Encyclopedia
Large language model (LLM)
What a large language model is, how it works in short, and its role in customer contact.
Large language model (LLM) is an AI model trained on large amounts of text to understand and generate natural language, and is the core technology behind modern chatbots, AI agents and conversational AI.
How an LLM works, in short
An LLM is trained on large volumes of text to predict the most likely next piece of text given what came before. Through that training it picks up grammar, facts and reasoning patterns. Modern models are then instruction-tuned so they follow requests rather than just continue text. Well-known families include Anthropic Claude, OpenAI GPT and Mistral.
What it can and cannot do
- Can: understand and generate language, summarise, translate, reason over a request, and, with tool-calling, decide which actions to take.
- Cannot reliably: act as a database of exact, current facts. It can produce confident but wrong output (a hallucination), which is why business systems ground it in a knowledge base and validated tools.
LLMs in customer contact
In customer contact, an LLM reads the customer's message, retrieves the relevant answer from the business's knowledge base, and can call tools to look things up or complete tasks. That grounding is what turns a general model into an accurate agent for a specific business.
From reference to practice
Turn what you just read into a working AI agent on your own customer contact. Start free, no credit card needed.