Abstract: The stability of highway cut slopes is of great significance to the overall safety and efficiency of expressways. Measuring the deep-seated displacement of slopes using inclinometers is an important method for detecting slope stability. The existing stability assessment approach relies on manual empirical judgment based on displacement-time history curves derived from collected data. However, when the slope volume is large, the data are extensive, and the displacement-time history curves exhibit significant fluctuations, manual empirical judgment becomes inefficient and fails to provide accurate assessments of slope stability. Therefore, to address the challenges of evaluating the variation patterns of deep-seated displacement and early warning in cut slope stability assessment, this paper utilizes the Hurst exponent from fractal theory to analyze the time-series data of deep-seated displacement from typical monitoring holes in a specific expressway and the inherent patterns of their Hurst values. The results indicate that when the cumulative deep-seated displacement of a monitoring hole shows no significant abrupt change, its Hurst value is less than 0.8, and the slope is in a critical state. When the cumulative displacement exhibits a sudden increase, the Hurst value reaches an extreme value greater than 0.8, indicating that the slope is in an unstable state, which suggests the area is still undergoing movement and may pose a threat to the operation of the highway. As the cumulative deep-seated displacement stabilizes, the Hurst value continues to decrease. Based on the Hurst exponent from fractal theory, the monitoring data of deep-seated slope displacement can be analyzed, and this method can serve as an auxiliary means for slope stability assessment.