To comprehensively address the mitigation and prevention of disasters in railroad embankments, landslides, and dam slope failures, a thorough slope stability analysis is essential. Various conventional approaches exist to investigate slope instability. Currently, geotechnical engineers have developed numerous soft computing techniques for conducting slope stability assessments. The current study investigates the machine learning based method including artificial neural network and Adaptive Neuro-Fuzzy Inference System (ANFIS) for slope stability assessment. To conduct this investigation, 700 databases were collected, incorporating input variables such as slope height, soil cohesiveness, internal angle of friction, slope angle, unit weight, and peak ground acceleration, with the aim of predicting the Factor of Safety (FoS) of soil slopes. The entire database was partitioned, allocating 70% for training the models and reserving the remaining 30% for model validation. To assess model accuracy, five statistical metrics, namely TIC, R2, RRSE, RAE, and LMI, were employed in both the training and testing phases. The statistical analysis reveals that the Adaptive Neuro-Fuzzy Inference System (ANFIS) model attains higher accuracy with values of (TIC = 0.0010; R2 = 0.9999; RRSE = 0.0105; RAE = 0.0087; LMI = 0.9913) during the training phase and (TIC = 0.0044; R2 = 0.9999; RRSE = 0.0308; RAE = 0.0204; LMI = 0.9796) during the testing phase.