Unveiling coping mechanisms in marketplace discrimination: The allure of artificial intelligence recommendations

Abstract Despite artificial intelligence's (AI) increased efficiency and accuracy in many contexts, algorithm aversion, that is, people's biased preference for human recommendations over those of algorithms, is a well‐documented phenomenon. In this research, we show a reversal of the algorithm aversion phenomenon, referred to as algorithm appreciation, in the prevalent context of marketplace discrimination. Specifically, the current research documents people's increased propensity to rely on AI‐based recommendations over those proposed by human counterparts in the aftermath of marketplace discrimination. Such an increased preference happens because it serves as a coping strategy for consumers who have faced discrimination in the marketplace from other human actors. The results of a series of three lab studies and one field study provide consistent support for the proposed effect and document the underlying psychological mechanism driving this effect through perceived embarrassment. Using a moderated‐mediation model, we identify a boundary condition of the effect by demonstrating that the focal effect, that is, algorithm appreciation, remains valid under public consumption but diminishes under private consumption. Employing the natural setting of the field, we replicate our findings with actual consumers making real choices. Our findings have important implications (e.g., integrating AI‐driven recommendation systems into firms' platforms in sectors susceptible to marketplace discrimination and developing ethical guidelines for AI systems) for managers and companies.