Artificial intelligence is an emerging technological domain capable of altering every aspect of our social interactions to ensure a sustainable society. The application of machine learning in predicting the quality of learning appears to be a promising field of research. Understanding what forecasts students’ learning poverty is crucial in promoting education quality and implementing practical and sustainable policies. This study aims to test a systematic procedure for implementing ML algorithms to predict learning poverty from the PISA 2022 test. Six distinct classifiers, including Decision Tree (DT), K-Nearest Neighbor (KNN), Artificial Neutral Network (ANN), Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM), were displayed. We established four metrics to validate these models and identify the best match. The final overarching goal of our research is to contribute to education issues. To this end, we offer new perspectives using machine learning algorithms to reach quality of education (SDG4).