The temporal and spatial evolution pattern of ecosystem sensitivity has been one of the most important components to restore and improve ecosystem stability, together with the driving factors and future development trends. The Loess Plateau is one of the typical areas with the most severe soil erosion in the world. Alternatively, the implementation of the "Grain for Green" project in 1999 has significantly improved the ecological environment. It is a high demand to examine the suitability of ecological governance measures using the temporal and spatial evolution characteristics of ecological sensitivity in the Loess Plateau over the past 20 years. In this study, an ecological sensitivity index system was constructed to couple the spatial distance index and the entire array-polygon indicator under the Driving-Pressure-State-Impact-Response framework. The temporal and spatial patterns and their driving forces of ecological sensitivity were investigated at five-time points from 2000 to 2020 in the Loess Plateau using a geographic information evolution map and a geographic detector. The CA-Markov model was used to simulate the ecological sensitivity for ecological zoning in 2030. The results showed that:1) Spatially, the ecological sensitivity of the Loess Plateau exhibited a distribution pattern of high in the northwest and low in the southeast, with the high sensitivity types decreasing towards the southeast direction and the aggregation weakening, while the low sensitivity types increased towards the northwest direction, but the aggregation strengthened at the original position. Temporally, there was a varying trend of first increasing and then decreasing. 2) The ecological sensitivity transfer types were mainly fluctuating stability. In individual sensitivity types, the insensitive transfer type was mainly stability, while the low, moderate, and severe sensitive transfer types were mainly fluctuating stability, and the extremely sensitive transfer type was mainly reduction. The largest area proportion of the ecological improvement zone (35.17%) was mainly distributed in the western part of the Loess Plateau. The areas with 0 and 1 changes accounted for 62.06%, indicating a stable and improving trend in the ecological environment. 3) The most significant effect of the natural factors on the ecological sensitivity among the driving factors. The NDVI, precipitation, and their interaction posed a significant driving effect on ecological sensitivity. 4) The ecological sensitivity can be expected as the distribution pattern of high in the northwest and low in the southeast in 2030, where the low sensitivity will be the main sensitive type. The key areas of ecological protection were obtained as the sandy land, agricultural irrigation region, and the mid-west of the loess sorghum gully region. Overall, the ecological environment of the Loess Plateau has improved after the ecological governance measures. This study can provide a strong reference for the ecological environment and high-quality development in the Loess Plateau.