China's "AI +" Blueprint: Three - Step Plan for AI in the Next Decade with Strategic Insights Unveiled

In ten years, China will be fully AI-enabled.

China will be fully AI-enabled in ten years.

China's AI has entered a new decade, evolving from empowering industries to reconstructing society.

In August 2025, the State Council issued the "Opinions of the State Council on Deeply Implementing the 'Artificial Intelligence +' Initiative" (hereinafter referred to as the "Opinions"), which outlined a strategic blueprint for AI development until 2035.

From the positioning of the document, development goals to key tasks, this "top-level design" reveals a new direction for China's AI development: It is no longer just a tool for industrial upgrading, but an infrastructure and the core of new productive forces to promote China's modernization.

To achieve this goal, the document proposes a "three-step" approach:

By 2027, achieve extensive and in-depth integration of artificial intelligence with six key sectors first, and the penetration rate of applications such as the new generation of intelligent terminals and intelligent agents will exceed 70%;
By 2030, China's artificial intelligence will comprehensively empower high-quality development. The penetration rate of applications such as the new generation of intelligent terminals and intelligent agents will exceed 90%. The intelligent economy will become an important growth pole for China's economic development, promoting the popularization of technology and the sharing of achievements;
By 2035, China will fully enter a new stage of development of the intelligent economy and intelligent society, providing strong support for basically realizing socialist modernization.

In ten years, AI will be as ubiquitous as electricity and the Internet, becoming the "underlying infrastructure" of society.

Short-term goal (2027): Breakthroughs in key areas, accelerated implementation in six sectors, and intelligent terminals and agents as key carriers

In the next two years, we need to achieve the transition from technological breakthroughs to scenario implementation first, focusing on six sectors: science and technology, industry, consumption, people's livelihood, governance, and global cooperation.

As the key carriers for promoting the wide application of artificial intelligence, the penetration rate of the new generation of intelligent terminals and intelligent agents needs to exceed 70%.

As Guo Tao, an angel investor and senior artificial intelligence expert, pointed out in an interview with Yicai Global, these sectors are characterized by clear data entry points, well - defined business closed - loops, and strong technology diffusion effects. They fit the logic of "promoting the whole by key points" and will be the core means for policy implementation.

In the science and technology sector (AI for Science, AI4S), AI is no longer just an auxiliary tool. It is expected to become a new paradigm for future scientific research and even open up a new research path for philosophy and social sciences.

For example, the document proposes for the first time to build large scientific models, promoting a paradigm revolution in scientific research from "0 to 1". At the same time, AI will also change the R & D chain, significantly shortening the distance between the laboratory and the market.

In the industrial sector, we need to promote the intelligent transformation of the "three pillar industries" (industry, agriculture, and services) and foster new "intelligent native industries".

Among them, "intelligent native enterprises" are the highlight of the policy. They build their businesses based on AI as the underlying logic. Just as "Internet native companies" (such as Meituan and Didi) emerged in the era of mobile Internet, new enterprise forms such as customer service platforms based on intelligent agents and AI - driven automated design companies will emerge in the AI era and become the key objects of future policy support.

Especially in the service industry, the application prospects of AI are the broadest. Intelligent agents and the new generation of intelligent terminals (AI mobile phones, AI PCs, and intelligent assistants) will become new service entry points.

In the consumer sector, AI not only promotes the improvement of production - end efficiency but also deeply reshapes the service and product forms, directly targeting individuals and families to improve the quality of life and consumption experience.

At the product level: Cars, mobile phones, computers, robots, home appliances, and wearable devices will all use AI as their "brains" to achieve the interconnection of all things.
At the service level: From digital convenience to "cognitive and emotional consumption", in the future, AI will not only "help you buy" but also accompany and inspire you.

In the people's livelihood sector, AI will truly "enter people's lives". From work, education, health to culture and social interaction, it will build a more intelligent way of working, a more personalized learning model, and a more warm - hearted society. The policy also encourages AI to create more works with "Chinese elements" in cultural production to strengthen cultural confidence.

