China's New Generation of Artificial Intelligence Technology Industry Development (2023)

Recently, the China New Generation Artificial Intelligence Development Strategy Research Institute of the Chinese Academy of Engineering released the "China's New Generation Artificial Intelligence Technology Industry Development 2023" (hereinafter referred to as the "Report"). The theme of this year's report is "Building Globally Competitive AI Industry Clusters".

The development of artificial intelligence industry in mainland China shows a clear trend of clustering

AI enterprises and their innovative activities constitute the micro-foundation for the development of AI industry clusters.

Statistics from the Ministry of Industry and Information Technology show that as of June 2022, the number of artificial intelligence enterprises in the mainland exceeded 3,000, ranking second only to the United States, and the scale of artificial intelligence core industries exceeded 400 billion yuan. The innovation and entrepreneurship activities of mainland artificial intelligence enterprises in intelligent chips, infrastructure, operating systems, tool chains, basic networks, intelligent terminals, deep learning platforms, large models and industrial applications have laid the foundation for the construction of independent and controllable technology systems and the improvement of the international competitiveness of the industry.

Platform enterprises, unicorn companies, small and medium-sized enterprises, start-ups, research universities, research institutes and investors cooperate with each other to jointly build a dynamic industrial innovation ecology, and the development of artificial intelligence industry shows an increasingly obvious trend of clustering.

The AI industry cluster is an innovation cluster based on cyberspace development

The "Report" research shows that unlike the traditional industrial cluster and innovation cluster concept, the AI industrial cluster is an innovation cluster based on cyberspace development. Based on the innovative development of physical space technology system, the industrial clusters and innovation clusters of the first three industrial revolutions have dependence on geospace, and the speed of innovation diffusion is relatively slow. The fourth industrial revolution stems from the innovative development of cyberspace technology system, innovation clusters are more dependent on cyberspace development, technology, products and services are innovating faster, and the application fields and geographical scope of innovation are more extensive.

The basic components of AI industry clusters include enterprise clusters, innovation resources, innovation systems, and cyberspace industry ecology. Among them, the innovation ecology of cyberspace industry is a unique element of artificial intelligence industry cluster under the background of the fourth industrial revolution. Artificial intelligence industry clusters include artificial intelligence industrialization innovation clusters and industrial intelligence innovation clusters. The AI industrialization cluster empowers the geospatial scattered industrial intelligent innovation cluster through the cyberspace industry ecology. The benign interaction between AI industrialization innovation clusters and industrial intelligence innovation clusters is the key driving force and mechanism for building globally competitive AI industry clusters.

The artificial intelligence industry cluster in the mainland is manifested as a spatial structure of "new innovation zones→ cities→ regions→ national →global". Different from traditional industrial parks and high-tech parks, the new innovation zones are generally located in the central and sub-central areas of large cities with strong scientific and technological innovation resources and industrial foundation, and are the habitat of artificial intelligence industrialization clusters and their industrial innovation ecology, emphasizing the creation of unlimited cyberspace industrial innovation ecology by relying on small physical space.

So far, the mainland's AI industry clusters are mainly distributed in key cities in the Beijing-Tianjin-Hebei region, the Yangtze River Delta, the Pearl River Delta and the Sichuan-Chongqing region. Through the introduction of external innovation resources and the activation of internal innovation resources, Xi'an in the western region, Wuhan and Changsha in the central region, and Shenyang, Dalian and Harbin in the northeast region have begun to take shape.

Enterprise clusters and their industrial innovation ecology

The report builds a sample database of China's smart economy, including 2,200 AI companies, 5,722 investors (investment institutions and non-investment institutions), 438 AI universities and 307 non-university scientific research institutions, 967 industry alliances, a total of 2,318 conferences held in China, 775 relevant policies issued by 31 provinces, municipalities and autonomous regions, and the planning and construction of 3,507 AI industrial parks. Through attribute data and relationship data analysis, the internal structure and development trend of the mainland artificial intelligence industry cluster were investigated.

The value network structure of the mainland artificial intelligence industry cluster is "polar core". The platform and its leading industrial innovation ecology constitute the "polar core" of the development of the mainland's artificial intelligence industry cluster. From the structural statistical indicators of the value network from 2014 to 2022, the cluster structure characteristics of the mainland artificial intelligence industry cluster are becoming more and more obvious. Super platforms represented by Huawei, Tencent, Baidu and Alibaba are the core nodes for the formation and development of artificial intelligence industry clusters in mainland China. In recent years, the R&D and industrialization layout of super platform in the fields of intelligent chips, infrastructure, operating systems, large models, machine learning platforms and application software have laid a solid foundation for the improvement of the international competitiveness of the mainland's artificial intelligence industry cluster.

