The prevalence of artificial intelligence (AI) increases social concern surrounding unethical consumer behavior in human–AI interaction. Existing research has mainly focused on anthropomorphic characteristics of AI and unethical consumer behavior (UCB). However, the role of algorithms in unethical consumer behavior, which is central to AI, is not yet fully understood. Drawing on social exchange theory, this study investigates the impact of algorithmic discrimination on UCB and explores the interrelationships and underlying mechanisms. Through three experiments, this study found that experiencing algorithmic discrimination significantly increases UCB, with anticipatory guilt mediating this relationship. Moreover, consumers’ negative reciprocity beliefs moderated the effects of algorithmic discrimination on anticipatory guilt and UCB. In addition, this study distinguish between active and passive UCB based on their underlying ethical motivations. This enhances the study’s universality by assessing both types of behaviors and highlighting their differences. These insights extend current research on UCB within the purview of AI agents and provide valuable insights into effectively mitigating losses caused by UCB behaviors, offering improved directions for facilitating AI agents to provide fair, reliable, and efficient interactions for both businesses and consumers.

Artificial intelligence (AI) technology has led to the gradual transformation of interpersonal interactions into human-AI interactions, and the differences between these two types of interactions have attracted widespread attention. Still, scholars have not yet agreed on consumers` responses to AI machine anthropomorphism. This paper explores the relationship between AI machine anthropomorphism and consumers’ usage intention based on the privacy computing theory and finds a U-shaped relationship between AI machine anthropomorphism and consumers’ usage intention, which is mediated by privacy concerns and public self-awareness and positively moderates the relationship. Consumers’ social class also influences their attitudes toward AI machine anthropomorphism, i.e., consumers from high social class show a high degree of AI machine anthropomorphism while those from low social class show favour for a low degree of AI machine anthropomorphism. This article enhances existing research by examining consumer characteristics through a granular lens.

The relationship between firm performance feedback and technology innovation has been studied extensively, but limited attention has been paid to factors that mediate this relationship. We examine how a parent firm addresses an affiliate’s performance level and its influence on the affiliate’s technology innovation. Integrating the behavioral theory of the firm and the concept of parent functions, we argue that the parent firm addresses the affiliate’s performance level in two ways: hierarchy management and resource allocation. Specifically, unlike the case of outperforming affiliates, the parent firm facilitates the technology innovation of underperforming affiliates through hierarchy management and resource allocation. Regression analyses of 2010–2020 data of listed affiliates belonging to Chinese business groups provide strong evidence supporting our conjecture. Our study sheds light on the importance of considering the parent’s influence when affiliates adopt technology innovation in light of its performance feedback.