This is China’s first systematic deployment of artificial intelligence as a national development engine through top – level design. In the next decade of the “AI +” era, traditional enterprises will undergo a reconstruction of organizational structure and capability framework, and service providers will witness a reshaping of business logic and delivery models. Cloud providers and chip manufacturers will shift from a computing power war to a closed – loop service model.
From the emergence of the “New Generation Artificial Intelligence Development Plan” in 2017 to the sweeping trend of large models in 2023, artificial intelligence technology has experienced wave after wave: a surge in parameters, model iterations, and capital chasing. AI has once become the hottest symbol of innovation.
However, the real turning point has only truly arrived now.
Recently, the State Council issued the “Opinions on Deeply Implementing the ‘Artificial Intelligence +’ Initiative” (hereinafter referred to as the “Opinions”). This is China’s first systematic deployment of artificial intelligence as a national development engine through top – level design, shifting from single – point breakthroughs to overall integration and rising from an industrial hot topic to a national strategy. This means that AI has entered an era of “use or not” and “how to use” rather than just “existence” and “strength”.
This is a reconstruction of production relations.
After carefully reading the content of the “Opinions”, the term “artificial intelligence” appears 85 times, and keywords such as “accelerate”, “integrate”, and “innovate” frequently appear, sending an unprecedentedly strong signal. The positioning of AI is no longer just a technological addition but the underlying operating system of the national economy. It will be deeply embedded in multiple core fields such as industry, agriculture, education, medical care, government affairs, transportation, and energy, becoming a new fulcrum for leveraging China’s industrial development.
More importantly, the policy for the first time proposes the concept of “Artificial Intelligence +”, replacing the previous superposition idea of “+ Artificial Intelligence”. This “+” sign means a role transformation, from passive integration to active reshaping. AI is no longer just a functional patch attached to industries but has become an intelligent foundation for building new business formats, reconstructing business logic, and driving the leap – forward development of governance capabilities.
As the train of the era is speeding up, participants standing at the carriage door need to be clear about where the opportunities lie in the “Artificial Intelligence +” era, how industries will change, and how to get on board.
1. What exactly does “Artificial Intelligence +” entail?
To analyze the changes brought about by this policy, we first need to figure out what this policy is all about.
After reading the “Opinions” through, we can summarize its core goal in one sentence: Deeply embed artificial intelligence into all systems of national operation within a decade.
Different from previous policies that emphasized “supporting the development of AI enterprises” or “promoting technological breakthroughs”, this “Opinions” clearly states that artificial intelligence should become a basic national capability, just like electricity and the Internet. In the future, every industry, every local government, and every enterprise must have and use it.
Meanwhile, the “Opinions” also provides specific action measures, mainly carried out through six action paths, starting from scientific and technological innovation, moving on to industrial transformation, consumption upgrading, public services, and finally to social governance and global cooperation.
It is worth noting that these paths are not independent of each other but are nested and support each other, forming a complete strategic system.
Specifically, AI for Science makes scientific research more efficient, which is the origin. Industrial transformation makes manufacturing, agriculture, and energy more intelligent, which is the key. Intelligent consumption changes products and user experiences, which is the interface. Livelihood services make education, medical care, and elderly care more inclusive, which is the scenario. Public governance makes cities and the country more intelligent, which is the ability. Global cooperation gives China a say in AI rules, which is the vision.
This design is like a “national AI infrastructure map”, with a scope far beyond science and technology itself.
In addition, this “Opinions” also for the first time sets a quantitative indicator of “AI terminal penetration rate”. It proposes that the penetration rate should reach 70% by 2027, 90% by 2030, and achieve full penetration by 2035 to build an intelligent society.
This is an unprecedented statement. In previous national strategies, there were penetration rate indicators for smartphones, broadband, and rural electricity, but it is the first time to include “artificial intelligence terminals” in the national – level KPIs.
The logic behind this is clear. China does not want only a part of people and enterprises to use AI but wants it to become a public infrastructure like water, electricity, and gas.
In other words, from now on, AI is no longer just the concern of technical personnel but a real – world problem that every industry manager must face.
Looking back at 2015 when “Internet +” was written into the government work report, it triggered a nationwide digital wave, giving birth to platform giants such as Pinduoduo, Meituan, Didi, and ByteDance. However, the core of that wave was “connection is king”, and the key was the connection between people and information to improve distribution efficiency.
This time, “Artificial Intelligence +” goes a step further. It is not just about connection but about cognitive and capability upgrades, making products understandable and decision – making, and enabling systems to have self – optimization capabilities.
This represents a fundamental paradigm shift, from using tools to making tools think. AI may not be just the next step in digitalization but a brand – new starting point.
Actually, this is not without warning, and the timing of the release of the “Opinions” also has a lot to say. You know, the real opportunity always appears when multiple variables reach their critical values simultaneously.
