Large Language Model (LLM)
A large language model (LLM) is an AI system trained on vast amounts of text to understand and generate human-like language. LLMs power chatbots, assistants, and answer engines, but they know nothing about your customers unless connected to your own first-party data.
Out of the box, an LLM reflects the public information it was trained on. It cannot tell you who will churn or what an account needs next, because that knowledge lives in your data, not the public web.
Techniques like retrieval-augmented generation connect an LLM to your owned, governed data so its responses are grounded in your specific business reality.
Put the vocabulary to work on your data
Get a free First-Party Data Readiness Review, or browse the full glossary and guide library.