A new semiactive control strategy for seismic response reduction using a neuro-controller and a magnetorheological (MR) fluid damper is proposed. The proposed control system adopts a clipped algorithm which induces the MR damper to generate approximately the desired control force. The improved neuro-controller, which was developed by employing the training algorithm based on a cost function and the sensitivity evaluation algorithm replacing an emulator neural network, produces the desired active control force, and then by using the clipped algorithm the appropriate command voltage is selected in order to cause the MR damper to generate the desired control force. In numerical simulation, a three-story building structure is semiactively controlled by the trained neural network under the historical earthquake records. The simulation results show that the proposed semiactive neuro-control algorithm is quite effective to reduce seismic responses. In addition, the semiactive control system using MR fluid dampers has many attractive features, such as the bounded-input, bounded-output stability and small energy requirements. The results of this investigation, therefore, indicate that the proposed semiactive neuro-control strategy using MR fluid dampers could be effectively used for control of seismically excited structures.