Grounding
Connecting Claude's responses to verified, real-world sources or data rather than letting it rely purely on what it learned during training. A grounded AI cites where its information comes from, references actual documents, or queries live data before responding — reducing the chance of hallucination. RAG (Retrieval-Augmented Generation) is the main technical approach to grounding. Grounding matters most when accuracy is critical and Claude's training data might be incomplete or outdated.
In practice
You're building a support bot and want Claude to answer from your actual help docs, not guess from training data. You load your docs into a Project and instruct Claude to answer only from them. When a customer asks about your refund policy, Claude cites your policy — not a generic answer. That's grounding: Claude's responses are anchored to real, verified information you control.
Related concepts