Core concepts
How does a large language model (LLM) work?
A large language model (LLM) works by learning patterns from training data and using context to generate sequences of text.
A model can produce explanations, summaries and other responses. Applications may connect it to search, documents or additional tools.
Its generated text is not inherently verified. Learned associations can be outdated, incomplete or wrong.
In Seenford
Why it matters in Seenford: answers come from clean sessions, not personalised ones, so personal context is not part of the check.
Example
A model writes an explanation based on a question and supplied documents.
Related terms
Updated
See what AI says about your brand →