Research universities, research institutes and new innovation organizations are important components of the platform-led industrial innovation ecosystem. The efforts of research universities and research institutes in the fields of basic research, technology development and talent training continue to enhance the international competitiveness of the mainland's artificial intelligence industry cluster. 18 domestic universities, including Tsinghua University and Peking University, have become the first batch of doctoral degree authorization points in integrated circuit science and engineering in China. As of March 2022, a total of 440 colleges and universities across the country have set up artificial intelligence undergraduate majors, 248 colleges and universities have set up intelligent science and technology undergraduate majors, and 387 ordinary colleges and universities have set up "artificial intelligence technology services" majors for higher vocational education (junior college).

Creating a new type of innovation organization to activate the vitality of government-industry-university-research-application collaborative innovation, and forming a new system and mechanism for basic research, technology research and development, applied innovation and industrial incubation seamless docking is an important way to promote artificial intelligence science and technology innovation and industrial development. So far, this report has found a total of 347 new innovation organizations in the field of artificial intelligence, which are widely distributed in Beijing-Tianjin-Hebei, Yangtze River Delta and Pearl River Delta. Among them, the artificial intelligence laboratories represented by Pengcheng Laboratory, Zhijiang Laboratory and Shanghai Artificial Intelligence Laboratory have become the most active new innovative organizations in the field of artificial intelligence industrialization.

"Polarization" and "diffusion" of innovation

As a general-purpose technology, the scientific and technological innovation and industrial development of artificial intelligence follow the law of "polarization" and then "diffusion". Based on the regional, application, technology and industrial field distribution of technical cooperation between 2,200 mainland AI backbone enterprises, the report depicts the "polarization" and "diffusion" of mainland AI industry clusters.

(i) Region

From the perspective of technical cooperation, Beijing, Guangdong Province and Shanghai constitute the three "poles" of the value network of the mainland's artificial intelligence industry cluster. At the same time, there are intensive AI technology partnerships between Beijing, Guangdong, Shanghai, Jiangsu, Anhui, Sichuan, Hubei, Hunan, Chongqing, Shandong and Fujian.

From the perspective of the flow of technical cooperation, the development of the mainland's artificial intelligence technology industry is still dominated by "polarization". The first and second places were internal technical cooperation between Beijing and Guangdong Province, accounting for 10.87% and 9.36%. Technical cooperation between Guangdong Province and Beijing Municipality, Beijing Municipality and Guangdong Province ranked third and fourth, accounting for 6.91% and 6.08%. To some extent, Beijing and Guangdong Province together constitute the north-south "double poles" of the development of Chinese industrial intelligent industrial clusters.

In terms of the flow of technical cooperation between cities, Beijing, Shenzhen, Guangzhou and Shanghai are the cities with the highest density of technical cooperation. In particular, the technical cooperation between Beijing, Shenzhen and Guangzhou has become the main direction of "polarization" and "diffusion" of artificial intelligence technology.

From the perspective of technical cooperation between cities, the first place is Beijing's internal technical cooperation, accounting for 10.87%; Second, third and fourth were technical cooperation between Shenzhen and Beijing, Shenzhen and Shenzhen, and Beijing and Shenzhen, respectively. From the perspective of the city, Beijing and Shenzhen constitute the north-south "poles" of the flow of AI technology cooperation.

(ii) Areas of application

With the acceleration of the pace of scientific and technological innovation, artificial intelligence and economic and social development have entered a new stage of comprehensive integration and development. In the application fields and industrial fields with high density of artificial intelligence technology cooperation, industrial intelligent innovation clusters have begun to appear.

Based on the quantitative analysis of the relationship data of 2,200 AI backbone enterprises, mainland AI has been widely used in 19 application fields, including enterprise intelligent management, intelligent marketing and new retail, smart finance, smart city, smart healthcare, new media and digital content, smart manufacturing, smart education, intelligent transportation, network security, smart logistics, smart cultural tourism, smart government, smart energy, smart hardware, intelligent networked cars, smart home, smart agriculture and smart security. The first place is smart city, accounting for 12.16%; The second place is enterprise intelligent management, accounting for 12.10%; Ranked third is intelligent manufacturing, accounting for 8.89%; The fourth and fifth places are intelligent marketing and new retail and intelligent networked vehicles, accounting for 8.41% and 8.07% respectively.

