Assistant Professor, University of Niagara Falls, Canada
Generative artificial intelligence is reshaping e-commerce by enabling consumers to seek product advice, compare brands, and receive personalized recommendations through conversational tools such as ChatGPT and Claude. Unlike conventional online shopping websites, generative-AI assistants do not merely present products or facilitate transactions; they interpret consumer needs, filter options, summarize product information, and influence how brands are framed during the purchase journey. This shift raises an important question about how the quality of AI-mediated shopping services should be evaluated. This presentation focuses on the E-S-QUAL model and recent AI-enabled service-quality research to examine the applicability of existing service-quality dimensions to generative-AI shopping environments. Parasuraman, Zeithaml and Malhotra (2005) established E-S-QUAL as a 22-item measure of electronic service quality across four dimensions such as efficiency, fulfilment, system availability, and privacy (Journal of Service Research). Building on this foundation and on subsequent empirical work that has extended service-quality measurement to AI contexts notably reported five-dimensional AI-enabled service quality scale spanning reliability, responsiveness, personalization and intelligence, transparency and trust, and empathy and emotional intelligence. The discussion highlights that while efficiency, fulfilment, system availability, and privacy remain relevant, the distinctiveness of generative-AI-mediated shopping raises questions about their applicability alone. Recent research on generative-AI shopping assistants likewise emphasizes the relevance of personalized recommendations, interaction design, and trust to consumer experience. The presentation offers insights into four considerations warranting closer attention in this context: recommendation quality, particularly alignment with consumers' preferences, budgets, needs, and purchase constraints; clarity about recommendation rationales and limitations; commercial transparency regarding sponsored or commercially influenced recommendations; and perceived algorithmic fairness. These considerations are discussed in relation to perceived usefulness, trust in the AI assistant, trust in recommended brands, brand consideration, and purchase intention. The discussion also considers privacy concerns, disclosure of commercial influence, and consumers' familiarity with AI. For brands seeking visibility in generative-AI shopping systems, the presentation argues that the accuracy, clarity, accessibility, and trustworthiness of the information informing AIgenerated recommendations not only traditional digital advertising and search-engine presence may influence brand discovery. The discussion concludes with implications for researchers developing empirical studies of conversational commerce and for managers seeking to build consumer trust in AI-mediated product discovery.
Shweta Dharia is an Assistant Professor in the Masters of Management Program at the University of Niagara Falls Canada and a Doctor of Business Administration candidate in Emerging Technologies and Applied AI at Golden Gate University. With over 10 years of teaching experience across Canadian postsecondary institutions, she specializes in AI governance, educational policy, business communication, and responsible AI adoption in higher education. Her work bridges academic research, industry experience, and institutional practice, with a focus on inclusive teaching, international student engagement, and ethics-by-design AI governance.