A company deployed an AI/ML solution to help customer service agents answer frequently asked questions that change over time. Agents should be able to ask questions and receive automatically generated answers to common customer inquiries. Which approach is the MOST cost-effective way to meet these requirements?

Implementing Retrieval-Augmented Generation (RAG) combined with prompt engineering is the most cost-effective solution. RAG allows the model to retrieve relevant information from a knowledge base (e.g., FAQs) and use it to generate answers, ensuring accuracy and up-to-date responses without retraining the entire model. Prompt engineering further refines the output. Regularly fine-tuning the model would be costly and time-consuming, especially as FAQs change frequently. Training the model using contextual data or pre-training and benchmarking it with contextual data are also resource-intensive processes that would need to be repeated often, making them less cost-effective for dynamic information.