In a continuously evolving educational landscape, even a traditionally human-centered discipline such as Physical Education is beginning to integrate digital transformation. The integration of Artificial Intelligence (AI) into educational assessment is gaining momentum, yet its adoption in Physical Education (PE) remains limited. This study investigates students’ perceptions and trust toward explainable AI (XAI) systems used in PE assessment. Using a quantitative design, data were collected through a structured questionnaire administered to over 250 middle school and high school students from Bacău and Bucharest, Romania. The instrument measured students’ attitudes toward AI-assisted evaluation, their understanding of AI decision-making processes, perceived fairness and transparency, and the extent to which explainability influences their trust in automated assessments. We conducted descriptive and inferential analyses to identify key trends and relationships between demographic variables and trust indicators. Generally, the results highlighted a positive perceptions of AI as a supportive tool in PE grading, yet some degree of skepticism persisted among certain student groups. Moreover, concerns about data privacy and the potential depersonalization of assessment were also noted. The findings underscore the importance of integrating explainability features into AI-assisted educational tools to foster student acceptance, equity, and confidence in technology-enhanced evaluation. This study aims to contribute to the growing body of literature addressing the ethical and pedagogical dimensions of AI in education.