Failures can occur in any interaction between humans and robots, and the extent of the failure depends on the domain of application and the context in which the failure occurs. Typically, interaction failures are attributed to errors on the robot’s side. This paper examines the concept of Intelligent Disobedience (ID) and how it can be leveraged for managing interaction failures caused by humans. Our work-in-progress focuses on the practical implementation of ID in robot teleoperation and highlights the benefits of interdisciplinary collaboration for this project.