In the governance sector, AI will comprehensively improve the modernization level of governance. In social governance, from smart cities to smart government affairs, human - machine co - governance will be realized. In security governance, a three - dimensional intelligent security system will be built to strengthen national security. In ecological governance, AI will be used to promote the construction of a "Beautiful China" and support the green transformation.

In the global cooperation sector, China will advocate a "universal sharing" approach, in contrast to the "security - restricted" model, and contribute Chinese solutions to global artificial intelligence governance.

Medium - term goal (2030): From key points to a broader scope, the intelligent economy becomes a growth pole

By 2030, the connotation of "AI +" will expand from the six sectors to all aspects of the intelligent economy, becoming an "important growth pole" driving the national economy.

By then, the penetration rate of applications of the new generation of intelligent terminals and intelligent agents will also increase from 70% in 2027 to over 90%. A 90% penetration rate means that almost every industry, every organization, and most individual users are using intelligent terminals or agents. AI will shift from scenario implementation to economic driving.

This trend is not a fantasy. Nature once reported the "Agent Moore's Law" proposed by the non - profit research institution METR: The capabilities of intelligent agents double approximately every seven months. At this pace, by the end of 2028, AI will be able to automatically complete many tasks that currently take humans a month to finish. If the penetration rate of intelligent agents can reach 70% in 2027, then driven by the exponential iteration of technology, it may be natural for the penetration rate to exceed 90% in 2030.

It is worth noting that the document also emphasizes the popularization of technology and the sharing of achievements, reflecting the awareness of social fairness and value distribution.

Long - term goal (2035): From a broader scope to a comprehensive coverage, strong support for the intelligent society and modernization

By 2035, AI will penetrate from being an important growth pole of the intelligent economy to the entire intelligent society, becoming a strong support for China's modernization.

The intelligent society is not only a change at the economic level but also a profound transformation at the social level. AI will penetrate into all aspects of people's lives, including public services, urban governance, and personal life, forming an efficient, convenient, and sustainable intelligent ecosystem.

The basic realization of socialist modernization indicates that the development of AI has become an important support for achieving the country's long - term development goals.

Models, data, computing power, and open - source: A four - in - one approach to accelerate the development of the AI industry

Recently, domestic AI companies have continuously released new - generation open - source large models. All the top 15 open - source AI models on the Design Arena ranking list are from China.

In the July open - source achievements of the Chinese AI community released by Hugging Face, many manufacturers, including Alibaba, Zhipu, Kunlun Tech, Dark Side of the Moon, Tencent, and Jieyue Xingchen, have successively open - sourced 33 large models.

Previously, the institution Interconnects (a high - quality content platform focusing on cutting - edge AI research) summarized 19 top domestic open - source model laboratories, including top - tier institutions like DeepSeek and some emerging academic laboratories that have emerged through technical reports and niche models.

The "Opinions" also clearly propose to support the construction of artificial intelligence open - source communities, promote the aggregation and opening of models, tools, and data sets, and cultivate high - quality open - source projects. In this regard, the government encourages universities to incorporate open - source contributions into student credit certification and teacher achievement evaluation, and supports enterprises, universities, and research institutions to explore new, inclusive, and efficient open - source application models.

Models, data, and computing power are becoming the "three high - points" of global AI competition. Through the combination of the open - source wave and policy guidance, China is forming a synergy among scientific research, industry, and application, promoting the AI industry into a new stage of accelerated development.

The "Opinions" clearly propose to comprehensively improve the basic support capabilities of models, data, and computing power.

In terms of models, it emphasizes strengthening basic theoretical research on artificial intelligence and innovation in model infrastructure, promoting the improvement of training and inference efficiency, exploring new forms of model application, and establishing a sound model capability evaluation system to ensure that model iteration is controllable and verifiable.

In terms of data, the opinions propose to build high - quality data sets, improve the data property rights and copyright system, promote the legal and compliant opening of copyright content formed by public - funded projects, explore a data cost compensation and revenue - sharing mechanism based on value contribution, and support the development of technologies such as data annotation and data synthesis.

