AI Paradox: The Negative Impact of AI Usage on Public Servants' Well-being

• In Event: AI in the Public Sector: Cross-National Evidence on Bureaucracy, Procurement, and Organizational Change

Abstract

In the last decades, Artificial Intelligence (AI) has been adopted to promote the efficiency and capacity of government in terms of decision-making and public service delivery. While AI usage may bring positive impact to citizens’ satisfaction, it may have unintended effects on public servants. This is an important question because public servants’ well-being can affect their productivity. Scholars have identified that AI adoption in the public sector will decrease the perceived discretion of public servants (Møller, 2025; Bullock, 2019; Young, Bullock, and Lecy, 2019). Despite these effects, there is a lack of causal evidence on how AI adoption affects public servants’ well-being and relevant mechanisms. Specifically, we are interested in: (1) what is the impact of AI usage in public sectors on public servants’ well-being; (2) what are the mechanisms, and (3) how does AI usage in public sectors affect frontline workers differently than non-frontline workers? By combining city-level panel data and a representative survey dataset from the China Family Panel Studies (CFPS) spanning 2010 to 2018 (N=5,179), we employ a staggered Difference-in-Differences (DID) approach to explore the causal effect of AI usage in the process of smart city reform on public servants’ well-being. We find robust evidence that the implementation of smart city reduces public servants’ well-being by 3.64% in terms of happiness and by 4.64% in life satisfaction. Further analysis indicates two channels of this causal relationship: increased workload and robot anxiety. We find that AI usage amplifies public servants’ workload, diminishes job satisfaction, decreases mental health, and heightens the likelihood of mild depression. It also erodes their confidence in their future lives, leading to surges in purchases of supplementary commercial insurance. Notably, these effects are more pronounced in frontline workers in public sectors.

Authors

• Shuping Wang, Syracuse University
  Presenting Author

• Andong Zhuge, Fudan University
  Non-Presenting Co-Author