Artificial intelligence accelerates the empowerment of thousands of industries

According to data released by the Ministry of Industry and Information Technology a few days ago, the development of artificial intelligence in mainland China has made positive progress, with the number of enterprises exceeding 4,500, the emergence of innovative achievements such as smart chips and general large models has accelerated, the intelligent infrastructure has been continuously consolidated, and the construction of digital workshops and smart factories has been accelerated, laying a good foundation for artificial intelligence to empower new industrialization.

Tao Qing, director of the Operation Monitoring and Coordination Bureau of the Ministry of Industry and Information Technology, said that it is necessary to give full play to the advantages of the mainland's complete industrial system, large industrial scale and rich application scenarios, with the deep integration of artificial intelligence and manufacturing as the main line, intelligent manufacturing as the main direction, and scenario application as the traction, to promote the intelligent transformation of the manufacturing industry, empower the industrial manufacturing system at a high level, and provide new momentum for high-quality development.

Accelerate breakthroughs in key technologies

Artificial intelligence is a strategic technology leading a new round of scientific and technological revolution and industrial transformation, and an important driving force for new industrialization. Only by accelerating breakthroughs in key technologies based on artificial intelligence can we consolidate the foundation of application empowerment.

At the 2024 IT market annual meeting held recently, Wang Peilin, deputy secretary-general of the Artificial Intelligence Special Committee of the China Computer Industry Association, said that the core technology foundation of artificial intelligence in mainland China is weak and there is a top-heavy industrial structure. At the application level, it has developed rapidly, but its R&D and industrialization capabilities at the basic level and key technology level are relatively weak, and it lacks core technologies and products with international competitiveness.

Tao Qing revealed that it is necessary to accelerate the breakthrough of basic key core technologies and products such as intelligent chips, large model algorithms, and frameworks around the underlying technologies of large models such as algorithms and computing power; Accelerate the deployment of intelligent IoT terminals and industrial clouds, and improve the ability of computing power supply, operation and management for the manufacturing industry; Guide ecological enterprises to accelerate the creation of a general model with global competitiveness, and cultivate an industry model for manufacturing scenarios; Build a high-level industrial database and stimulate the value of industrial data elements.

"Only by integrating the three core elements of computing power, algorithms and data can we bring 'artificial intelligence +' to the industry and truly realize the intelligent empowerment of the industry. In April this year, we released the business strategy of the Hairuo large model, planning to invest 5 billion yuan to accelerate the rapid implementation of the Hairuo large model in 100 cities relying on the digital innovation ecological community. Yin Ping, industry director of Inspur Cloud, said.

In recent years, the computing power industry in mainland China has developed rapidly. According to Shan Zhongde, Vice Minister of the Ministry of Industry and Information Technology, by the end of 2023, the total scale of data center racks in use in mainland China will exceed 8.1 million standard racks, and the total scale of computing power will reach 230 exascale floating point operations per second. The scale of intelligent computing power has reached 70 trillion floating point operations, with a growth rate of more than 70%.

"Computing power has always been a key factor restricting the development of artificial intelligence." Fu Changwen, President of CCID Consulting Co., Ltd., said that the current scale of computing power in mainland China ranks second in the world, but there are still problems such as unbalanced computing power structure. With the rapid development of artificial intelligence and the continuous advancement of industry and enterprise digitalization, the demand for intelligent computing centers is increasing. In 2022, the gap in China's intelligent computing power will be 52.6%, and in 2023, it will reach 78.4%.

Wang Peilin believes that the cultivation and expansion of new quality productivity is inseparable from the deep integration and efficient operation of "artificial intelligence +" and "data element ×". The in-depth promotion of these two actions will bring a historic opportunity for industrial transformation to the mainland, and is expected to create a number of digital industrial clusters with global influence and promote a qualitative leap in the modern industrial system.

"The digital economy driven by digital technology is developing rapidly in an all-round way, and innovation based on digital technology has become a key support for the formation of new quality productivity. At present, the national basic data system and the management of data elements have entered a substantive stage, and new technologies, new concepts, and new systems in the digital field are constantly emerging. Su Bo, director of the National Intelligent Manufacturing Expert Committee, said.

Expand multiple application scenarios

Zhang Li, President of China Electronic Information Industry Development Research Institute, said that in 2023, IT technology represented by generative artificial intelligence will trigger a new revolution in the intelligent era and become the driving force for promoting application innovation in thousands of industries. The goal of the "Artificial Intelligence +" action is to actively apply artificial intelligence to empower the real economy, improve the efficiency of the real economy, promote the high-quality development of the real economy, and bring new industries, new models and new momentum to various fields of the economy and society.

The mainland has a wide and rich range of artificial intelligence application scenarios, which has created a huge space for "artificial intelligence+". Artificial intelligence has been widely used in imaging, retail, manufacturing, finance, medical and other fields, changing the production mode and economic form, improving production efficiency, reducing production costs, and effectively enhancing the international competitiveness of the industry.

