Glossary
AI terms in plain English
Every word you’ll meet in AI news, explained without the jargon.
- Agent
- An AI that can plan steps and use tools to complete a task for you, not just answer questions.
- AI (artificial intelligence)
- Software that does tasks we’d normally expect to need human judgement.
- API
- A way for one piece of software to talk to another, used by developers to build AI into their own apps.
- Benchmark
- A standard test used to compare how well AI models do at certain tasks.
- Bias
- When an AI’s answers unfairly favour or disadvantage certain views or groups.
- Chatbot
- A program you talk to in everyday language, such as ChatGPT, Claude or Grok.
- Claude Code
- Anthropic’s AI tool that works on the files on your computer, reading, editing and running things with your permission.
- CLI (command-line interface)
- A text window where you type instructions to your computer, or to a tool such as Claude Code.
- Context window
- How much text an AI can keep in mind at once, measured in tokens.
- Fine-tuning
- Extra training that adapts a general AI model to behave a certain way or do a specific job.
- Generative AI
- AI that creates new content, such as text, images, audio or code, from your request.
- Hallucination
- When an AI states something false or made up in a confident tone.
- Inference
- The moment an AI model is actually used to produce an answer, as opposed to being trained.
- LLM (large language model)
- The kind of AI model behind chatbots, trained on huge amounts of text to predict what comes next.
- MCP (Model Context Protocol)
- A standard that lets AI apps connect to tools and data sources in a consistent way.
- Memory
- A feature where an AI app remembers facts about you between conversations.
- Model
- The trained “brain” behind an AI product, such as Claude Sonnet 5.5 or Grok 4.7.
- Multimodal
- An AI that can work with more than one kind of input, such as text, images and audio.
- Parameters
- The internal numbers a model learns during training, often used as a rough size measure.
- Prompt
- What you type to an AI to tell it what you want.
- Prompt injection
- A trick where hidden text in a web page or document tries to make an AI follow the attacker’s instructions.
- RAG (retrieval-augmented generation)
- A method where an AI looks up relevant documents first, then writes its answer from them.
- Reasoning model
- A model that spends extra time “thinking” through steps before answering.
- System prompt
- Hidden instructions that set how an AI behaves before you say anything.
- Temperature
- A setting that controls how random or creative an AI’s output is.
- Terminal
- The text window on your computer where you type commands.
- Token
- A small chunk of text, roughly ¾ of an English word, that AI reads and writes.
- Training data
- The text, images and other material a model learned from.