Abstract: The fracturing effect of coal and rock mass is the main basis for the design of supercritical CO₂ (SC-CO₂) fracturing projects. In order to accurately predict the effects of SC-CO₂ fracturing in coal and rock mass, based on the dynamic propagation characteristics between two opposite cracks originating from the adjacent fracturing boreholes, six geological factors and four construction factors that affect the SC-CO₂ fracturing effects in coal and rock mass are screened and determined. A hybrid artificial intelligence model that integrates the support vector machine (SVM), dragonfly algorithm (DA) and differential evolution algorithm (DE) is proposed、the relationship between the SC-CO₂ fracturing effects and the influencing factors is constructed using the SVM, the hyper-parameters of SVM are optimized using the DA and the differential evolution algorithm, the performance of the hybrid artificial intelligence model is evaluated using the correlation coefficient, root mean square error and average absolute error as the evaluation indices, and the sensitivity of the model input variables is analyzed by the MIV method. The results show that the proposed DA-DE-SVM prediction model can predict the effects of SC-CO₂ fracturing well of coal and rock mass. The R value of the training set is 0.9572 and that of the testing set is 0.9316. The importance of factors affecting SC-CO₂ fracturing effects is from high to low: horizontal distance between adjacent fracturing boreholes > vertical stress > fracturing fluid injection rate > vertical distance between adjacent fracturing boreholes > tensile strength > horizontal stress > fracturing fluid temperature > coal permeability coefficient > initial pore pressure > coal elastic modulus. The research results may provide important guidance for the parameter optimization design and engineering application of the SC-CO₂ fracturing technology.