AI

LLM

A large language model — AI trained on massive text to predict and generate language.

An LLM is a giant next-word prediction engine: given everything before it, it guesses the most plausible next chunk of text. It doesn't 'look up' facts like a database — it absorbed statistical patterns from billions of documents, which is why it's fluent, fast, and sometimes confidently wrong. When you call Claude, ChatGPT, or Gemini, you're using an LLM. Treat it like a very well-read intern: brilliant output, but verify anything that matters.