AI & prompting, in plain English
Short, jargon-free definitions of the terms that come up when you're writing prompts and running AI across a team.
Prompting & AI basics
Prompt engineering
Prompt engineering is the practice of writing and refining the instructions you give an AI model so it returns more accurate, useful and on-task answers.
System prompt
A system prompt is the top-level instruction given to an AI model before your message — it sets the model's role, rules and tone for the whole conversation.
Few-shot prompting
Few-shot prompting is giving an AI model a handful of worked examples inside the prompt so it copies the pattern, format or style you want.
Zero-shot prompting
Zero-shot prompting is asking an AI model to do a task with no examples — just a clear instruction — relying on what it already knows.
Chain-of-thought prompting
Chain-of-thought prompting asks an AI model to reason step by step before giving its answer, which improves accuracy on problems that need working-out.
Context window
A context window is the maximum amount of text an AI model can consider at once — your prompt plus its reply — measured in tokens.
Temperature (in AI)
Temperature is a setting that controls how predictable or varied an AI model's output is — low values stay focused and consistent, high values add creativity and surprise.
AI hallucination
An AI hallucination is when a model states something false or made-up as if it were fact — fluent, confident and wrong.
Large language model (LLM)
A large language model is an AI system trained on huge amounts of text to predict and generate language — the engine behind tools like ChatGPT, Claude and Gemini.
Prompt chaining
Prompt chaining is breaking a big task into a sequence of smaller prompts, where each step's output becomes the input to the next.
AI for business
AI governance
AI governance is the set of policies, approvals and records a business uses to make sure AI is used responsibly — and to prove it.
Shadow AI
Shadow AI is the use of personal or unsanctioned AI tools for work, outside any oversight from the organisation.
LLM gateway
An LLM gateway is a single control layer that sits between your team and multiple AI models, routing requests and applying controls.
Prompt management
Prompt management is the practice of storing, approving and sharing a team's prompts so AI output stays consistent and on-brand.
AI data loss prevention (AI DLP)
AI DLP (data loss prevention) is the practice of stopping confidential or regulated data from being sent into public AI models.
Token cost
Token cost is what AI models charge per "token" — a chunk of text — processed as input and output. It's the main driver of an AI bill.
AI usage policy
An AI usage policy is a written statement of what staff may and may not do with AI, and how the firm keeps that use safe.
Model neutrality
Model neutrality is the ability to use AI across multiple providers — Claude, ChatGPT, Gemini — without being locked to any one of them.
Prompt library
A prompt library is a shared, curated set of approved prompts a team reuses for its common tasks.