This study introduces a novel approach to predicting the pressure and flow rate of flexible electrohydrodynamic (EHD) pumps using Kolmogorov-Arnold Networks (KAN). KAN, inspired by the Kolmogorov-Arnold representation theorem, replaces fixed activation functions with learnable spline-based activation functions, enhancing its ability to approximate complex nonlinear functions compared to traditional models like Multi-Layer Perceptron (MLP) and Random Forest (RF). The KAN model was evaluated on a dataset of flexible EHD pump parameters and compared with RF and MLP models. KAN achieved superior predictive accuracy, with Mean Squared Errors (MSE) of 12.186 and 0.012 for pressure and flow rate predictions, respectively. The symbolic formulas extracted from KAN provided insights into the nonlinear relationships between input parameters and pump performance, demonstrating its exceptional accuracy and interpretability. This makes KAN a promising alternative for predictive modeling in electrohydrodynamic pumping, particularly in applications requiring precise pressure control and a deep understanding of underlying relationships.This study introduces a novel approach to predicting the pressure and flow rate of flexible electrohydrodynamic (EHD) pumps using Kolmogorov-Arnold Networks (KAN). KAN, inspired by the Kolmogorov-Arnold representation theorem, replaces fixed activation functions with learnable spline-based activation functions, enhancing its ability to approximate complex nonlinear functions compared to traditional models like Multi-Layer Perceptron (MLP) and Random Forest (RF). The KAN model was evaluated on a dataset of flexible EHD pump parameters and compared with RF and MLP models. KAN achieved superior predictive accuracy, with Mean Squared Errors (MSE) of 12.186 and 0.012 for pressure and flow rate predictions, respectively. The symbolic formulas extracted from KAN provided insights into the nonlinear relationships between input parameters and pump performance, demonstrating its exceptional accuracy and interpretability. This makes KAN a promising alternative for predictive modeling in electrohydrodynamic pumping, particularly in applications requiring precise pressure control and a deep understanding of underlying relationships.