Soil erosion is the most crucial factor in land degradation in semi-arid areas. Natural climatic, edaphic, and topographic conditions, and human activities have made Morocco vulnerable to soil erosion. At the watershed scale, soil erosion assessment is of great importance in conservation strategies. The aim of this study is to assess soil erosion risk in Tensift watershed using the RUSLE model. Remote sensing (RS) and GIS were used to generate maps of RUSLE factors such as rainfall erosivity (R), soil erodibility (K), slope length and steepness (LS), cover management (C), and conservation practices (P) factors. The relationship between land use/cover, slope, elevation, geology, and soil erosion was analyzed. The annual soil erosion map made it possible to evaluate the average soil loss at 44.03 t/ha/year in the Tensift watershed. Soil loss assessment indicates that 30.43% of the watershed was subject to high-to-extremely high soil losses greater than 25 t/ha/year, while 69.57% of the watershed was still experiencing light and moderate soils losses less than 25 t/ha/year. The lower part of the watershed mainly experienced slight soil erosion, whereas the upstream portion was subject to high-to-extremely high soil erosion rates. Slope gradient and LS factor are correlated with soil loss with correlation coefficients of 0.57 and 0.55, respectively. The relationship between land use and soil erosion indicates that barren land and scrubland on steep slopes have a high rate of soil erosion. Thus, sustainable soil erosion prevention methods should be implemented in the upper part of the Tensift watershed to reduce soil erosion.