This paper presents a theoretical and experimental study to control grasping force of specific artificial hand (Otto Bock 8E37), which it uses by amputees. The hand has two rigid fingers actuated by a DC motor through a multi-gears system. The aim of this work is to give the amputees a feeling of slipping while the hand grasping an object. The mathematical model has been derived to simulate the hand mechanism and analyze the generated signal of contact force between fingertip and the grasped object through a slippage phenomenon. The experimental work consisted of modifying the artificial hand design to aid load cell mounting process in order to measure the grasping force indirectly, then acquiring the measured signal to the PC. An artificial neural network (ANN) was trained on the patterns of the force signals. These patterns were prepared by using force sensors with modified design of the artificial hand for detecting the slippage of the different shapes grasped object. The Neural Network training results have been evaluated and discussed under different conditions, which affect the controller operation such as network error, classification percentage and the response time delay.