I. Policy Background: AI Industry Enters New Stage of 'Ecosystem Construction' from 'Technological Breakthroughs'
In 2026, the AI industry has moved beyond the technology validation phase and entered the stage of large-scale commercial application. According to MIIT data, the core AI industry scale reached 500 billion yuan in 2025, a year-on-year increase of 35%, but the industrial ecosystem still faces issues such as fragmentation and insufficient collaboration. The release of this White Paper aims to bridge the gap between 'technology and application', promote the construction of an open and collaborative ecosystem, and lay the foundation for the 800 billion yuan target by 2028.
In recent years, AI policies have gradually shifted from 'encouraging technology R&D' to 'promoting ecosystem construction'. The 'Next-Generation Artificial Intelligence Development Plan' (2020) proposed a 'three-step strategy', the 'Action Plan for the Innovative Development of the Artificial Intelligence Industry' (2023) focused on the three key elements of computing power, algorithms, and data, while the 2026 White Paper emphasizes 'open collaboration'—breaking down barriers between enterprises, universities, and research institutions to form an integrated 'industry-academia-research-application' ecosystem. This shift marks the AI industry's transition from 'single-point breakthroughs' to a new stage of 'systematic collaboration', with policy orientation placing greater emphasis on enhancing overall industrial competitiveness.
II. Interpretation of the Core Content of the White Paper: Three Directions Driving Industrial Upgrading
1. Clear Targets: Core Scale to Reach 800 Billion by 2028, Focusing on 'High-Quality Growth'
The White Paper proposes that by 2028, the core AI industry scale will reach 800 billion yuan, with a compound annual growth rate (CAGR) of over 25%. This target not only emphasizes scale expansion but also focuses on 'high quality'—namely, technologically self-reliant, rich application scenarios, and efficient ecosystem collaboration. Compared to 500 billion yuan in 2025, the 800 billion yuan target means an average annual growth of about 20% over the next three years, posing higher requirements for each link in the industry chain. Notably, the definition of 'core industry' in the target is clearly 'core AI technologies and products', not the broad AI-related industry, highlighting the policy's emphasis on 'hard technology'.
2. Ecosystem Construction: Open Collaboration Becomes the Core Keyword
For the first time, the White Paper positions 'open collaboration' as the core direction of the industrial ecosystem, proposing a 'platform + ecosystem' model: on one hand, supporting leading enterprises to build open platforms to provide basic resources such as computing power, algorithms, and data; on the other hand, encouraging SMEs to develop innovative applications based on open platforms to form an ecosystem pattern of 'leading enterprises guiding and SMEs supplementing'. For example, Huawei's Ascend and Alibaba's DAMO Academy have already launched open platforms, and more enterprises will join in the future. This model aims to address the pain points of 'large enterprises monopolizing resources and SMEs lacking technology', promoting AI technology from 'laboratories' to 'markets'.
3. Key Areas: Three Levers of Computing Power, Application, and Security
The White Paper identifies three key areas: first, computing power infrastructure, promoting the development of domestic computing power chips, servers, and data centers to reduce the cost of large model training; second, application scenario implementation, focusing on AI applications in industries such as industrial, medical, education, and finance to promote deep integration of 'AI + industry'; third, security and ethics, establishing an AI security assessment system to prevent technological risks. Among these, 'computing power' is placed first as the 'cornerstone' of the AI industry, highlighting the policy's emphasis on 'self-reliance'; 'application scenarios' are the 'next phase' promoted by the policy, emphasizing the value of 'technology implementation'; 'security and ethics' are the guarantee for the healthy development of the industry, preventing technology abuse.
III. Industry Impact: New Opportunities for Upstream and Downstream of the Industry Chain
1. Computing Power Segment: Domestic Chip and Server Manufacturers Benefit
Computing power is the 'cornerstone' of the AI industry, and the White Paper emphasizes the importance of 'self-reliant computing power'. Currently, the domestic computing power market still relies on imported chips (such as NVIDIA A100), while the market share of domestic chips (such as Enflame T20, Cambricon Thinker) is gradually increasing. Under policy promotion, domestic computing power manufacturers will face development opportunities: on one hand, the performance of domestic chips is continuously improving—for example, the inference performance of Enflame T20 reaches 80% of NVIDIA A100, with a 50% cost reduction; on the other hand, policy support for data center construction promotes the domestication of computing power infrastructure. For example, enterprises such as Accelink (computing power optical modules), Sugon (servers), and Cambricon (AI chips) are expected to see improved performance as the share of domestic computing power increases. In addition, emerging business models such as computing power leasing and trading platforms may emerge, bringing new growth points to the industry chain.
