Abstract：In the epoch of the digital economy, the swift progress and iterative evolution of artificial intelligence (AI) technology have substantially transformed income distribution structures. Nevertheless, extant research mainly concentrates on the influence of AI technology on income inequality at the urban and enterprise levels, with a relatively scant focus on income inequality at the county level.
This research, firmly rooted in Romer’s knowledge spillover theory, utilizes balanced panel data from 1,607 counties in China over the time span from 2010 to 2022. It delves into the non-linear relationship and the underlying mechanisms between AI technology and income inequality at the county level, with a particular emphasis on two dimensions: the urban-rural income gap and regional income disparities. The core research results are as follows: (1) The non-linear effect of AI technology on both the urban-rural income gap and regional income disparities manifests as an inverted U-shaped curve. Initially, it widens income inequality, and then it narrows. At present, AI technology has, on the whole, aggravated income inequality at the county level. (2) Mechanism examinations disclose that AI technology exerts its influence on income inequality mainly through channels such as industrial structure upgrading, employment levels, and population mobility. Moreover, the degree of innovation and entrepreneurship activities can act as a moderating variable. (3) Heterogeneity analysis demonstrates that the non-linear impact of AI technology on income inequality varies markedly depending on factors like economic development, educational attainment, government support, and unique county-level characteristics. (4) Extended analysis indicates that although AI technology can significantly propel common prosperity at the county level, it is of great importance to continuously improve the digital literacy of county residents, especially those in rural areas, in order to close the internal digital divide.
In contrast to the extant literature, this study proffers three principal contributions. First, taking into account the real-world conditions of county-level development and the national strategic configuration of AI, this research analyzes the impact of AI technology on intra-county income inequality. This endeavor furnishes novel perspectives for fostering integrated urban -rural development and common prosperity at the county level. It acknowledges the unique economic, social, and geographical features of counties, which are often overlooked in broader-scale studies, and thus fills a crucial gap in understanding how AI can be harnessed for local-level inclusive growth. Second, given that in its nascent stages, AI technology is highly dependent on capital, technology, and talent, and its development is characterized by urban-rural and regional disparities, this study constructs a non-linear analytical framework to investigate the relationship between AI technology and county-level income inequality. It delves deeper into the mechanisms by which industrial structure upgrading, employment levels, and population mobility shape income inequality, and also explores the moderating function of innovation and entrepreneurship activities. By doing so, it uncovers the complex interplay of various factors that influence the distributive effects of AI at the county level, which has not been comprehensively explored before. Third, while existing research predominantly centers on the development of AI technology at the provincial, urban, or enterprise levels, there has been a dearth of prior studies quantifying the development level of AI technology within counties. This study bridges this void by leveraging Python to extract the distribution of AI enterprises in counties from relevant platforms, followed by meticulous manual data processing, cleaning, and matching procedures. These efforts significantly enhance the authenticity and reliability of the data.
Overall, this study offers both theoretical underpinning and practical guidance for the development of AI technology, the strategic layout of AI-related industries, and the establishment of inclusive development trajectories in Chinese counties.
高静 李丹 陈峰 冯浩. 人工智能技术与县域收入不平等：“抑制剂”还是“加速器”[J]. 浙江大学学报(人文社会科学版), 0, (): 1-.
Gao Jing Li Dan Chen Feng Feng Hao. Artificial Intelligence Technology and County Income Inequality: “Inhibitor” or “Accelerator”?. JOURNAL OF ZHEJIANG UNIVERSITY, 0, (): 1-.
