AI customer support terminology

An AI agent is software that uses a large language model to understand requests and take autonomous action, rather than following a fixed script.

RAG is a technique where an AI model retrieves relevant information from a knowledge base before generating an answer, so responses stay grounded in real, current facts.

A knowledge base is the collection of documents, articles, and FAQs an AI agent retrieves from to answer questions accurately, using retrieval-augmented generation.

Escalation is the process of handing a conversation from an AI agent to a human — automatically, based on rules like sentiment, topic, or an explicit request.

Conversational AI refers to systems that understand and generate natural language to hold open-ended dialogue, as opposed to rule-based chatbots limited to predefined flows.

Ticket deflection is resolving a customer's question before it becomes a support ticket a human has to handle — typically via self-serve answers or an AI agent.

Multi-channel support means answering customers wherever they reach out — chat, WhatsApp, email, social messaging, phone — from one unified system with shared context.

Lead scoring is ranking prospective customers by how likely they are to convert, based on signals like conversation intent, stated budget, or engagement — often automated by an AI sales agent.