the sustainability of AI demand and providing a strong counterattack to the AI bubble theory.

DeepSeek's latest model, Model One, and V4 series are expected to be significantly optimized through algorithms such as the MOE architecture and MHC training methods, reducing AI computing power consumption. This will enhance the performance and market competitiveness of domestic chips such as the Ascend 910B, further opening up the imagination space for domestic replacement of computing power and strengthening the investment value of the artificial intelligence sector.

The conversation discussed the importance of AI applications in 2026, especially the release of the Qianwen APP marking a breakthrough in transactional AI assistants, demonstrating consumers' high acceptance of AI and reducing market education costs. The Qianwen APP showcases the commercial potential of AI in e-commerce, local life, transportation, and other areas, providing reference cases for other companies to implement AI applications. In addition, the ability to integrate ecosystems has become a key factor in AI competition, as platforms with mature business ecosystems and user bases will accelerate the embedding of AI capabilities, forming a moat of AI plus scenarios. China's huge Internet user base and data dividend will further strengthen its endowment advantage in the artificial intelligence industry.

Since 2026, Generative Optimization (GO) has become the focus of the market, aiming to enhance the visibility of content in AI-generated answers, with the goal of increasing the citation rate and recommendation priority of brand or product information, and achieving zero-click exposure. With the arrival of the AI era, consumer habits have shifted from manual searches on multiple platforms to one-stop AI consultations. 83% of users use AI to obtain consumer information, with satisfaction exceeding traditional searches and shopping time reduced by over 30%. The GO strategy not only provides advertising monetization paths for AI platforms, but also solves the monetization difficulties of AI applications, with an estimated market size in China reaching 24 billion yuan by 2030. At the same time, the GO strategy promotes the independent and controllable development of the domestic AI industry chain, accelerating the iteration of domestic AI large models and the maturity of the industry chain, which is beneficial for the domestic artificial intelligence sector.

In 2026, as the opening year of the Fifteenth Five-Year Plan, the artificial intelligence sector will benefit from policy catalysis, including promoting the integrated development of industry, technology, and talent, accelerating commercialization, strengthening the logic of domestic substitution, and promoting the empowerment of the manufacturing industry. It is expected to continue to see an upward trend in valuation in the spring market, with outstanding investment value.

The Science and Technology Innovation Board's Artificial Intelligence Index focuses on five major industries including electronics and computers, featuring high purity, high elasticity, and cutting-edge technology. It particularly highlights the weight in the field of digital chip design, representing China's leading AI technology capabilities and reflecting the combination of national strategic support and technological innovation.

The dialogue provides a detailed analysis of the high valuation background and rationality of the artificial intelligence index on the Science and Technology Innovation Board. It points out that although the current valuation is high, it is still near the historical median, and the PEG indicator indicates that it still has investment value. At the same time, it emphasizes the advantages of a dollar-cost averaging strategy in diversifying market volatility, reducing investment threshold, and overcoming human weaknesses. It suggests that investors choose suitable investment methods based on their own situation, and seize the investment opportunities in the Science and Technology Innovation Board artificial intelligence ETF.

The discussion centered around selecting investment strategies based on market conditions, recommending the core-satellite approach. This method involves using broad-based indexes as core assets to provide a safety net, while incorporating satellite assets such as artificial intelligence to capture excess returns. It is suitable for investors seeking a balance between stability and opportunity, and it is recommended to adjust asset allocation based on individual investment goals and risk tolerance.

The dialogue detailed the ETF grid trading strategy, which is based on market volatility, by setting buy and sell price points to achieve buying low and selling high, aiming to reduce timing risks, average costs, and lock in profits. The strategy is divided into two parts, buying and selling. When buying, buy more at low prices and less at high prices. When selling, sell more at high prices and less at low prices. Through the rule of buying and selling in batches, the strategy can flexibly respond to market fluctuations and achieve stable returns.