Chatbot Agency—Model Testing

The factors hypothesised in the conceptual model on chatbotChatbot agency are tested in an empirical study hereby. We have invited consumers who had used a chatbotChatbot at least once in their daily life to complete a survey, sharing their opinion about the experienceExperience they had concerning chatbotChatbot agency. The chatbotsChatbot were used to contact the customer services in 91% of the cases, demonstrating the increasing role of chatbotChatbot agents as front service line providers. The results from the regression modelling are clear in showing that: (1) InformativenessInformativeness and accuracyAccuracy predetermine chatbotChatbot functionalityFunctionality perception. (2) The higher the social presenceSocial presence was perceived to be, the higher was the enjoymentEnjoyment of interacting with a chatbotChatbot. (3) FunctionalityFunctionality and enjoymentEnjoyment had positive impact on ease of useEase of use and qualityQuality perception, and thus, on satisfactionSatisfaction. (4) The greater the satisfactionSatisfaction was, the higher the brand loyalty was and the intention to use chatbotsChatbot in future. Interestingly, however, personal carePersonal care did not play a role hereby, opposite to the proposition we had. CompetenceCompetence did not load on functionalityFunctionality, but modulated enjoymentEnjoyment. Given, we have asked about userUser experienceExperience after the actual use of a chatbotChatbot in a daily life situation, a reasonable question arising hereby is whether chatbotsChatbot currently available on the market offer the desired personal carePersonal care and competenceCompetence. This is a serious question challenging the UXUser Experience (UX) design to reconsider the contemporary AIArtificial Intelligence (AI) systems, and to ensure that these provide the desired agency, distinguishable with intelligence, empathyEmpathy, and social interaction.