Accurate prediction of part quality is crucial for ensuring the performance of machined products. The ability to accurately predict part quality aids in optimizing machining process parameters, thereby increasing the durability and reliability of the final product. Recently, machine learning (ML) methods have been proposed to accomplish these tasks. However, the widespread application of these methods in real-world practice is hindered by their large computational resource requirements and lack of explainability. In this paper, we investigate the use of the Kolmogorov-Arnold Network (KAN), a new machine learning approach with high interpretability for part quality prediction in machining. Our validation with machining data from the turning of Ti6Al4V materials shows that KAN can surpass other baseline ML methods, including the state-of-the-arts Artificial Neural Networks (ANN), in terms of prediction accuracy. Moreover, KAN offers higher interpretability compared to other ML methods, as it allows one to visualize and interpret the decision-making process of the model, making the model validation task less challenging.