Background and Purpose. In today's retail sector, AI chatbots have played a crucial role by providing 24/7 customer service assistance, boosting sales through prompt and accurate responses to customers' questions, and offering personalized product recommendations based on customers' preferences (Ashfaq et al., 2020). According to Marketsandmarkets.com, the global chatbot market size is expected to increase from $5.4 billion in 2023 to $15.5 billion in 2028. Despite the significant impact of AI chatbot technology on the retail industry, the lack of studies has impeded our understanding of consumers' antecedents (reasons for and reasons against factors) and consequences (willingness to buy and eWOM) regarding attitudes toward using AI chatbots, particularly in the context of apparel retail. To address this gap, this study aims to develop and test a conceptual model of the potential antecedents and consequences of consumers' attitudes toward using AI chatbots. Conceptual Framework and Hypotheses. This study integrated Westaby's Behavioral Reasoning Theory (BRT) (2005) by incorporating three factors of service quality (responsiveness, reliability, and assurance), along with five barriers from the Innovative Resistance Theory (IRT) (Ram & Sheth, 1989) (usage, risk, value, image, and tradition barriers) to develop a conceptual framework to understand apparel consumers' AI chatbot usage behaviors. Hypotheses were formulated regarding the impact of reasons for factors (a) responsiveness, (b) reliability, and (c) assurance, as well as reasons against factors (a) usage barrier, (b) risk barrier, (c) value barrier, (d) image barrier, and (e) tradition barrier on consumers' attitudes toward using AI chatbot services. These attitudes, in turn, were hypothesized to influence their willingness to purchase apparel with the assistance of AI chatbots and electronic word-of-mouth (eWOM). Additionally, this study hypothesized whether technology familiarity moderates the relationships between all three reasons for factors, all five reasons against factors, and attitudes toward using AI chatbots. Methodology and Data Analysis. Data were collected from approximately 700 participants through Amazon Mechanical Turk (MTurk), an online panel. After careful screening, a total of 632 participants were deemed eligible for statistical analysis. Among the participants, 65% were male, with the majority (32.1%) aged between 26 and 30. A total of 58 measurement items were adapted from previous studies (e.g., Chaoualia & Souiden, 2019; Leong et al., 2020; Olya et al., 2021) and were assessed using a 5-point Likert-type scale. Anderson and Gerbing's (1988) twostep approach was employed using Mplus version 8 to establish both measurement and structural models. Confirmatory factor analysis (CFA) was first employed to validate the measurement model. Subsequently, path analysis was performed to test all hypothesized relationships. The hypothesized relationships were tested based on two separate structural models: one focusing on the main effects and the other on the moderating effects.