Accompanied by the continuous development of Internet information technology and the promotion of technologies such as big data, artificial intelligence, and machine learning, AI-enabled services are gradually integrated into daily life, and remediation in the face of AI service interactions has also become an inevitable research issue. This study examines the impact of remediation in the event of service failures, particularly in the context of intelligent robot service failures that lead to negative customer experiences. Through the dissection of service failure remediation methods, combined with the customer satisfaction index model, a theoretical research model is constructed from the two aspects of willingness to use and satisfaction, and five hypotheses of remediation are proposed, and then experiments are designed to test them. The results of regression analysis indicated that the mean values of perception and willingness to use in the low anthropomorphism group were 4.875 and 5.052. The mean values of perception and willingness to use in the high anthropomorphism group were increased by 0.718 and 0.649 compared to those of the low anthropomorphism group. The customer’s satisfaction in terms of robot anthropomorphism in the high anthropomorphism was 4.055, and in the low anthropomorphism was 3.410, which indicated that there were positive correlations between the degree of anthropomorphism and both willingness to use and satisfaction have a positive effect, and all five hypotheses of this paper are proved. The purpose of this study is to provide an in-depth analysis of the use of AI anthropomorphization in service failure and service remediation and to offer guidance and references for AI services and related development and design decisions.