LLM (Large Language Model)

A Large Language Model, put simply, predicts the next piece of text each time. By training that on vast amounts of text, it learns patterns in language so well that it can summarize, translate, write, classify and reason. Well-known examples are the models behind ChatGPT, Claude and Gemini.

Important to understand: an LLM knows nothing for certain, it estimates the most likely answer. That explains why it is sometimes brilliant and sometimes produces nonsense with conviction (see hallucination). The model also only knows what was in its training, not your company data, unless you provide it explicitly via RAG.

For a business, the model itself is rarely the interesting part, it has become a commodity. The difference lies in what you build around it: your data, your processes and the boundaries within which the model is allowed to work.