2. Application Segment: B2B Software and Service Providers Become New Mainstream
As policies promote the implementation of 'AI + industry', B2B application scenarios have become investment hotspots. According to the position adjustment trends of public funds, capital has shifted from computing power to the application end, with commercial realization becoming the keyword of the next phase. The White Paper clearly identifies 'application scenarios' as a key area, focusing on industries such as industrial, medical, education, and finance. For example, enterprises in fields like industrial AI (e.g., intelligent manufacturing), medical AI (e.g., new drug R&D), and financial AI (e.g., intelligent risk control) will see growth. Taking industrial AI as an example, Yonyou Network's (industrial internet) AI applications have covered over 1,000 manufacturing enterprises, helping customers reduce production costs by 15%; Weining Health's (medical informatization) AI new drug R&D platform has helped enterprises shorten the R&D cycle by 30%. The performance growth of these enterprises will attract investor attention. In addition, niche segments in the AI application end (such as AI digital humans, AI short video generation) may also explode, bringing new opportunities for individuals and enterprises.
3. Ecosystem Segment: Open Platforms and the Rise of SMEs
The construction of open platforms will drive SMEs to participate in the AI ecosystem. The White Paper proposes an ecosystem pattern of 'leading enterprises guiding and SMEs supplementing', encouraging SMEs to develop innovative applications based on open platforms. For example, individual developers can develop AI applications through Huawei's Ascend open platform, and enterprises can obtain technical support through Alibaba's DAMO Academy open platform. This model will lower the threshold for SMEs to enter the AI industry and promote the popularization of AI technology. In addition, niche segments such as AI agent custom development and AI side projects (e.g., AI picture book creation, AI programming) may explode. For example, AI agent custom development has become a new trend, with individual developers earning 100,000 yuan per month from orders, and enterprises competing to purchase; after the upgrade of AI picture book creation tools, novice mothers can earn 20,000 yuan per month by 'telling stories'. These niche segments will become new growth points in the ecosystem segment.
IV. Investment Opportunities: Layout Logic Under Policy Benefits
1. Computing Power Industry Chain: Focus on Domestic Substitution and Cost Reduction
The computing power segment is a policy focus, and sub-sectors such as domestic chips, optical modules, and servers are expected to benefit. First, the self-reliance of domestic chips is a core policy goal, and enterprises like Cambricon and Enflame Technology will gradually increase their market share as technology breakthroughs occur; second, computing power optical modules are key components of data centers, and Accelink's (computing power optical modules) performance is directly related to data center construction, with its revenue expected to grow as the share of domestic computing power increases; finally, server manufacturers (such as Sugon) will benefit from data center construction, especially the increasing demand for domestic servers. Investors can focus on the performance improvement of these enterprises and market opportunities brought by the reduction in the cost of domestic computing power.
2. Application End: The 'Commercialization Inflection Point' for B2B Software and Service Providers
The application end is the 'next phase' promoted by the policy, and B2B software service providers (such as industrial software, medical software) will reach the peak of commercialization. The White Paper emphasizes the deep integration of 'AI + industry', and AI applications in industries such as industrial, medical, and finance will become key focuses. For example, Yonyou Network's (industrial internet) AI applications have entered a large-scale stage, with revenue increasing by 25% year-on-year in the first half of 2026 and net profit increasing by 30%; Weining Health's (medical informatization) AI new drug R&D platform has cooperated with 10 pharmaceutical companies and is expected to see a 40% revenue growth in 2026. The performance growth of these enterprises will attract investor attention. In addition, niche segments in the AI application end (such as AI digital humans, AI short video generation) may also explode, such as AI digital human live streaming becoming a new trend, with ordinary people earning 50,000 yuan per month using virtual images; AI short video batch generation tools are popular, with personal bloggers earning 50,000 yuan per month through 'script + editing' automation. These niche segments will become new growth points at the application end.
3. Ecosystem Segment: Benefits of Open Platforms and Niche Segments
The construction of open platforms will create new investment opportunities, such as AI agent custom development and AI side projects. The White Paper proposes an ecosystem pattern of 'leading enterprises guiding and SMEs supplementing', and open platforms will become the core of the ecosystem. For example, Huawei's Ascend open platform and Alibaba's DAMO Academy open platform will attract SMEs to participate in the AI ecosystem. In addition, AI side projects (such as AI programming, AI picture book creation) will become new ways for individuals to increase income, such as zero-code beginners earning 30,000 yuan per month from AI programming side jobs, and novice mothers earning 20,000 yuan per month by 'telling stories' after the upgrade of AI picture book creation tools. These niche segments will become new growth points in the ecosystem segment, bringing new opportunities for investors.
V. Conclusion: New Opportunities in the AI Industry Driven by Policy
The release of the MIIT White Paper marks the AI industry entering a new stage of 'ecosystem construction'. Under policy promotion, the three segments of computing power, application, and ecosystem will face development opportunities, and investors can focus on sub-sectors such as domestic computing power, B2B applications, and open platforms. As the 800 billion yuan target by 2028 approaches, the AI industry will usher in a new cycle of 'high-quality growth', with the layout logic shifting from 'technological breakthroughs' to 'commercial implementation', bringing long-term opportunities for investors. At the same time, the policy emphasizes 'open collaboration', meaning that upstream and downstream of the industry chain will cooperate more closely to form an integrated 'industry-academia-research-application' ecosystem, which will further promote the healthy development of the AI industry. For investors, it is necessary to pay attention to policy orientation, layout sub-sectors that meet the direction of 'high-quality growth', and seize new opportunities in the AI industry.
