A recent discovery in AI tutoring technology highlights a critical flaw in systems that rely on Retrieval-Augmented Generation (RAG). After three weeks of testing, a user reported receiving an outdated answer from the AI tutor, which was misleading due to its lack of temporal awareness. This issue stems from the system's inability to prioritize the most current information, instead defaulting to the most similar document in its knowledge base. The knowledge base, by its very nature, is constantly evolving, making this oversight a significant concern. The solution to this problem didn't lie in the AI model or the retriever itself, but rather in the gap between them. To address this issue, a temporal layer was developed to filter the retrieved information based on its relevance to the current time. This innovative approach ensures that users receive the most up-to-date and accurate information, which is essential for applications like AI tutoring. The temporal layer is a critical component in bridging the gap between the retriever and the model. By incorporating this layer, the system can now provide users with the most current information, even in a constantly changing knowledge base. This development has significant implications for the field of AI tutoring and highlights the importance of temporal awareness in RAG systems. The implementation of the temporal layer has shown promising results, with users receiving accurate and relevant information. This achievement demonstrates the potential of AI technology to adapt and improve over time, and it serves as a model for future developments in the field.