人工智能的绿色发展效应:技术赋能和结构优化

Green Development Effects of Artificial Intelligence:Technological Empowerment and Structural Optimization

人工智能作为中国抢占科技战略高地的重要抓手,能否显著释放绿色发展效应备受关注,也是理论研究亟待探讨的重要命题。融合非连续性技术创新理论意涵,利用超效率EBM模型测算中国省级绿色发展效率,系统考察人工智能的绿色发展效应。研究发现:(1)人工智能会直接赋能绿色发展,赋能效果呈边际效应递增的非线性特征。以地区高校平均科技产出和《中国制造2025》政策冲击为工具变量强化内生性控制后,结论仍成立。(2)人工智能主要通过技术创新效应和产业结构优化效应提升绿色发展效率。(3)中国转型经济背景下资源禀赋和要素密集度差异使得人工智能绿色发展效应释放具有异质性,要素市场化配置水平、科技人力资源积累程度越高,技术、资本要素越密集,人工智能对绿色发展效率的赋能效果越强。(4)拓展分析发现,人工智能对提升绿色发展效率具有时滞性,且依托国内技术来源的人工智能,比依托国外技术来源的人工智能对绿色发展的时滞效应更明显。因此,应把握人工智能高速发展的战略契机,在开放融合创新中塑造技术竞争新优势,进而带动绿色发展转型。

The prevailing“black development”model in China,characterized by high energy consumption,pollution,and low productivity,has contributed to rapid economic growth.However,it has come at a substantial environmental cost,jeopardizing long-term economic benefits and well-being.To address these challenges and achieve sustainable growth,the potential of artificial intelligence(AI)as a key tool for seizing the strategic high ground in science and technology is of paramount interest.This paper investigates whether AI can leverage its technological innovation advantage to promote green development,guide China’s economic transformation,and foster a symbiotic relationship between green,low-carbon practices,and economic development.Using a panel dataset of 30 Chinese provinces(excluding Tibet and Hong Kong,Macao,and Taiwan)spanning 2010 to 2020,we integrate theoretical insights into discontinuous technological innovation.Employing the super-epsilon-based measure(super-EBM)model and the entropy value method,we assess the green development efficiency and the comprehensive AI development index in China,respectively.Through empirical analysis,we uncover the following key findings:(1)AI directly empowers green development,with the empowerment effect exhibiting non-linear marginal effects.Our results withstand rigorous endogeneity controls using regional university science and technology output and the policy shock of“Made in China 2025”as instrumental variables.(2)AI enhances green development efficiency predominantly through its technological innovation and industrial structure optimization effects.(3)The release of AI’s green development effect varies due to differences in resource endowment and factor intensity within China’s transition economy.Specifically,a higher level of factor market allocation,an increased accumulation of scientific and technological human resources,and greater intensity in technology and capital factors reinforce the impact of AI on green development empowerment.(4)Further analysis reveals a time lag in AI’s capacity to enhance green development efficiency,with AI relying on domestic technology sources exhibiting a more pronounced time lag effect than AI relying on foreign technology sources.Unlike previous research that has primarily focused on the economic development effects of AI and the potential for intelligent technology to drive green development,this paper fills the gap by offering a systematic exploration of the release of AI’s green development effects and its realization path.Specifically,we analyze AI’s direct empowerment effect on green development through intelligent production methods,industrial chain changes,and economic system optimizations,while also investigating its underlying mechanisms of technological innovation and industrial structure optimization.Furthermore,we introduce a panel threshold regression model to examine the non-linear characteristics of AI-enabled green development and analyze the moderating role of factor market allocation and human resources in science and technology.This allows us to capture regional variations in AI’s green development effect release based on factor intensity status.Finally,by adopting a dynamic perspective,we identify and analyze the time lag effect of AI on green development between domestic and foreign technology sources.In conclusion,this paper sheds light on the green development effect and dynamic mechanisms of AI,offering valuable policy insights for government authorities.These insights can inform decisions related to green development pathways,the cultivation of AI’s technological advantage,economic structure optimization,and the transformation of production and lifestyle.