能量转换视角下人工智能关键核心技术产业化路径解析

人工智能是驱动新一轮科技革命和产业变革的中坚力量，探究人工智能关键核心技术产业化路径对于我国数字经济发展、数字化转型和创新型国家建设具有重要意义。基于能量转换视角，探讨人工智能技术产业化路径形成机理，识别人工智能关键核心技术产业化路径并对其进行解析。研究发现：①人工智能技术产业化过程存在能量转换，技术创新能量、催化孕育能量、商业转化能量和业态塑造能量共同构成能量转换的核心环节，决定技术产业化路径具体过程；②人工智能关键核心技术包括机器学习、计算机视觉、自然语言处理等八大技术领域，不同属性核心技术构成相应技术集群，形成以识别、交互和执行为主题的技术产业化路径；③技术集群属性是影响技术产业化路径的关键因素。从能量视角对人工智能关键核心技术进行有效识别和归纳，为我国科学推进人工智能技术产业化进程提供相应思路启示与政策响应。

An Analysis of the Industrialization Path of Key Core Technologies of Artificial Intelligence from the Perspective of Energy Conversion

It is of strategic significance to explore the industrialization path of key core technologies of artificial intelligence for developing the digital economy, promoting digital transformation and building an innovative country in the background of new round of scientific and technological revolution and industrial revolution. Artificial intelligence technology has made great progress and some key core technologies have gradually broken through application threshold and entered the stage of industrial application. However, industrialization of artificial intelligence technology is still in the early stage of development, so the industrialization path is not completely clear, and the commercial value of artificial intelligence is difficult to be fully released. The academic circles have conducted a series of discussions on technology industrialization using patent data, literature data, and R&D data from the perspective of technology transfer and transformation of scientific and technological achievements, but the existing research can′t meet the actual needs of the highly complex and interrelated artificial intelligence technology industrialization. This paper classifies the complex energy flow process of ecosystem operation from the perspective of energy conversion, hoping to provide new perspective for research on the industrialization path of key core technologies of artificial intelligence.#br#This paper selects data from 2000 to 2020, and a total of 7 811 patent documents related to artificial intelligence were retrieved and sorted out in the DII database. Based on the specific co-occurrence situation with reference to the standard system framework in the White Paper on Artificial Intelligence Standardization (2018), the 239 technical keywords of artificial intelligence are further extracted, which is adapted to the existing clustering results, and finally this paper draws the artificial intelligence core technology network map.#br#The results show that the industrialization path of artificial intelligence core technology has characteristics of energy conversion. Technological innovation energy, catalytic energy, commercial conversion energy and business shape energy constitute the core link of energy conversion, which determines specific process of the path of technology industrialization. The key core technologies of artificial intelligence include eight technical fields such as machine learning, computer vision, and natural language processing. The core technologies of different attributes constitutes the corresponding technology clusters, forming a technological industrialization path with the themes of "identification", "interaction" and "execution". The attribute of technology cluster is a key factor that affects the path of technology industrialization.#br#This paper reveals that the industrialization of artificial intelligence core technology is not equal to a simple technology transfer and routine replication, but a cumulative practice process of a series of complementary innovations and specialized technology systems, as well as a process of energy gathering and transformation. Managers should focus on technology clusters composed of multiple core technologies, and promote collaborative innovation among technologies by innovating R&D models and optimizing internal structures. It is necessary to keep exploring and cultivating powerful technological innovation energy, catalytic breeding energy, commercial transformation energy and business format shaping energy, so as to support the implementation of industrialization path of core artificial intelligence technologies. Managers should expand energy conversion channels from multiple perspectives to improve energy conversion efficiency. It is necessary to explore more novel and reasonable energy conversion methods such as technology incubation and technology empowerment from the perspective of technological innovation to maximize the overall value. It is also essential to improve the rationality of value matching and innovative human-machine collaboration methods from the perspective of scene application.#br#