Retrieval-augmented generation (RAG) is an AI information gathering method that directs models to retrieve information from a specified dataset or knowledge base. Businesses can give algorithms access to data from thier own repositories to supplement the large language model (LLM) to generate answers based on the most relevant documents you provided. This method has been show to increase relevancy and accuracy. Including your own supplemental data has also been shown to help reduce AI hallucinations without exposing your private data to third parties