Rotating machinery has been a major contributor to the development in manufacturing industry. Ensuring the efficiency and maintenance on these machineries has been an objective to safety and productivity. This article offers a comprehensive review of the recent progressions in cross-machine fault diagnosis methodologies, with a specific focus on their role in determining faults of machine components. Initially, the paper introduces the concept of cross-machine fault diagnosis, underscoring its significance within contemporary industrial frameworks. Besides, the review scrutinizes the transformative influence of machine learning and artificial intelligence on fault diagnosis. Examining diverse machine learning models, including neural networks, support vector machines, and adversarial networks, addressing their roles in identifying faults within components of machines. Subsequently, an assessment of the advantages and disadvantages of each methodology for further enhance the comprehension of those techniques’ implications. Furthermore, the paper also compiles datasets frequently being used which are publicly accessible. In conclusion, the paper reflects on the prospective paths of cross-machine fault diagnosis, stressing the necessity for further exploration in realms like automated diagnostic systems, real-time monitoring, anomaly detection and predictive maintenance. This critique strives to furnish scholars and professionals with a comprehensive insight into the prevailing landscape of cross-machine fault diagnosis, thereby steering forthcoming research and practical applications in this domain.