Amazon’s AWS AI team has developed RAGChecker, a new research tool for evaluating Retrieval-Augmented Generation (RAG) systems that combine large language models with external databases. This tool offers a more detailed and nuanced approach to evaluating these systems, which are crucial for AI assistants and chatbots that need access to up-to-date information. RAGChecker is based on claim-level entailment checking and breaks down responses into individual claims for more accurate evaluation.