This paper presents an investigation into the utilization of physics-informed neural networks for parameter identification in the domain of magnetorheological dampers. MR dampers are known for their controllable rheological properties, making them integral components in various engineering applications such as vibration control and structural dynamics. Efficient utilization of MR dampers relies on accurate characterization of their material properties, necessitating robust parameter identification techniques. The proposed methodology integrates physics-informed neural networks, a class of neural networks that embed physical principles into their architecture, enabling the incorporation of governing equations and boundary conditions during the training process. This fusion of physics-based constraints with machine learning facilitates the extraction of meaningful parameters from experimental data, enhancing the accuracy of the identification process. Through a series of simulations and experiments, this study assesses the efficacy of physics-informed neural networks in capturing the complex nonlinear behaviour exhibited by MR dampers. The neural network is trained on a dataset comprising experimental observations of the damper’s response under varying conditions. The results demonstrate the capability of physics-informed neural networks to discern and infer key material parameters. The findings presented herein contribute to the growing body of research on the application of machine learning techniques in structural dynamics and control. The demonstrated results of physics-informed neural networks in parameter identification for MR dampers showcase their potential as a valuable tool for engineers and researchers seeking to optimize the design and control of these adaptive devices in real-world engineering applications.