Love deepens hatred! AI immorality and users’ love-becomes-hate phenomenon

With the widespread application of artificial intelligence in the service industry, understanding consumers’ changing attitudes towards AI is crucial. This study explores the impact of AI immorality on the shift from “love-becomes-hate” in consumers, using three scenario-based experiments and framed within the theoretical perspective of the Relationships Breakdown model. In Study 1, cognitive dissonance and perceived betrayal were found to mediate the relationship between AI immorality and the “love-becomes-hate” transition. Studies 2 and 3 further demonstrated that technological mindfulness positively moderated the relationship between AI immorality and cognitive dissonance (Study 2). Specifically, consumers with high levels of technological mindfulness experienced more pungent cognitive dissonance when confronted with AI immorality. Additionally, social judgment positively moderated the relationship between AI immorality and perceived betrayal, with negative social judgment exacerbating the degree of perceived betrayal (Study 3). Overall, AI immorality can lead to a “love-becomes-hate” phenomenon among consumers. This study theoretically fills the gap in the mechanism of AI immorality impact on consumer psychology, expands the application field of the relationship breakdown model, and deepens the understanding of consumer emotional transformation; in practice, it provides enterprises, AI developers, and marketers with an important guide for risk control of AI ethical development, which helps to maintain good human-machine relationship and brand image.