| With the rapid development of artificial intelligence and machine learning,AI chatbots have become a digital application with great commercial potential,affecting people’s life and work.Many companies have begun to invest in and deploy AI chatbots,especially in the field of customer service.AI chatbots serving as AI customer service are expected to effectively replace some human resources,reduce costs and increase efficiency for enterprises.In the field of e-commerce,AI customer service is used to provide users with product information,recommend preferential products,solve after-sales problems,and even conduct customer relationship management.The interaction mode based on natural language enables AI chatbots and users to understand each other,so as to meet user needs and enhance user experience.The resulting benefits make the market scale of AI customer service expanding rapidly.However,enterprises are also facing resistance from users in the process of deploying AI customer service.The emergence of "switching to human customer service" phenomena has brought loss of profits and increased labor costs to enterprises.Despite the many benefits of AI customer service,due to the complexity of the service environment and the heterogeneity of customer needs,AI customer service interactions will inevitably experience service failures,and the phenomenon of frequent user switching behavior to human service channels has become one of the great challenges that AI customer service has to face.In order to enable AI customer service to better replace human customer service,companies and developers focus on the anthropomorphic design of AI customer service,so that it has human characteristics and stimulates positive social reactions from users.This study also starts from this clue,based on social response theory and from the perspective of expectation,to carry out this empirical study to explore how the customer’s perception in the process of interacting with the AI customer service with anthropomorphic characteristics affects their expectation of service recovery in the face of failure under different service tasks,and the impact of this expectation on the customer ’s decision-making process whether to initiate a re-interaction,or to switch to a human agent.In this study,a scenario-based online experiment combined with questionnaire survey was used to explore the above research questions,and 404 valid experimental data were analyzed in detail by using SPSS26.0 and Smart PLS3.0 software.The research shows that: in the service failure scenario,the personification of AI customer service in the initial interaction will have a positive impact on the customer-recovery expectations,which represents the estimation of the possibility that the customers can meet their functional and emotional needs when interacting with AI chatbots again.When the expectation value is high,it will promote the customer to actively interact with AI again and prevent users from switching to human agents.In addition,when the customer thinks this service task is more important,the effect of emotional service recovery expectation on the customer’s switching intention will be weakened.This study correlates service failure scenarios with customer switching behavior.Through theoretical construction and empirical data analysis,it explains the impact mechanism of customer switching behavior in the environment of AI service interaction failure,further expands the research scenario of AI chatbot anthropomorphism,and innovatively classifies user service recovery expectations into functional and emotional dimensions,enriches the connotation of the concept of customer-recovery expectation in the context of AI customer service,in addition,deepens the understanding of the mechanism of customer switching behavior.It also provides theoretical guidance and practical suggestions for the practice of technology development and enterprise management,so as to promote the rapid development of AI technology in the field of customer service. |
