This paper provides an analytical overview of the educational potential of artificial intelligence technologies, with a particular focus on self-directed learning in chemistry. The study explores how AI-based platforms facilitate adaptive explanations of complex chemical concepts, enable personalized learning pathways, and support the development of practical skills through virtual laboratory environments. The implementation of tools such as ChatGPT, ChemGPT, PhET Interactive Simulations, Wolfram Alpha, and Semantic Scholar in university-level chemistry education is examined through comparative analysis.