Focus on basic research on shortcomings and business landing problems.

Chen Siqi, a reporter from Southern Finance and Economics, reported that Shenzhen reported that my country's first special artificial intelligence industry legislation- "Regulations on the Promotion of Artificial Intelligence Industry in Shenzhen Special Economic Zone" (hereinafter referred to as the "Regulations") was officially announced and planned to be implemented from November 1 this year.

On September 5th, the Seventh Standing Committee of the People's Congress of Shenzhen announced that the "Regulations on the Promotion of the Artificial Intelligence Industry of Shenzhen Special Economic Zone" passed on August 30 and implemented from November 1. The "Regulations" proposes an innovative product access system. For low -risk artificial intelligence products and services that have not yet formulated standards for national and local standards but meet international advanced product standards or standardized, they are allowed to conduct first trials through testing, testing, and pilots.

According to data from the new generation of artificial intelligence Development Research Center of the Ministry of Science and Technology, in 2021, the scale of my country's artificial intelligence core industry exceeded 400 billion yuan, and more than 3,000 enterprises. At present, there is no special artificial intelligence industry legislation at the national level. Shenzhen's artificial intelligence industry has a good foundation, with more than 1,300 artificial intelligence -related enterprises, and the number of companies ranks second in the country.

The official website of the Standing Committee of the Shenzhen Municipal People's Congress announced that the "Regulations" is an important legislation of our city in the emerging field. It promotes the development of the industry as the starting point. Through a series of system design, it promotes and exerts artificial intelligence as the new round of scientific and technological revolution and industrial changes. The important role of the core driving force provides the protection of the rule of law in order to promote the development of the artificial intelligence industry in Shenzhen, and also explores the relevant legislation at the national level.

The first legislation clarify the concept of artificial intelligence and industrial boundary

《条例》根据国务院《新一代人工智能发展规划》等相关文件，在借鉴国内外一系列相关研究成果的基础上，从技术角度明确人工智能是“利用计算机或者其控制的设备，通过感知环境、 Get knowledge and derive interpretation methods to simulate, extend or expand human intelligence. "

At the same time, the Regulations clarify the boundaries of the artificial intelligence industry, and the core industries such as the research, development and production, system application, and integrated services related to software and hardware products related to artificial intelligence, as well as artificial intelligence technology in people's livelihood services, social governance, economic development, etc. Related industries formed by fusion applications in various fields are included in the field of artificial intelligence industry.

The lack of complete artificial intelligence industry statistical monitoring system is one of the issues that are generally reflected by enterprises and industry organizations that urgently need to solve. At present, artificial intelligence has not been included in the category of the national economic industry, and the national level lacks a unified artificial intelligence industry statistics. The standards and statistical calibers of the number, output value and even industrial scale of artificial intelligence enterprises are also different.

For reference, in February this year, Chengdu took the lead in issuing the "Classification Classification of the Artificial Intelligence Industry of Chengdu (Trial)", which clearly defined the artificial intelligence industry by artificial intelligence software development, artificial intelligence hardware and equipment manufacturing, artificial intelligence services and artificial intelligence Fusion application composition.

On the basis of clarifying the connotation and extension of artificial intelligence in the "Regulations", the "Regulations" stipulates that the statistical classification standards of artificial intelligence industry should be established, and the statistical classification catalog of the artificial intelligence industry should be established.

Focus on basic research shortcomings and business implementation problems

It is understood that the artificial intelligence industry chain can be divided into basic layers, technical layers and application layers. According to the "Artificial Intelligence Standardized White Paper (2021 Edition)" released by the China Institute of Electronic Technology Standardization, my country's artificial intelligence has achieved certain results in terms of technology and applications, but there are still obvious shortcomings in key areas such as smart chips and smart sensors. In terms of algorithms, chips, terminals, and scene applications, it relies heavily on foreign deep learning framework ecology.

Shenzhen has more than 1,300 artificial intelligence -related enterprises, and the number of enterprises ranks second in the country. According to the previous statistics of the Shenzhen Artificial Intelligence Industry Association, as of the end of 2021, 19.8%of Shenzhen artificial intelligence enterprises were at the basic layer, 16.0%of the enterprises were at the technical layer, and 64.2%of the enterprises on the application layer were concentrated on public safety and intelligent manufacturing In areas such as smart homes and smart transportation, the contradictions of re -application and light foundation are quite prominent.

In view of this, the "Regulations" pointed out that it will encourage colleges, scientific research institutions, enterprises, and other organizations to carry out long -term and interdisciplinary artificial intelligence basic research. At the same time, the scientific and technological innovation department shall focus on the key core links of artificial intelligence, establish a key core technical research mechanism with market demand -leading and deep integration of government, property, research, and research, and build a supporting policy system covering the full cycle of artificial intelligence core technologies.

