The seismic responses of a single-story reinforced concrete (RC) frame building model under control using an active mass damper (AMD) are demonstrated through a real-time hybrid simulation (RTHS) method. In this study, the RTHS test is carried out by using a hydraulic actuator and a shaking table under synchronization. Most parts of the target RC frame model are provided as an analytical model for an online computer simulation, and only the single column of the first story is prepared as an experimental substructure. A hydraulic actuator deforms the actual RC column, and uncertainty or nonlinearity of the RC column's behavior is focused on in this RTHS test. At the same time, a control device of AMD is actually tested under a situation of installing it on the target building's floor. The floor response of the target building model is generated using a shaking table . A control motion of the AMD is manipulated based on an online simulation of the entire RC building model. Firstly, a time delay compensation of the hydraulic actuator is considered. Time delay parameters are identified using a combination model of a time lag and a first-order delay. A PID controller and a time series compensator are applied to improve actuator performances. Next, the reproducibility of the RTHS test using two-individual actuators is evaluated. The tracking of a restoring force and deformation of the actual RC column specimen generated by the hydraulic actuator and floor motion responses reproduced on the shaking table are investigated. To improve the online numerical simulation based on the measured force responses of the RC column specimen, a high-pass filter (HPF) is applied for a force correction to utilize its phase-lead property. The effect of this HPF force correction is evaluated in both a linear region and a strong nonlinear region of the actual RC column specimen. Finally, the RTHS test results are compared to fully numerical simulations and the control effect of the AMD to increase the damping effect for the target RC building model is also investigated.