AI Glossary

The Plain-English AI Glossary

Every AI term you'll see in articles, podcasts, and tweets — explained without jargon. Bookmark this page.

The AI world has its own vocabulary. Most of it sounds harder than it is. Below are the terms that come up most often, with a one-paragraph definition and a real example. Each term links to a full explainer if you want to go deeper.

Prompt Engineering

The skill of writing inputs that get better outputs from AI. Less mysterious than it sounds.

System Prompt

Hidden instructions that shape every conversation. The AI's "operating manual" before you say a word.

Hallucination

When an AI confidently invents information that isn't true. The single most important AI failure mode to know.

Context Window

How much text an AI can "see" at once. The limit that determines whether it can read a book or just a chapter.

Token

A chunk of text the AI processes — roughly ¾ of a word. Why "rate limits" exist.

Temperature

A setting that controls how predictable vs creative an AI's outputs are.

Zero-Shot Prompting

Asking an AI to do something without giving it any examples. The default mode for most users.

Few-Shot Prompting

Giving the AI 1–3 examples in your prompt so it copies the pattern. The single fastest way to improve outputs.

Chain-of-Thought

A prompting technique that makes AI "think step by step" — produces dramatically better reasoning.

Role Prompting

"You are a..." prompts that change how the AI responds. The single biggest beginner unlock.

More entries coming weekly. Subscribe to our Tuesday newsletter to get notified when we add new terms.

One new prompt every Tuesday.

Free. No spam.

Copied