In the context of reducing dropout rates in massive open online courses (MOOCs), adaptive learning is recognized as one of the key tools to remedy this phenomenon by personalizing the learning experience and responding to learners’ individual and specific needs. This article uses two optimization algorithms, namely Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO), to suggest adaptive optimal learning paths for at-risk learners. Using these optimization techniques inspired by natural phenomena, this paper seeks to compare the performance of these two approaches in terms of efficiency in generating adaptive paths. In other words, the study will compare the performance of these two approaches in terms of the quality of the courses generated, computation time and ability to adapt to changes in the learning environment. The results of this research will provide valuable information on the use of ACO and PSO in the context of adaptive learning, helping to improve dropout reduction strategies in MOOCs.