Brain-computer interface systems (BCIs) aim to provide a better quality of life for people with disabilities by bypassing natural pathways between the brain and the central nervous system. A number of physiological sensors are known to develop a BCI system, out of which electroencephalogram (EEG) is popular among practitioners as it is convenient and effective. Motor Imagery (MI) is defined as the “mental rehearsal of movements without the movements being executed”. BCI systems have utilized MI to control the action of devices such as wheelchairs or an object on the monitor (e.g., cursor). To date, most BCI systems research focused on synchronous control applications, where the system initiates the time period during which the user is expected to perform the MI task. Alternatively, asynchronous BCI systems are way close to the ideal conditions, where the users don’t need to follow systems instructions, and they can intend MI activity whenever they want. In such systems, detecting a non-intentional activity period (i.e., no-control state) is equally important. Developing a classifier that classifies MI tasks efficiently and discriminates intentional MI tasks from the no-control state activity without having training samples for the no-control state is always critical. The aim of this work is to develop a self-paced asynchronous BCI system with robust non-control state detection using an endogenous brain switch, which can determine if the user wants to use the system, thereby reducing the number of false positive operations significantly. The details of the model formulation, salient results, and significance of the study will be presented in this seminar.