This paper reviews the application of AI in exoskeleton-assisted rehabilitation of the lower limbs, focusing on 37 peer-reviewed papers. The review categorizes these papers based on robotic application scenarios and AI methodologies, providing a detailed summary of input features, AI model performance, enrolled populations, exoskeletal systems used, and specific tasks. The innovative aspect of this review lies in offering a clear understanding of the suitability of different algorithms for specific tasks, aiming to guide future developments and support informed decision-making in the field of lower-limb exoskeleton and AI applications. The introduction highlights the clinical relevance of lower-limb rehabilitation, noting the growing number of individuals benefiting from rehabilitation and the potential of exoskeletons to improve therapy quality and patient independence. It discusses the integration of wearable robotic technologies and AI in rehabilitation, emphasizing the role of exoskeletons in enhancing locomotory economy, joint strength, and endurance. The methods section outlines the literature search process, including the use of specific keywords and databases, and the selection criteria for the included studies. The results section provides a comprehensive overview of the reviewed papers, grouped by AI methodology (Reinforcement Learning, Neural Networks, Support Vector Machine, and Decision Tree). The paper then delves into the specific applications of these AI methodologies in exoskeleton-assisted rehabilitation, detailing the functional blocks of a lower-limb exoskeleton, including robot control, locomotion classification, intention detection, and human joints trajectory prediction. It presents case studies and experimental results for each methodology, demonstrating their effectiveness in various rehabilitation tasks. Finally, the paper discusses the major outcomes of the literature analysis, including advantages, limitations, and future directions, concluding with a summary of the contributions and an appendix listing acronyms used throughout the text.This paper reviews the application of AI in exoskeleton-assisted rehabilitation of the lower limbs, focusing on 37 peer-reviewed papers. The review categorizes these papers based on robotic application scenarios and AI methodologies, providing a detailed summary of input features, AI model performance, enrolled populations, exoskeletal systems used, and specific tasks. The innovative aspect of this review lies in offering a clear understanding of the suitability of different algorithms for specific tasks, aiming to guide future developments and support informed decision-making in the field of lower-limb exoskeleton and AI applications. The introduction highlights the clinical relevance of lower-limb rehabilitation, noting the growing number of individuals benefiting from rehabilitation and the potential of exoskeletons to improve therapy quality and patient independence. It discusses the integration of wearable robotic technologies and AI in rehabilitation, emphasizing the role of exoskeletons in enhancing locomotory economy, joint strength, and endurance. The methods section outlines the literature search process, including the use of specific keywords and databases, and the selection criteria for the included studies. The results section provides a comprehensive overview of the reviewed papers, grouped by AI methodology (Reinforcement Learning, Neural Networks, Support Vector Machine, and Decision Tree). The paper then delves into the specific applications of these AI methodologies in exoskeleton-assisted rehabilitation, detailing the functional blocks of a lower-limb exoskeleton, including robot control, locomotion classification, intention detection, and human joints trajectory prediction. It presents case studies and experimental results for each methodology, demonstrating their effectiveness in various rehabilitation tasks. Finally, the paper discusses the major outcomes of the literature analysis, including advantages, limitations, and future directions, concluding with a summary of the contributions and an appendix listing acronyms used throughout the text.