[Objective] Beiliu City, located in the karst region of Guangxi, is highly susceptible to collapse and landslide hazards due to rainfall erosion and runoff, posing severe threats to life and property. Therefore, it is crucial to establish a collapse and landslide risk assessment system tailored to the karst areas of Guangxi to provide a scientific basis for early warning and disaster prevention in Beiliu City. [Methods] Beiliu City was selected as the study area, and a collapse and landslide database was constructed. Using slope units as the evaluation foundation, multi-source data were systematically collected and analyzed. Key evaluation indicators, including groundwater type and runoff intensity index, were selected, and a fully connected neural network model was employed to evaluate the susceptibility to collapse and landslide hazards. Considering the region's vulnerability to rainfall and karst erosion, the study introduced the soil erosion modulus for hazard assessment, and combined it with the vulnerability of disaster-bearing bodies to construct a risk assessment model for Beiliu City. [Results] The results of the risk assessment indicated that the areas of high and very high risk covered 252.22 km2, accounting for 10.27% of Beiliu City's total area. These high-risk zones are primarily distributed in Longsheng Town, Xinfeng Town, Pingzheng Town, and Liujing Town, characterized by eroded and denuded hills and tectonic erosion low mountains. The loose rock-soil structure, high soil erosion modulus, dense population, and concentrated buildings make these areas highly vulnerable to collapse and landslide threats. [Conclusion] The ROC curve and field investigation validation demonstrated that the collapse and landslide susceptibility model for Beiliu City achieved a precision of 0.9664, with a risk assessment accuracy of 89.3%. These results confirm that the constructed model has high accuracy and practical applicability, aligning well with the actual situation. It provides a valuable scientific basis for early warning and disaster prevention in Beiliu City. Moreover, the successful application of this assessment system offers an effective risk evaluation method and practical experience for other karst regions.