Teacher Readiness for AI Integration: A Professional Development Framework

Artificial intelligence (AI) technologies are increasingly embedded in classroom practice, yet teacher readiness to adopt them varies widely and is shaped by an interacting set of psychological, institutional, and demographic factors. This study proposes and empirically illustrates a Professional Development (PD) Framework for AI integration by applying supervised and unsupervised machine learning techniques to survey-based indicators of teacher readiness (N = 428). Four classifiers — logistic regression, support vector machine, random forest, and gradient boosting — were trained to predict three-level readiness (Low, Moderate, High) from digital literacy, self-efficacy, school ICT support, attitude toward AI, prior technology training, professional development hours, experience, and demographic covariates. Logistic regression achieved the strongest generalization (accuracy = .61, macro F1 = .54, 5-fold cross-validated F1 = .56), while random forest attained the highest raw accuracy (.63). Feature-importance analysis identified digital literacy, self-efficacy, attitude toward AI, and school ICT support as the four strongest predictors of readiness classification. K-means clustering on principal components identified three latent teacher profiles — cautious adopters, confident-but-under-supported adopters, and experience-anchored moderates — each requiring differentiated PD strategies. The results inform a tiered, profile-responsive PD framework that aligns institutional support, coaching intensity, and AI-pedagogy content with teachers' actual readiness profiles rather than a one-size-fits-all training model. Implications for teacher education, instructional leadership, and future longitudinal and multi-site replications are discussed.