摘  要
　　人工智能逐渐成为新一轮科技革命的重要力量，对提升企业全要素生产力、推动经济社会发展具有重要意义。党的二十大报告强调要推动战略性新兴产业发展，实现我国产业发展新态势，并促进产业集群融合，构建新一代信息技术、人工智能等新增长动能。人工智能作为新一轮科技和产业革命的重要动力，能够为企业发展创造出更多的新机会，是推动企业全要素生产率提升的重要抓手。为此，在加快现代化经济体系背景下，深入剖析人工智能技术应用对企业全要素生产率影响效应，探究其影响机理，对推动企业智能化转型，实现高质量发展具有重要意义。通过政策梳理和文献分析，探究人工智能和企业全要素生产率提升现状，以政策和实际需求为导向，在加快建设现代化经济体系的背景下，借助新一代人工智能发展实现现代化产业体系建设，剖析企业全要素提升的难题，发挥人工智能通用技术的新动能。
　　本文基于劳动力结构理论，构建人工智能影响企业全要素生产率的理论分析框架，并选取2011-2022年沪深A股上市公司数据，通过机器学习方法构建人工智能指标，并运用计量分析方法实证检验了人工智能技术对企业全要素生产率的影响效应和作用机制。主要得出以下结论：（1）企业人工智能的应用能够显著提升企业全要素生产率，并通过一系列稳健性检验，包括替换被解释变量、倾向性得分匹配、工具变量以及PSM_DID；（2）企业劳动力结构和创新能力在人工智能对企业全要素生产率的影响中发挥中介作用。人工智能显著提升劳动力结构，对研发人员投入有较好的促进效果。人工智能显著提升企业创新能力，对专利获取起到促进作用。研究结果表明人工智能通过提升劳动力结构和企业创新能力进而提高企业全要素生产率水平；（3）对于不同类型和不同行业的企业，人工智能对企业全要素生产率的影响效果不同。具体来说，企业异质性视角下，国有企业应用人工智能技术对企业全要素生产率的提升效果要优于非国有企业。行业异质性分析视角下，技术密集型行业企业影响效果明显，资本密集型行业企业影响效果要优于劳动密集型企业；（4）基于供应链溢出视角，人工智能对企业全要素生产率的影响效果在供应链上存在差异，供应链向后溢出的作用效果更为明显。
　　
　　关键词：人工智能 企业全要素生产率 劳动力市场 供应链溢出
Abstract
　　Artificial intelligence is gradually becoming an important force in the new round of scientific and technological revolution, and is of great significance in enhancing the total element productivity of enterprises and advancing economic and social development. The report of the 20th Party Congress emphasizes the need to advance the development of the strategic emerging industries, to realize the new dynamics of China's industrial development, and to promote the convergence of industrial clusters, and to build a new generation of information technology, artificial intelligence, and other new growth dynamics. Artificial intelligence, as an essential driving power of the new round of scientific and technical and industrial revolutions, can create more new opportunities for enterprise development, and is an important hand in promoting the improvement of the total factors productivity of enterprises. For this reason, in the context of accelerating the modernization of the economic system, in-depth analysis of the effects of the application of artificial intelligence technology on the total elemental productivity of enterprises, and investigate the effect mechanism, is of major significance to driving the intellectual transformation of enterprises and achieve the development of high quality. Through policy combing and bibliographic analysis, we explore the status quo of artificial intelligence and enterprise total elemental productivity enhancement, oriented by policy and actual demand, in the context of speeding up the establishment of a modernized industrial system and realizing the construction of a new-generation industrial framework with the help of new-generation artificial intelligence development, we analyze the difficulties of enterprise total elemental enhancement, and give play to the new kinetic energy of the general-purpose technology of artificial intelligence.
　　Based on the working force architecture theory, this essay builds the framework of a theoretical analysis of the effect of artificial intelligence on the total elemental productivity of enterprises, and picks the data of A-share listed companies in Shanghai and Shenzhen from 2011 to 2022, constructs the indicators of artificial intelligence through machine learning method, and empirically tests the effect and mechanism of the influence of artificial intelligence on the total elemental productivity of enterprises by applying the method of econometric analysis. The following main conclusions are drawn: (1) the application of enterprise AI can dramatically enhance enterprise total factors productivity, and passes a series of firmness tests, including replacing the explanatory variables, propensity score matching, instrumental variables, and PSM_DID; (2) the enterprise labor force structure and innovation capacity play a mediating role in the impact of AI on enterprise total parameters productivity. Artificial intelligence significantly improves labor force structure and has a better promotion effect on R&D personnel investment. Artificial intelligence remarkably enhances the firm's innovation ability and has a facilitating effect on patent acquisition. The results of the study suggest that AI improves the level of total enterprise gross factor productivity by enhancing the workforce structure and corporate creativity; (3) For different types of companies and different professions, the effect of AI on corporate total element productivity is different. Especially, under the viewpoint of firm heterogeneity, the effect of applying AI technology on the total elemental productivity of state-owned firms is better than that of non-state-owned firms. Under the viewpoint of industry heterogeneity analysis, the influence effect of enterprises in skill-intensive industries is obvious, and the influence result of enterprises in capital-intensive industries is better than that of labor-intensive enterprises; (4) based on the supply chain spillover perspective, the influence result of AI on the total factor productivity of the enterprise differs in the supply chain, and the effect of the supply chain's backward spillover effect is more obvious.

　　Keywords：Artificial Intelligence；Firm total factor productivity；Labor market；Supply chain spillovers

　　67