(iii) Industrial fields

Among the three industries, the tertiary industry with the highest distribution density of artificial intelligence technology cooperation cooperation is the tertiary industry, accounting for 75.49%; It is followed by the secondary industry, accounting for 23.82%. In the tertiary industry, the first place is information transmission, software and information technology services, accounting for 28.46%; In second place is scientific research and technical services, accounting for 21.17%; In third place is the leasing and business services industry, accounting for 10.75%; The fourth and fifth places were finance, wholesale and retail respectively, accounting for 10.68% and 9.62%.

In the secondary industry, manufacturing accounted for the highest proportion, at 87.36%. In the manufacturing industry, the first place is the computer, communication and other electronic equipment manufacturing industry, accounting for 28.16%; The second place is the automobile manufacturing industry, accounting for 25.41%; The third place is the electrical machinery and equipment manufacturing industry, accounting for 9.30%.

(iv) Technical categories

The deep integration and development of artificial intelligence and economy and society has led to the systematization, complexity and specialization of artificial intelligence technology. So far, artificial intelligence has developed into a complex technical system including 17 technologies, including big data and cloud computing, Internet of Things, intelligent robots, intelligent recommendation, 5G, blockchain, speech recognition, virtual/augmented reality, smart chips, computer vision, natural language processing, biometrics, space technology, optoelectronic technology, autonomous driving, human-computer interaction and knowledge graph. At the same time, with the innovative application of artificial intelligence in 19 application fields, the evolution of technical systems increasingly shows a trend of specialization.

(v) Focus areas for cluster development

From the distribution of regions, applications, technologies and industrial fields of innovation "polarization" and "diffusion", with the innovation of artificial intelligence science and technology, innovation clusters in key industrial fields including intelligent manufacturing, intelligent chips, intelligent networked vehicles, scientific and technological research and development and services, smart medical care and smart education are the forefront of the development of artificial intelligence technology industrial clusters. For example, since 2019, in the development of the smart chip industry, a number of smart chip R&D and design companies have emerged, including Ali Pingtouge, Baidu Kunlunxin, Huawei HiSilicon, Bicheng Technology and Yiwei Semiconductor. Platform enterprises build a technical foundation for innovation clusters by building a root technology innovation system and a collaborative software and hardware innovation ecology including smart chips, operating systems, machine learning platforms and pre-trained large models.

In the intelligent networked vehicle industry, a software and hardware collaborative industry innovation ecosystem with Baidu and Huawei as "dual-core" has been formed. The addition of traditional automobile enterprises, new forces in car manufacturing, small and medium-sized enterprises and start-ups has accelerated the development of intelligent networked vehicle industry clusters. At the same time, the social experiment research of China's New Generation Artificial Intelligence Development Strategy Research Institute shows that the mainland has a good social atmosphere in the development of the intelligent networked vehicle industry. With the deep integration of artificial intelligence and the automotive industry, the intelligent networked vehicle industry cluster will become the representative of the integrated development of artificial intelligence and the real economy.

With the development of generative artificial intelligence, the innovative application of artificial intelligence in scientific and technological research and development, services and education will trigger a new round of integration of industry and education, which is a new frontier for the development of artificial intelligence industry clusters. The transformation of innovative production methods brought about by artificial intelligence not only brings about the rapid development of the industry, but also brings new changes in the paradigm of scientific and technological innovation and the paradigm of education.

Challenges and policy recommendations

Artificial intelligence is the focus of global technological and industrial competition. Driven by deep science and technology innovation, an artificial intelligence innovation alliance is being formed in China to compete with the United States. Building an independent and controllable technology system and a collaborative innovation ecosystem for software and hardware is the strategic goal of cultivating and developing globally competitive AI industry clusters.

Despite unprecedented achievements, in the process of building a globally competitive AI industry cluster, we are still facing challenges brought by the US technology blockade, the shortcomings of the technical system and the relatively slow technological upgrading of the head platform enterprises. Accelerating the development of new platforms with industrial empowerment capabilities and their leading industrial innovation ecology, high-level planning and development of new innovation zones, building a highly open innovation system, promoting technical cooperation with countries around the world, and promoting the integration of general artificial intelligence and special artificial intelligence are strategic supports to cope with challenges and accelerate the international competitiveness of artificial intelligence industry clusters. (Source/China Academy of Information and Communications Technology)

Protection statement: This number selects content with high-quality communication value, and may be partially abridged or modified. We respect the copyright of high-quality original content very much, if the selected content in this number cannot contact the original author himself, please contact us.