Technologically, large models have completed multi – modal integration, and the concept of Agent has moved towards productization. Industrially, pilot applications in various industries have taken shape, and data and model resources have gradually accumulated. In terms of governance, from algorithm management measures and generative AI management rules to the initial formation of industry evaluation systems. Strategically, the focus of Sino – US technological competition has shifted to AI, and the global governance order is facing reconstruction.
Taking action at this time is the golden period for large – scale implementation. The country needs to unify standards, guide resources, layout computing power, and release scenarios, and promote AI to become a real social infrastructure from a global perspective.
2. Where are the opportunities? Six major industrial AI sectors become new high – ground
So, where are the real opportunities behind this policy?
Actually, the answer lies in every aspect of “Artificial Intelligence +”.
In the part of scientific and technological innovation, the document clearly proposes to promote the participation of artificial intelligence in the entire process of scientific research, accelerate the construction of scientific large models and the ability to process complex scientific research data, and promote the transformation of scientific research paradigms from “experience – driven” to “model – driven”.
This part particularly emphasizes interdisciplinary cross – over and the construction of intelligent scientific research infrastructure, which means that AI is no longer just an application object but a tool and collaborator in scientific research itself.
In the future, fields such as AI products and platforms for scientific research scenarios, interdisciplinary AI talent cultivation, and scientific data service platforms will witness rapid development and become new directions for universities, research institutions, and technology – based enterprises.
In the industrial field, the document for the first time systematically proposes the concept of “intelligent native” enterprises, encouraging enterprises to embed AI into strategic planning, organizational structure, and business processes to build a fully intelligent production system.
This signal is very clear. That is, AI should not just be a “tool” but the “operating system” of enterprise operations. The “Opinions” also mentions the AI application paths in multiple industries such as manufacturing, agriculture, and services, emphasizing the integrated collaboration of technology, talent, and processes.
This means that whether it is traditional industry, agriculture, or modern service industries, new space will be released in this AI – driven reconstruction. New models such as AI + ERP systems, intelligent industrial software, agricultural AI solutions, and “unmanned services” will rise rapidly.
In the consumption part, the policy focuses on the dual upgrade of intelligent terminals and service consumption, proposing the directions of “cognitive and emotional consumption” and “companion and assistant applications”, emphasizing the reshaping of the interaction mode between intelligent products and consumers.
The main push is to let AI penetrate into lifestyles in all aspects, bringing the AI consumption experience into the “human – machine co – existence” era. For example, smart homes, smart cars, wearable devices, virtual companions, AIGC entertainment content, and personalized e – commerce recommendations mentioned in the “Opinions”.
Behind this are a large number of market opportunities in intelligent terminal hardware manufacturing, AI application content entrepreneurship, and new human – machine interaction technologies.
The action opinions in the livelihood field are particularly practical. The “Opinions” emphasizes promoting AI to empower scenarios directly related to people’s lives such as education, medical care, employment, culture, and elderly care, and specifically proposes to develop new roles such as “intelligent teachers”, “intelligent health assistants”, and “intelligent vocational training”. For education technology companies, digital medical platforms, and human resources service agencies, this is a brand – new blue ocean.
In terms of social governance, the policy promotes the role of AI to a higher governance level. The document emphasizes building a new model of urban operation with human – machine co – existence, promoting the in – depth application of AI in government services, public safety, ecological governance, etc., and clearly states that it is necessary to promote “urban intelligent upgrading” and “construction of AI government service platforms”.
Behind this, two important messages are conveyed: one is that AI will directly participate in the allocation of public resources, and the other is that the data – driven governance model will become the mainstream. In the future, whether it is the intelligent transformation of urban infrastructure or the upgrade of county – level government systems, a large number of AI solutions and local implementation services will be needed, giving rise to a new round of investment boom in government technology, urban brains, and public data applications.
Finally, the document expands its vision globally, proposing to build artificial intelligence into an “international public product”, promoting the open – source sharing of AI technology, international standard setting, and global governance cooperation.
This part conveys China’s strategic intention to participate in the reshaping of the global AI landscape. For Chinese enterprises, this means that the expansion of the overseas market is not limited to products and services but also includes the output of underlying chips, models, algorithm frameworks, and standard setting, both in terms of hard and soft power. The overseas expansion of AI infrastructure, the internationalization of model – as – a – service platforms, and the global contribution of open – source projects will become new growth points for China’s technological strength.
It can be seen that “Artificial Intelligence +” is not just a slogan but a systematic design map covering the underlying logic of national development. What it brings is a systematic reshaping of the future society. Technology and institutions, efficiency and fairness, individuals and society, industries and the country, all these elements will be recombined under the framework of “AI +”.
And the opportunities lie in the gaps of this major reshuffle. Those who can find their own “+” may stand at the forefront of this transformation.
