The impact of generative AI shopping assistants on E-commerce consumer motivation and behavior: Consumer-AI interaction design

Generative AI (Gen AI) shopping assistants have been extensively studied—both theoretically and empirically—for their impact on consumer experiences in developed-country e-commerce platforms. However, cultural, economic, and technological differences may constrain applicability in developing-country contexts. This study examines both the “lights and shadows” of Gen AI shopping assistants in developing countries, focusing on how these assistants shape the consumer motivation–behaviour process. We conduct a review from the perspectives of human–computer interaction (HCI), cognitive psychology, and marketing to assess the current state and challenges of Gen AI shopping assistants. Based on this review, we have developed the Motivation–Expectation Management Model (MEMM) to complete the following cycle: Motivation → HCI → Expectation Confirmation → Satisfaction and Repurchase → (feedback to) Motivation. We then collect data from consumers using the “Taobao Wenwen” Gen AI shopping assistant within the Taobao app in China and employ a mixed-methods approach to test the significance, importance, and necessity of the MEMM. (1) Extrinsic motivation exerts a greater influence on personalzation and UX than intrinsic motivation; (2) Mediation chains linking user experience, expectation confirmation, satisfaction, and repurchase intention are significant, with some relationships supported across significance, importance, and necessity analyses, while others are only partially consistent. In summary, MEMM provides both theoretical and empirical grounding for studying Gen AI shopping assistants in developing-country contexts, helps elucidate the consumer–Gen AI interaction mechanisms at play, and offers strategic guidance for sustaining a continuous cycle of interaction optimisation in e-commerce markets.