Is treatment with rapid maxillary expansion and facial mask (RME/FM) stable in the long term? What are the patient-related factors that influence the long-term stability of this treatment? What is the role of machine learning in predicting Class III growth? Is it possible to control mandibular growth in patients treated with RME/FM? Are bone-anchored devices more effective than conventionally anchored ones for the treatment of Class III malocclusion? These are some of controversial topics that will be covered during this presentation. The first topic that will be covered will be growth in untreated Class III subjects, why it is so important to treat all patients affected by Class III malocclusion during the early developmental phases, and how we can use AI to predict craniofacial growth in untreated Class III subjects in the long-term. Additionally, the long-term effects produced by RME/FM with respect to untreated Class III controls will be analyzed together with the mechanisms involved in the control of mandibular growth. The presentation will then focus on the patient-related factors that can influence long-term stability of Class III treatment. In particular, the role of treatment timing (prepubertal vs pubertal and early prepubertal vs late prepubertal) on the long-term effects produced by RME/FM will be discussed. Additionally, the pre-treatment prognostic craniofacial features for the prediction of successful or unsuccessful long-term outcomes of RME/FM therapy will be illustrated. The findings of a recent Consensus Conference on the efficacy of bone-anchored versus conventionally anchored devices for Class III malocclusion will be presented.