Guosen Ansheng releases 2026 investment outlook and recommends these 24 AI concept stocks.

Renowned investment bank Pu Yin Ansheng has released its investment outlook for 2026, stating that the investment logic for AI in 2026 is expected to be fully upgraded, moving from speculation on technical concepts to a period of industrial value realization. Investors will no longer blindly buy into concepts, but will evaluate whether AI technology can be transformed into actual revenue and profit, focusing on whether there are clear business models and scaling potential. Investment strategies will shift from betting on individual technology hot trends to a full industry chain layout, covering computing power, basic models, data services, vertical applications, etc., to achieve overall efficiency improvement and long-term value creation. As emphasized in the "14th Five-Year Plan" proposal, "Artificial Intelligence +" requires empowering technological innovation and industrial upgrading. The research report points out that China and the United States have formed differentiated paths in AI development. As infrastructure gradually matures, AI investment has entered the second stage, with market attention shifting to how to translate computing power into specific applications. However, the path to application implementation and monetization remains uncertain, and the ultimate winner is still unknown. China's AI development path highlights a clear focus on the real economy and national strategic orientation. China vigorously promotes the integration strategy of "Artificial Intelligence +", strengthening its role in empowering technological research and development and industrial upgrading. Compared to countries such as the United States, which are enterprise-led and more focused on consumer entertainment applications, large model capabilities competition, and capital market logic driven by AI narratives, China's AI strategic layout is more systematic and practical, with clear policy emphasis on leveraging AI technology to drive fundamental transformation of manufacturing and a fundamental improvement in total factor productivity. China's AI development is expected to achieve "overtaking on the bend". Earlier this year, the Chinese large language model DeepSeek brought a change to the AI large language model path, no longer relying solely on "computing power", but achieving better results through optimized algorithms, breaking the narrative that AI competitions belong only to the United States. In the long run, the emergence of DeepSeek will accelerate the development of AI. More efficient and cost-effective AI models are expected to emerge, which will not only help promote the widespread application of AI but also spur more business models. In terms of AI technology application development, China has advantages in talent accumulation, data acquisition, policy support, commercialization, etc., which will help China achieve a global leading position in AI development. In terms of specific investment targets, Pu Yin Ansheng is optimistic about both AI infrastructure and application layer companies. Companies in the infrastructure layer can provide short-term certainty and benefit from the long-term improvement of AI technology. Among the investment targets in the infrastructure layer, major technology companies have invested heavily in AI over the past two years, occupying favorable positions in terms of technological reserves and racecourse layout, driving their own profit growth. In terms of the application layer, we have seen an increasing number of companies promoting the transformation of AI from passive tools to active agent mode (AI Agent). In this trend, more companies in the application layer may have opportunities for explosive growth. It is expected that with the rapid development of AI, more outstanding companies in various niche tracks of AI will emerge. Based on this, Pu Yin Ansheng has identified 24 AI technology and application-related Hong Kong stock companies that are expected to be favored by investors in the short term for reference: (Character limit reached. Translation is complete.)