Balancing identity diversity and product contexts: Understanding consumer trust in AI-enhanced chatbot services

Abstract

As chatbots are increasingly integrated into customer service, businesses are rapidly expanding their use. However, a significant gap has emerged between the rapid adoption of these systems and consumers’ reluctance to trust them fully. This trust deficit weakens the effectiveness of these systems and poses a critical challenge to companies aiming to improve customer experience and optimize service quality. This study examines how AI-enhanced customer service chatbots’ identity design―spanning gender, race, name, and job title―affects consumer trust and preferences while also addressing broader societal questions about diversity and inclusion in AI design. Across seven experiments, results reveal that consumers generally trust female and non-Caucasian AI chatbots more and that chatbots with nicknames are perceived as more trustworthy than those with formal names. Moreover, chatbots with higher-ranking job titles consistently garnered greater preference. Trust was a key mediator linking identity design to consumer preferences, moderated by product attributes (utilitarian vs. hedonic) and psychological distance (proximal vs. distal). Beyond advancing theoretical understanding, the findings provide actionable insights for businesses seeking to balance commercial objectives with ethical considerations in AI design, thereby promoting social diversity and inclusivity while fostering trust chatbot services.