A large language model, or LLM, is an AI model that works with language. It reads text, predicts likely next tokens, and uses those predictions to write answers, summaries, code, plans, and explanations.
In practice
LLMs are useful because they can apply language patterns across many tasks without being trained separately for every single task. The same model can draft an email, explain a concept, classify a message, or help debug code when the prompt gives it enough direction.
What to watch
An LLM is not a database or a human expert. It can sound confident when it is wrong, so important answers need grounding, source checking, or human review.