Abstract
The rise of conversational AI agents has gradually changed the way users shop online. Technologies like Natural Language Processing (NLP) and Artificial Intelligence (AI) are likely to accelerate this trend further. Users, however, express a preference for human interaction and exhibit reluctance in engaging with AI agents due to a prevailing perception of their impersonality and absence of the human touch. To stimulate a more human-like interaction between AI agents and users, humanoid embodiment of the agents may result in evoking social presence in e-tail platforms and influencing user perception. This study investigates the interplay of user acceptance, social presence, and trust in the context of conversational AI agents (n = 158), extending the Technology Acceptance Model. The findings suggest that social presence emerges as a critical factor, significantly shaping user attitudes by fostering a sense of social interaction. Transparent agent recommendations contribute positively to trusting beliefs. Concerns regarding data privacy persist, highlighting the need for robust data protection measures. This study also offers theoretical contributions and practical insights for e-tail businesses.