In normal daily activities, ligaments are subjected to repeated loads, and respond to this environment with creep and fatigue. While progressive recruitment of the collagen fibers is responsible for the toe region of the ligament stress-strain curve, recruitment also represents an elegant feature to help ligaments resist creep. The use of artificial intelligence techniques in computational modeling allows a large number of parameters and their interactions to be incorporated beyond the capacity of classical mathematical models. The objective of the work described here is to demonstrate a tool for modeling creep of the rabbit medial collateral ligament that can incorporate the different parameters while quantifying the effect of collagen fiber recruitment during creep. An intelligent algorithm was developed to predict ligament creep. The modeling is performed in two steps: first, the ill-defined fiber recruitment is quantified using the fuzzy logic. Second, this fiber recruitment is incorporated along with creep stress and creep time to model creep using an adaptive neurofuzzy inference system. The model was trained and tested using an experimental database including creep tests and crimp image analysis. The model confirms that quantification of fiber recruitment is important for accurate prediction of ligament creep behavior at physiological loads.