Tenderness is the most important factor affecting consumer satisfaction with beef. As demand for high-quality beef continues to grow, tenderness standards are receiving increased attention. Warner-Bratzler Shear Force (WBSF) is an objective measure widely used to assess meat tenderness and is a key tool for evaluating beef quality. In this study, we developed a machine learning-based model to predict WBSF across various cuts of Hanwoo beef, considering critical factors such as aging time, quality grade, sex, texture, and marbling. We collected WBSF measurements from 37 bulls, 104 cows, and 245 steers, resulting in a dataset of 15,326 entries spanning 33 different beef cuts. Five machine learning algorithms were applied: linear regression, principal component regression (PCR), gradient boosting regression (GBR), tree-based regression, and ridge regression. Among these, GBR demonstrated the highest predictive performance, achieving an MSE of 0.40, an R2 score of 0.64 and an MAE of 0.779, which represents a substantial improvement over conventional linear regression (MSE 0.66, R2 0.42) and other baseline models. Correlation analysis also confirmed that aging days exhibited the strongest relationship with WBSF (r = –0.41, p