THE INFLUENCE OF CHATBOT ANTHROPOMORPHISM AND TRUST ON CHATBOT ADOPTION INTENTION IN E-COMMERCE

Abstract

As Artificial Intelligence (AI) continues to reshape customer service in e-commerce, chatbots are becoming pivotal in enhancing user experiences and streamlining customer–business interactions. With AI increasingly taking over jobs traditionally performed by humans, understanding how consumers perceive and adopt AI-driven systems becomes critical. This study examines the role of Chatbot Anthropomorphism (CA) in fostering Chatbot Trust (CT), and how trust subsequently drives Chatbot Adoption Intention (CAI). It also investigates whether Artificial Intelligence Anxiety (AIA) moderates this relationship. Grounded in the Stimulus–Organism–Response (S– O–R) model, data were collected from 300 e-commerce users via an online survey and analyzed using Structural Equation Modeling (SEM). The measurement model demonstrated an excellent fit (SRMR = 0.041; RMSEA = 0.046; CFI = 0.982; TLI = 0.974), and all constructs showed satisfactory reliability and validity (CR ≥ 0.70, AVE ≥ 0.50). Results confirm that CA significantly enhances CT (β = 0.610, p < 0.001), and CT strongly predicts CAI (β = 0.680, p < 0.001). However, AIA does not moderate the CT–CAI relationship (β = –0.010, p = 0.821), with moderation tested via interaction term in multiple regression. This suggests that anxiety may play a diminished role in contexts where trust is already high. These findings highlight that incorporating humanlike cues, such as warmth, responsiveness, and personalization, into chatbots is critical for building trust and promoting adoption. For managers, the study suggests prioritizing trust-building design features while recognizing that AIA may not be a major barrier for digitally confident user groups, thereby refining strategies for chatbot deployment in e-commerce.