By 2026, AI applications are expected to move from functional to highly effective, potentially creating synergies with the domestic computing power industry chain. Investment opportunities should focus on two key themes throughout the year: AI applications and domestic computing capacity.
According to Zhitong Finance APP, Galaxy Securities released a research report stating that by 2026, AI applications will transition from functional to highly efficient and may create synergies with the domestic computing power industry chain. Investors should focus on dual investment opportunities in AI applications and domestic computing capacity throughout the year. Key areas of focus include: 1) Large scale models and MaaS providers. 2) Domestic computing capacity and data center industry chain. 3) AI+Marketing. 4) AI+Industrial Software. 5) AI+Healthcare. 6) AI+Office. 7) AI+ERP. 8) AI+Finance.
Galaxy Securities’ main points of view are as follows.
The computer industry is off to a strong start to the year, with expectations for a reversal in sector sentiment. Year-to-date, the Computer Industry Index has increased by 18.04%, ranking 3rd among SW Level 1 industries. During the same period, the Shanghai Composite Index rose 3.96%, the CSI 300 index rose 2.42%, the ChiNext index rose 4.56%, and the STAR 50 index rose 11.66%. AI applications are the central driving force in this phase of the upward trend in the computing sector. The wind AI application index rose 19.25%, the commercialization logic of AI applications will gradually form a closed loop, and the computer sector sentiment is expected to drive the recovery from the lowest point.
The AI application sector is witnessing intensive catalysts with GEO reshaping traffic dynamics for the AI era. MiniMax and Zhipu AI, two major model companies listed in Hong Kong, have shown strong performance even after their IPOs. NVIDIA and pharmaceutical giant Eli Lilly and Co. have reached a collaboration worth up to $1 billion over five years to create an institute aimed at advancing fundamental models for AI-assisted drug discovery. OpenAI launches ChatGPT Health and announces acquisition of healthcare startup Torch. AI applications are receiving significant attention as they receive dense catalytic support. Traditional SEO (Search Engine Optimization) is gradually giving way to GEO (Generative Engine Optimization), driving the commercialization of AI applications from the technical validation stage to the realization of commercial value.
While B2B AI applications are expected to take off first, B2C AI applications should be viewed from a long-term value investment perspective. The commercialization logic of AI applications is likely to expand rapidly within the B2B context, with GEO leading the AI applications market as brands compete for traffic control in the AI era, thereby increasing GEO’s commercial value. Meanwhile, the cost-reducing and efficiency-improving effects of AI models in internal corporate applications are beginning to emerge. Investors should focus on the explosive growth of AI applications in B2B scenarios such as AI + Marketing, AI + Industrial Software, AI + Healthcare, and AI + Finance. Additionally, traditional high-quality B2C product application companies that already have user bases and brand recognition can further solidify their competitive advantage through AI-powered products. Long-term value investment goals in B2C applications require attention.
The demand for data centers is being released with intensive bidding, and the country’s computing capacity is about to enter a new cycle. From the fourth quarter of 2025, domestic AIDC bidding started to recover and showed an upward trend. By 2026, major domestic internet companies plan to accelerate data center deployment at a faster pace than in 2025. Once H200 supplies resume, it will improve the efficiency of large-scale model training and further facilitate the implementation of AI applications at the inference level, increasing demand for domestic computing power chips. Domestic computing capacity is expected to enter a new cycle.
risk warning
Risks associated with AI application iterations being slower than expected. increased competition in the industry; Legal and Regulatory Risks. Supply chain risk. And downstream demand is weaker than expected.
