Imagine a world where every student has a personalized AI tutor, guiding them through their educational journey. This isn't science fiction; it's the potential of Large Language Models (LLMs) combined with Knowledge Tracing (KT) in education. Knowledge Tracing is like a digital detective, figuring out what a student knows and doesn't know based on their learning interactions. LLMs, the brains behind AI assistants like ChatGPT, supercharge this process. This exciting combination could revolutionize learning by providing personalized content, assessments, and feedback. Recent research explored various ways to combine these powerful technologies. LLMs can adapt to different subjects, generate practice questions, even grade open-ended responses. They can also help overcome some of the limitations of traditional Knowledge Tracing methods, such as the 'cold-start' problem where there's limited initial student data. However, some big hurdles remain. Building these AI tutors is computationally expensive, and tailoring them to different educational settings is a challenge. Moreover, ensuring student privacy while collecting detailed learning data is crucial. Despite these obstacles, the potential of LLMs and KT in education is enormous. As research progresses and these technologies become more refined, we could see a future where learning is truly personalized, effective, and accessible to all.