At the 14th Beijing International Film Festival held not long ago, two old films released more than 40 years ago were re-premiered as new 4K films. According to reports, in the digital restoration process, the engineers of the Volcano Engine technical team used AI identification technology to reduce the restorer's repetitive work, help optimize the clarity, fluency, and color of the film, and remove flaws. Ma Kede, assistant professor of the Department of Computer Science at City University of Hong Kong, said that with the widespread application of generative AI visual model technology in more film and television scenes, this technology trend may become the mainstream of 4K film restoration in the future.

In Weihai City, Shandong Province, a fishing boat sailed into the port and was immediately caught by the fishery administration verification and recording system on the shore. This high-tech "receptionist" developed by China Unicom quickly confirms the identity of the ship and completes the verification within 5 minutes, which greatly shortens the time to determine the identity of the ship and improves the accuracy. In the 16 months since the fishery administration verification and recording system was launched, the average recognition rate accuracy of the AI "ship face" algorithm has reached more than 86%, and 8,580 AI early warnings, data collection of more than 3,100 fishing vessels, and more than 19,153 times of machine self-learning have been completed, improving the management level of smart fishing ports.

Artificial intelligence can also improve the efficiency of quality management and provide convenient services for consumers to purchase goods. For example, Dewu APP took the lead in using AI technology to explore quality consumption assurance, and established the first artificial intelligence inspection and identification production line in China. The AI computing engine is trained by the inspection and identification research accumulated by the Dewu APP for many years and the massive physical commodity data, which can help improve the efficiency of inspection and identification.

Li Chengjie, Vice President and Chief Digital Intelligence Officer of Mengniu Group, introduced that Mengniu has gone through three stages: the digitalization of goods, the digitalization of people and the digital and intelligent transformation driven by artificial intelligence, transforming the organization and production methods, and stimulating new quality productivity. In the digitalization 1.0 stage, Mengniu has achieved digital coverage of the entire chain from milk source, factory to consumer; In the Digital 2.0 stage, Mengniu is deeply engaged in consumer digitalization, and in the current Digital Intelligence 3.0 stage, Mengniu has implemented the "AI-driven double flywheel" strategy, using artificial intelligence to comprehensively support the supply side and the consumer side.

"It is expected that the application scenarios around artificial intelligence and data elements will continue to be released, promote the cultivation, growth and development of new quality productivity in the IT field, and bring new development momentum to all walks of life. At the same time, the deep integration of artificial intelligence and data elements will create new sparks in the industrial, economic and social fields. Su Bo said.

Tao Qing revealed that it is necessary to deepen the integration and application of artificial intelligence technology in the whole process of manufacturing, and greatly improve the intelligent level of R&D, pilot test, production, service, management and other links. For key industries that have a great impact on the national economy, have a strong driving force, and have a good digital foundation, we will carry out special actions for artificial intelligence to empower new industrialization, accelerate the intelligent upgrading of key industries, and improve the intelligence level of key products and equipment such as high-end equipment, key software, and intelligent terminals.

Cultivate and expand the industrial ecology

The development of artificial intelligence in mainland China also faces many problems. According to Wang Peilin's analysis, the degree of standardization and standardization of the mainland artificial intelligence industry is not high, and the lack of a unified standard system in the industry not only affects the interoperability and compatibility of enterprises, but also hinders the standardization and large-scale development of the entire industry. Although the scale of artificial intelligence talent training in mainland China is gradually expanding, it is still facing problems such as the scarcity of high-level leading talents and the loss of basic research talents, and it is urgent to build a more complete multi-level talent training system.

In addition, with the integration of big data and artificial intelligence, how to effectively protect user privacy, ensure data security, and meet the big data needs of AI model training has become an important issue. Cross-border collaboration and resource integration are not close enough, the deep integration and collaborative innovation between various industries and industry, academia, research and application are not sufficient, and the resource sharing and benefit distribution mechanism needs to be improved.

In response to these problems, relevant enterprises have carried out a series of exploration and practice. According to reports, Inspur Cloud took the lead in launching the exploration and practice of new data infrastructure in China, and met the data circulation needs such as secure storage, trusted transmission, and joint modeling of authorized data by building an urban trusted data space and integrating various technical capabilities such as privacy computing, sandbox computing, data authorization, and blockchain, and provided strong technical support for the safe and credible circulation of data elements in various places.

Yang Haiming, chief expert of digital transformation of China Unicom Digital Technology Co., Ltd., introduced that China Unicom Digital Technology has formed an enterprise transformation layout of "China Unicom Digital Technology + Industry Corps + Research Institute", and also played a leading role in the field of network security of China Unicom, and jointly built an ecological chain of external empowerment with partners.

Tao Qing said that it is necessary to improve the support system and cultivate and expand the industrial ecology; Strengthen the guidance of standards, and establish and improve the standard system of artificial intelligence-enabled new industrialization. Optimize the layout of manufacturing innovation centers, industrial technology basic public service platforms and key laboratories in the field of artificial intelligence, strengthen collaborative innovation in the industrial chain, strengthen the construction of artificial intelligence open source ecology, and improve mechanisms for artificial intelligence talent training, security and international cooperation.

Zou Debao, executive deputy general manager of CCID Consulting Artificial Intelligence and Big Data Research Center, proposed that relevant departments should give policy support to meet the needs of artificial intelligence innovation and development to promote new industrialization and create a better industrial environment. (Reporter Huang Xin, Li Hefeng)

Source: Economic Daily