For the first time, a low-cost and eco-friendly adsorbent prepared from pomegranate peel was used for the efficient removal of aniline blue (AB) dye from wastewater. After carbonization at 500 °C for 1 h, chemical activation using HCl was done. Textural characterization and adsorbent properties were analysed using FTIR, SEM along with EDX and pH ZPC . The influence of various operating parameters such as initial solution pH, dosage time, adsorbent mass and initial dye concentration for AB removal was investigated. Kinetic studies were conducted at room temperature (30 °C) by varying guest molecules concentrations. Guest-host interaction was maximum (90.78%) at pH 4. Langmuir, Freundlich, Temkin and Dubinin-Radushkevich isotherms were employed to design and optimize the adsorption data. The data obtained agreed well with the Freundlich isotherm and followed pseudo-second-order kinetics. The maximum AB uptake was predicted from Langmuir model (27.322 mg/g at 30 °C for 0.1 g pomegranate peel activated carbon). Thermodynamic studies performed in the temperature range of 30-50 °C indicated a decrease in randomness at the solid-liquid interface during physic-sorption, and the process was exothermic. Multiple regression (MR) and state-of-the-art artificial intelligence technique, namely epsilon-insensitive loss function-support vector regression (ε-SVR), were used for modelling. Prediction performance and analyses of the developed MR and ε-SVR models were done using statistical parameters such as AARE, R, RMSE, SD, MAE, Q 2 LOO and Q 2Ext . Parity plots between the adsorption data and the predicted data demonstrated that the SVR-based model was a robust model with high accuracy and generalizability.