One Size Does Not Fit All: Mechanisms of Employees’ Acceptance of Robotic Lower-Level Managers

Abstract

Organizations are facing fundamental changes in digitized work environments. These changes include the introduction of android and humanoid social robots designed to assist employees. The current research addresses whether and under what conditions employees accept such robots as lower-level managers, not just as assistants. The authors extend research on social robots with the expectation-disconfirmation paradigm. They examine how the interplay of expected and experienced performance-related and relational characteristics of robot-supervisors affect employees’ willingness to work with the robotic manager. Results from an online experiment with 7061 U.S. office workers reveal different effects for perceived performance-related and relational characteristics of a robot manager. For performance-related characteristics, employees’ willingness to work with robot managers follows an S-shaped pattern. A slight overfulfillment of expectations is associated with the highest levels of readiness. For relational characteristics, the results instead follow a degressive curve. Increasing positive experiences are associated with decreasing positive evaluations of willingness. Overall, employees indicate a preference for android robots over humanoid robots as managers. This research thus suggests tactics for the successful implementation of mixed human-robot teams and the effective introduction of social robots into lower-level management roles.