"Difficult to land" is another bottleneck restricting the development of artificial intelligence. Ding Ning, deputy dean of Shenzhen Artificial Intelligence and Robotics Research Institute, previously pointed out in an interview with a reporter from Southern Finance and Economics that the incompatibility of the industry -university -research evaluation system is a major obstacle to the commercialization of technology.

On the one hand, the evaluation system of the industry and enterprises is determined by the market. Most enterprises pursue ready -made mature technology, and it is difficult to invest sufficient time, funds, and resources to make technology full iterative, trial and error, and cost control; Primitive innovation, academic achievements, and academic value are evaluated. The transformation process from academic to industry lacks a corresponding evaluation system. Therefore, college professors often stop at the publication of the paper and promote the motivation of academic innovation achievements in the market.

In response to the above issues, the "Regulations" on the one hand clearly establishes an incentive mechanism that is conducive to promoting the transformation of scientific and technological achievements, and promotes various types of innovative subjects to carry out scientific and technological achievements; on the other hand, it will establish scientific and technological talents guided by innovative value, abilities, and contributions. Evaluation system, the economic and social benefits of the transformation and creation of scientific and technological achievements are an important evaluation factors for the evaluation of artificial intelligence talent titles. It is worth noting that the access registration system that has previously implemented some artificial intelligence products has occurred with the development of new technologies and new products. For example, the new crown epidemic has accelerated the development of medical intelligence. Medical AI is considered to be the most likely to be the first to realize commercialization in artificial intelligence applications. Relatively delay.

In this context, the "Regulations" clearly stipulates that the innovative product access system, on the basis of hierarchical classification supervision, stipulates that low -risk artificial intelligence products and services that have not yet formulated standards but meet the standards of international product standards or standards are allowed. Carry out the first test through testing, testing, pilot and other methods. In response to the current field of smart medical care, the "Regulations" also made specialized regulations to explore the establishment of a registered medical device clinical trial ethical review and mutual recognition mechanism, and encourage medical institutions to use assistance decision -making, video or data processing, medical data analysis and mining, and Medical assistant and other artificial intelligence products and services.

Establish an artificial intelligence ethics governance mechanism

"At present, the main risks brought by artificial intelligence are not necessarily immature in technology, but the rapid development of AI technology on ethical and social and cultural impacts have surpassed our cognition." The 2022 world artificial intelligence governance recently held recently At the forum, Hong Zhiwei, a member of the Social Science and Humanities and Sciences Projects of the United Nations Education Organization in China.

In recent years, the rapid development of artificial intelligence has brought about problems such as precise push, financial fraud, and "information cocoon houses". The most common is that some technology companies have grasped consumers' online browsing, shopping travel and other data. Through correlation analysis, they will deliver commercial advertisements in a targeted manner, raising the prices of mature customers, and "big data killing".

In response to the above -mentioned risk performance, the Regulations clarify the bottom line of artificial intelligence products and services, and stipulate that relevant organizations or individuals are prohibited from endangering national security and social public interests, infringing personal privacy and personal information rights; Use algorithm technology to implement acting of consumer rights such as price discrimination or consumer fraud based on user preferences, payment capacity, and consumer fraud.

In March of this year, the General Office of the Central Committee of the Communist Party of China and the General Office of the State Council issued the "Opinions on Strengthening the Governance of Scientific and Technological Ethics". In response to the inadequate institutional mechanisms of scientific and technological ethics governance, incomplete systems, and uneven field development, the government should be impressed to improve the government Science and technology ethics management system and compact the responsibility of science and technology ethics management of the main body of innovation.

The "Regulations" also clearly stipulate the responsibilities of government, industry organizations, enterprises and other related entities. At the government level, we must improve the supervision mechanism in the field of artificial intelligence, establish the Shenzhen Artificial Intelligence Ethics Committee, and be responsible for formulating ethical security specifications and supervision and implementation of the field of artificial intelligence; industry organizations need to carry out industry self -discipline management, guide and urge the industry operators to compete in accordance with the law to compete in accordance with the law. To maintain the market competition order; enterprises should include compliance with ethical security specifications into the professional specifications of the unit, and incorporate ethical security risk education, laws and regulations into the training of artificial intelligence practitioners in the unit.

According to the "Shenzhen New Generation Artificial Intelligence Development Action Plan (2019-2023)", by 2023, Shenzhen will strive to build a leading area for the first -class artificial intelligence application to cultivate more than 20 technological innovation capabilities at the leading level in China Enterprises, the size of the core industry of artificial intelligence exceeded 30 billion yuan, driving the scale of related industries to 600 billion yuan. In this context, the promulgation and subsequent landing of the Regulations will provide a solid protection of the rule of law for Shenzhen to cultivate a high -quality and efficient industrial development environment and a good industrial development environment and a good industrial ecology, and promote the development of the artificial intelligence industry.

- END -