Regular physical activity is essential for physical and psychological wellbeing of pregnant women, yet many engage in insufficient exercise, leading to complications like gestational diabetes and hypertension; interventions that enhance self-efficacy, particularly home-based and non-weight-bearing exercises, can promote activity levels and mitigate these risks. The objectives are to create a framework for identifying factors related to pregnancy complications, evaluate supervised versus unsupervised exercise effectiveness through quantitative analysis, and inform clinical practices on the benefits of therapeutic exercise for enhancing maternal and fetal health. This review includes original articles from randomized controlled trials (RCTs) or controlled clinical trials (CCTs) that assess the effectiveness of physical activity interventions in healthy pregnant women aged 18 and older, focusing on various exercise forms compared to non-active interventions. Studies lacking data completeness or relevance, those involving pregnant teenagers, subjects with underlying health issues, incomplete or inaccessible data, studies without specific numerical outcomes relevant to pregnancy-related health and studies reporting acute exercise complications or using non-comparable formats are excluded. AI technologies can revolutionize prenatal physical activity programs by providing personalized behavior change techniques, real-time monitoring, predictive health insights, and adaptive support, ultimately enhancing both physical and psychological outcomes for pregnant women. These innovations enable tailored, data-driven interventions that improve adherence, optimize health recommendations, and support overall maternal and fetal well-being. Through real-time monitoring, virtual coaching, and data standardization, AI can transform prenatal healthcare, supporting healthier pregnancies and improved maternal and fetal outcomes.