A concrete example
When you ask Claude to summarize an article, the LLM reads the text, predicts a plausible answer, and adjusts it to the context.
Why it matters
Almost every other word in the glossary starts here: prompt, token, context, hallucination, RAG.
It's hiding behind ChatGPT, Claude, Gemini, Perplexity, and a lot of assistants built into your tools.
Don't mix it up with
AI model: An AI model is the engine that reads your request and produces an answer.
Token: A token is a small chunk of text the AI counts to measure what it reads, writes, and bills.
Common mistakes
- Thinking an LLM always knows where its information comes from.
- Treating it like a perfect database.
- Forgetting it can be convincing and wrong.
Quick checklist
- First I check whether the word names a concept, a tool, a risk, or a metric.
- I tie it to a concrete case: When you ask Claude to summarize an article, the LLM reads the text, predicts a plausible answer, and adjusts it to the context.
- I keep the main trap in mind: Thinking an LLM always knows where its information comes from.
Quick questions
What is LLM in AI?
An LLM is an AI model trained to understand and generate language.
Where will I run into LLM?
It's hiding behind ChatGPT, Claude, Gemini, Perplexity, and a lot of assistants built into your tools.
Which word should I read next?
Start with AI model, Token, Context window.