This proof-of-concept study will be embedded into our line of research in this field. Our overall aim is to render the control of a myoelectrical forearm prosthesis (MCP) intuitive. Currently, an MCP typically has two integrated electrodes, through which the user provides input to the prosthesis. With these two degrees of freedom, individuals can control up to 20 different grip patterns. This control is however not intuitive. For example, to achieve a so called “cup holder grip”, a user explained he must conduct four different consecutive actions consisting of quick and slow single and double impulses, provided with muscles in his stump. This process takes him 15 sec. and a lot of cognitive input. Our proposed solution is to record electromyographic data (EMG) not only from two muscle groups in the stump, but from several major muscle groups in the stump as well as the intact upper extremity. AI will then be used to detect patterns and predict the required corresponding movement of the prosthesis. In a previous project, the developed algorithm could successfully distinguish between two different movement intentions. In this project, the algorithm will be refined, built into a prosthesis, and actually tested by prosthesis users.