Introduction: As the global community strives to ensure the health and well-being of mothers and newborns, AI emerges as a powerful ally in this noble endeavor. Through this systematic review, we seek to provide a comprehensive overview of the state of AI-driven mortality prediction, offering insights that may shape the future of maternal and neonatal healthcare and bring us closer to the goal of ensuring safe pregnancies and healthy beginnings for all. Material and methods: We systematically reviewed the literature, restricting our search to publications from the past decade, and utilized the five major scientific databases as primary sources. Results: Out of the initial pool of 671 works, a total of 18 primary studies were meticulously chosen for in-depth analysis. It was evident that a predominant focus of these studies revolved around the prediction of neonatal mortality, predominantly employing machine learning models, with Random Forest being a popular choice. The top five frequently utilized features for model training encompassed birth weight, gestational age, the child's gender, Apgar score, and the mother's age. The development of predictive models for mitigating mortality during and after pregnancy holds immense potential, not only for enhancing the quality of life for mothers but also as a potent and cost-effective tool for reducing mortality rates. Conclusion: Drawing from the findings of this systematic review, it becomes evident that substantial scientific endeavors have been undertaken in this domain. However, it is equally apparent that numerous unexplored research avenues and opportunities await further exploration within the research community.