This paper investigates two methods for the prediction of onset of slip in gait during the transition from single support to double support. The first method employs an optimization-based gait prediction simulation to determine the threshold of walking velocity that results in stable gait for one stride length and one subject. In order to determine the threshold, the simulation is carried out with progressively increasing walking velocities (initial conditions) until the gait becomes unstable. The zero moment point (ZMP) and support region are employed as criteria for stable gait, i.e. as long as the ZMP remains within the support region throughout the duration of gait, the gait is said to be stable. The second method employs a probabilistic simulation to predict the likelihood of slip, where the likelihood of slip is related to the available and required friction for gait.