In terms of computing power, the opinions clearly support the innovation of artificial intelligence chips and the construction of ultra - large - scale intelligent computing clusters, improve the national integrated computing power network, give full play to the role of hubs such as the "East - to - West Computing" project, and encourage standardized and scalable computing power cloud services.

Regulation and security: China's AI governance enters a new institutionalized stage

From "model hallucinations and algorithmic discrimination" to the filing system, regulation is about to be comprehensively upgraded.

The "Opinions" not only emphasize building security capabilities in all aspects such as algorithms, data, computing power infrastructure, and application systems but also clearly propose to face up to and address the typical risks brought by the current AI development (the word "security" is mentioned 12 times in the full text), such as the black - box problem (unexplainability) of models, hallucinations (generating false or unreliable information), and algorithmic discrimination (bias against groups or individuals).

It highly echoes the international concepts of "explainable AI" and "responsible AI", reflecting China's overall idea of balancing security and development in artificial intelligence governance.

In fact, various national departments have also issued many policy documents on artificial intelligence governance this year.

For example, on March 14, four departments, including the Cyberspace Administration of China, the Ministry of Industry and Information Technology, the Ministry of Public Security, and the National Radio and Television Administration, jointly issued the "Measures for Marking Artificial Intelligence - Generated Synthetic Content", which will come into force on September 1, 2025. The measures require that all AI - generated synthetic content must be legally marked with an "electronic watermark".

Specifically, service providers are required to add explicit and implicit marks to content such as text, pictures, audio - video, and virtual scenes generated by AI to ensure public right to know and traceability; dissemination platforms need to verify the marks and remind the public; applications need to be reviewed for compliance before being launched; users must actively declare and mark when publishing content, and it is strictly prohibited to tamper with or delete the marks.

On March 21, the Cyberspace Administration of China and the Ministry of Public Security issued the "Measures for the Safe Management of the Application of Facial Recognition Technology", which came into force on June 1, 2025.

The core of the measures is to strictly regulate the use of facial recognition. It requires clear purposes, necessity, and the principle of minimization, obtaining separate consent and protecting the rights and interests of minors; data should be stored locally and kept for a limited period, and important applications must be filed; for identity verification, facial recognition should not be forced as the only method, and alternative means should be provided; the collection of data in public places should be reasonable and legal, and it is prohibited to install equipment in private spaces; the system must take encryption and security protection measures, and those who violate the regulations will be held legally accountable.

Conclusion

As early as 2017, the State Council issued the "New - Generation Artificial Intelligence Development Plan", which became the first national - level AI strategy.

However, the "Opinions on Deeply Implementing the 'Artificial Intelligence +' Initiative" has a more distinct focus and systematicness. It focuses on the integrated application of "artificial intelligence +" for the first time and makes a systematic deployment in six sectors such as science and technology, industry, and consumption, reflecting the special - policy guidance for a single cutting - edge technology.

Moreover, the opinions pay more attention to practicality. In response to problems such as emphasizing hardware over software, fragmented applications, and low activity in open - source communities in the application and implementation of artificial intelligence, a series of targeted measures are proposed. It adheres to industry - specific policies, proposing the development directions of integration with artificial intelligence for each sector respectively, forming the main line of thinking for the "artificial intelligence +" in each industry.

Links to policy documents:

Opinions of the State Council on Deeply Implementing the "Artificial Intelligence +" Initiative

https://www.gov.cn/zhengce/content/202508/content_7037861.htm

Answers to Journalists' Questions by Responsible Comrades of the National Development and Reform Commission on the "Opinions on Deeply Implementing the 'Artificial Intelligence +' Initiative"

https://www.gov.cn/zhengce/202508/content_7037920.htm

Measures for Marking Artificial Intelligence - Generated Synthetic Content

https://www.gov.cn/zhengce/zhengceku/202503/content_7014286.htm

Measures for the Safe Management of the Application of Facial Recognition Technology

https://www.cac.gov.cn/2025 - 03/21/c_1744174262156096.htm

New - Generation Artificial Intelligence Development Plan

https://www.gov.cn/gongbao/content/2017/content_5216427.htm

Reference links:

https://www.yicai.com/news/102794683.html