3. Welcome to the new decade of the “AI +” era
The implementation of a national – level strategy is not just about policies but also the beginning of the redefinition of industrial structures.
“Artificial Intelligence +” is redefining the boundaries of capabilities. Data needs to be re – integrated, systems need to be reconstructed, products must be intelligentized, and the service delivery mode will also be comprehensively upgraded. This means that every link in the industrial chain, from traditional manufacturing enterprises to SaaS service providers, from cloud providers to chip companies, from development platforms to middle – tier systems, needs to re – find its own positioning and value.
The next few years will be the most critical stage of this “reconstruction competition”.
In the past decade, most industries in China have completed the basic digitalization process of cloud – based and systematic transformation. However, this time, “Artificial Intelligence +” is not just about adding a plug – in to the old system but rewriting the enterprise architecture with intelligent capabilities, which poses higher requirements for traditional enterprises.
For example, manufacturing enterprises no longer just purchase AI visual inspection systems but need to build a production intelligent agent that can schedule tasks, automatically optimize parameters, and coordinate multiple devices. Energy enterprises no longer just install monitoring equipment but need to establish a real – time energy efficiency control system with prediction and strategy feedback capabilities. Educational institutions need to use AI teachers to replace some standardized teaching tasks and write personalized learning curves into the underlying logic of the teaching system.
This requires enterprises to build new capabilities, namely data asset governance capabilities, intelligent agent collaboration frameworks, and cross – departmental “product + technology” joint modeling mechanisms. Only in this way can AI not be just an add – on for enterprises but a second operating system growing inside the organizational structure.
For service providers and ISVs, change is also imminent.
In the past, most AI companies appeared in the form of tools, models, and components. However, under the policy framework of “Artificial Intelligence +”, AI suppliers need to complete a role transformation from “developers” to “joint creators”.
For example, an industrial AI company cannot just provide “algorithms + deployment” but needs to jointly build vertical models with manufacturing enterprises, continuously conduct data annotation, and jointly optimize business logic.
This means that service providers not only need to have technical capabilities but also industry understanding; they should not only be able to “deliver models” but also “co – build systems”; the model form should shift from static release to dynamic iteration; and the product form should shift from API tools to AaaS (Agent as a Service) intelligent agent services.
In this trend, what really determines whether a service provider can win a bid is no longer just algorithmic capabilities but its comprehensive performance in co – construction mechanisms, industry scenario adaptation, and compliance governance capabilities.
Cloud computing platforms are facing another systematic change.
The most direct resource opportunity brought by “Artificial Intelligence +” belongs to cloud service providers and intelligent computing centers. The policy clearly proposes to build a “cloud – edge collaborative system for intelligent agents”, establish a national – level intelligent computing scheduling center, and encourage local governments and industries to jointly build “intelligent computing service bases”.
Behind this, a clear signal is released: AI is no longer the exclusive resource of large enterprises but should become a basic capability that small and medium – sized enterprises can also afford, just like public water and electricity.
For cloud providers, they not only need to optimize the cost structure of the training and inference cycles, achieve key breakthroughs such as multi – model concurrency, low – bit quantization, and edge inference; in terms of service models, they also need to upgrade from IaaS to PaaS, SaaS, and even AaaS to promote the integrated delivery of AI capabilities.
In this reconstruction of intelligent infrastructure, database and chip manufacturers also face a historic opportunity.
In the field of databases, as RAG (Retrieval – Augmented Generation) becomes the mainstream for large – model implementation, databases are also moving from traditional transaction processing to a “structured + unstructured” hybrid processing model. AI databases that natively support vector retrieval, graph neural networks, and streaming embedding updates are becoming the new favorites of government and enterprise customers.
In the field of chips, Chinese domestic NPU, GPU, and AI accelerator manufacturers are facing the dual benefits of “domestic substitution” and “diversification of computing power”. The policy encourages the construction of heterogeneous computing power pools and the development of edge inference capabilities, making AI chips not just an option but an indispensable key element for building intelligent agents.
At the same time, chip and database manufacturers cannot just sell chips or operators but need to package “scenarios, optimization, and scheduling” and extend upstream to system integration, becoming systematic deliverers for the implementation of the AI industry.
Generally speaking, in the next decade after the implementation of the new policy, traditional enterprises will undergo a reconstruction of organizational structure and capability framework, service providers will witness a reshaping of business logic and delivery models, and cloud providers and chip manufacturers will shift from a computing power war to a closed – loop service model.
In the next stage, the logic of the AI implementation industrial chain will shift from competing in capabilities to competing in systems. Those who can quickly build their own “intelligent native capability matrix” will be able to take the lead in this reshuffle period and occupy the industrial high – ground.
This article is from the WeChat official account
该文观点仅代表作者本人，36氪平台仅提供信息存储空间服务。