Direct Torque Control or simply (DTC) is a sophisticated control strategy that offers notable advantages for driving electric motors. However, it suffers from certain shortness’s caused by the use of (PI) regulators, which may lead for a certain deterioration of the overall system performance by causing: torque Ripples, high parametric sensitivity, low dynamic Performance at transient and Steady State periods. In order to overcome these issues, this paper adopts Artificial Intelligence (AI) technics with its advantageous features such as the high dynamics, smooth operations, stable and robust performance. To make a performance improvement with Artificial Neural Networks (ANNs) for a Direct Torque Controlled Squirrel cage induction motor (SCIM) fed by two-level voltage inverter. And at the end, a comparison study in made between the proposed (ANN) controller and classic (PI) regulator, the obtained results through several tests with MATLAB/Simulink platform has demonstrated and proved categorically the superiority of the intelligent controller.