The objective of this review was to investigate the variety of machine learning (ML) methods and their ability to detect individuals at risk of or suffering from sarcopenia. Sarcopenia is a progressive disease characterized by a significant decline in muscle mass, muscle strength, and physical performance, primarily affecting older adults. PubMed databases were searched from July to August 2024 for trials related to the sarcopenia prognosis and artificial intelligence. The keywords used were “sarcopenia,” “artificial intelligence,” “machine learning,” and “risk factors.” Eleven studies were included involving 15,799 participants. All studies were able to identify sarcopenia with moderate to high accuracy values and with different ML methods. The machine learning methods that standout were deep neural networks, LightGBM, Decision Tree, CatBoost (CAT), and k-nearest neighbors. As for the risk factors, it is found that the most important were age, body mass index, waist circumference, chronic diseases, and some socioeconomic features. In conclusion, machine learning methods have the ability to extract valuable insights from data enabling accurate predictions for the early detection of sarcopenia. These results indicate that ML methods can be used by health professionals for faster and time-saving methods to identify sarcopenia.