In the present study, optimization and prediction models for fungal bioleaching for effective metal extraction from waste printed circuit boards (WPCBs) of mobile phones were developed employing central composite design (CCD) of response surface methodology (RSM), and two artificial intelligence (AI) models, i.e., artificial neural network (ANN) and, support vector machine (SVM), respectively. Two continuous process parameters, such as pH (4–9) and pulp density (1–10 g/L), and the bioleaching approaches, viz., one-step and two-step, were experimentally optimized for the extraction of targeted metals, i.e., Cu, Ni, and Zn from WPCBs by mixed cultures of Aspergillus niger and Aspergillus tubingensis. Datasets were then used for predictive modelling using AI tools. Results showed that the highest simultaneous bioleaching of Cu, Ni, and Zn, with an extraction efficacy of about 86%, 51%, and 100%, respectively, achieved at an optimal condition of pH 5.7 and pulp density of 3 g/L following the two-step bioleaching approach. Effective metal extraction in the two-step approach could be attributed to the abundant production of organic acids with a content of about 16.3 g/L, 8.4 g/L, and 0.5 g/L of citric acid, oxalic acid, and malic acid, respectively. Further, the predictive modelling revealed that the ANN model was found to predict the fungal bioleaching responses more accurately as compared to the SVM model with R2 values exceeding 0.96 for all targeted metals. This research demonstrates the applicability of the optimization and prediction models for efficient metal extraction from WPCBs using mixed Aspergillus spp. following the two-step approach.