This study empirically examined perception differences between teachers and students regarding AI-based learning platforms through a learning analytics lens. A survey of 161 teachers (89 elementary, 72 secondary) and 431 students (226 elementary, 205 middle school) with at least six months of platform experience was conducted across six regions at digital innovation leading schools. The instrument comprised 22 items across six domains based on Davis’s (1989) Technology Acceptance Model and Wang et al.’s (2023) AI readiness framework, demonstrating robust reliability (Cronbach’s α=.74~.88). Results revealed: (1) significant perception gaps across all domains, with notably large effect sizes for usage patterns (d=0.98) and concerns (d=0.88); (2) school level × user type interaction effects, where teacher-student differences in continuance intention were pronounced at elementary but attenuated at middle school levels; and (3) differential predictors of continuance intention, with perceived usefulness as the strongest predictor for both groups, yet readiness showing null effects for teachers but positive effects for students. These findings underscore the necessity for differentiated implementation strategies tailored to distinct stakeholder characteristics within AI-based educational ecosystems. 

Keywords : AI-based learning platform, learning analytics, teacher-student perceptions, technology acceptance model, AI readiness