This paper reports adapting traditional Pacman game with Machine Learning technology Case-based Reasoning (CBR) to provide student learning motivation in the AI subject teaching. The paper explains how the study transforms the SOAR tutorial EATER game?s playing rules as a platform to experience that CBR (or AI) agent plays with human player in a shared running-map for finding the CBR agent?s best solution as competition with human player and Greedy agent. The advantage of AI project is that the experiment has to adopt human competition experience. There are many uncertain what CBR can achieve, which wait for the individual students to discover. The experiment results are different individually; the educators do not have to worry about the cheating in student assignments. The study found that the fun of interactive computer games would stimulate students? motivation in learning CBR concepts in AI.