| With the continuous development of artificial intelligence technology,the relationship between human and AI is getting closer and closer.The interaction between AI applications and users has become a hot topic worth discussing.Despite the widespread use of artificial intelligence technology in consumers’ work and life,many users are still skeptical of the application of AI.However,the behavior of user distrust will affect the willingness of users to use AI applications and affect the work efficiency of AI.The problem of user trust has become one of the biggest obstacles to further improve the efficiency of human-AI interaction.This paper systematically reviews the theories of social interaction between human and AI applications and the research results related to user trust.According to the current research status and lack of research,the research questions and hypotheses of this paper are proposed,and a research model including AI application characteristics,user perception and user trust is constructed.This study makes up for the lack of user perception perspective in current research,and provides empirical evidence for AI application characteristics,user perception and user trust relationship.In this study,two AI service robots were used to collect data through an online inter-group experiment design to test the research model.The results show that:(1)robots with human-like appearance are considered to have low levels of anthropomorphism and intelligence,and the design of robot interaction function does not significantly improve these levels;(2)The intelligence and anthropomorphism level of AI applications perceived by users have a significant impact on user trust,that is,the higher the anthropomorphism level and intelligence level of AI applications perceived by users,the higher the level of trust of users;(3)The intelligence level of the AI application perceived by users positively affects the anthropomorphism level of the application perceived by users. |
PDF Full Text Request