AI glossary
AI terms, in plain language
Every term you keep running into when people talk about AI — defined simply, with a real example and why it matters for your work.
- AI agent(agent, agents)
- An AI agent is an AI system that plans and takes multiple steps — often using tools or external systems — to accomplish a goal, rather than answering a single prompt.
- Context window(context length)
- A context window is the maximum amount of text (measured in tokens) an AI model can take into account in a single conversation.
- Hallucination(confabulation)
- A hallucination is when an AI produces a response that sounds plausible and confident but is factually wrong or entirely made up.
- Large language model(LLM)
- A large language model is an AI system trained on very large amounts of text to understand and generate human-like language.
- Prompt(prompting)
- A prompt is the text instruction you give an AI tool describing what you want it to do.
- Structured output(JSON output, structured data)
- Structured output is an AI response returned in a defined format such as a table, bulleted list, or JSON, rather than free-flowing paragraphs.
- System prompt(system message, custom instructions)
- A system prompt is a background instruction that sets an AI's role, tone, and rules for an entire conversation, above the individual messages you send.
- Token(tokens, tokenization)
- A token is a small chunk of text — often a word or piece of a word — that language models use as their basic unit of input and output.