AI Glossary: AI Terms and Tools Explained


Artificial intelligence brings new tools, technologies and terms into our work and everyday lives. What is a prompt? How does zero-shot prompting work? What is the difference between Microsoft 365 Copilot and Copilot Studio?

The Prompts AI Glossary explains key AI concepts in clear English, with practical examples and links to related tools. It helps you understand what AI can do, where it has limits and how to use it with a clear purpose.

Whether you are getting started, working in marketing, developing software or introducing AI into a business, use this glossary as a place to build your knowledge.



AI and prompt engineering basics


Start with the instructions you give to an AI system. Then compare three ways to show it what you expect: no examples, one example or several examples. These techniques are useful for tasks such as writing, classifying information and following a consistent format.




What is a Prompt?


A prompt is the question, instruction or task you give to an AI system. Learn how context, a clear goal and a useful output format can improve your instructions.

What is a prompt? Meaning and practical examples →




Zero-shot Prompting


Give an AI model a task without a completed example of the answer. Find out when clear instructions are enough and when examples may help.

What is zero-shot prompting? →






One-shot Prompting


Add one example to show the expected style, structure or label. See how to choose an example that supports your task.

What is one-shot prompting? →




Few-shot Prompting


Use several input and output examples to show a pattern. Explore a practical approach to tasks that need consistent categories or formatting.

What is few-shot prompting? →





Chain of Thought (CoT)


A prompting technique that encourages AI models to solve complex problems through intermediate steps. Learn how Chain of Thought works, how it differs from AI reasoning, and when it can improve problem-solving.

What is Chain of Thought (CoT)? →





More prompting techniques


Prompt chaining breaks a task into connected prompts with reviewed intermediate results. Follow a practical workflow from notes to a proposal draft.

Role prompting sets a useful perspective for a specific task. Learn how to combine a role with a concrete brief, examples and clear limits.



AI tools and platforms


Explore tools for everyday work, software development, workflow automation and custom AI applications. Each glossary page explains the main purpose of the tool, an example business scenario and important questions to ask before using it.




Microsoft 365 Copilot


An AI assistant for work in the Microsoft 365 environment. Understand its role in supported apps and why available features, data access and setup matter.

What is Microsoft 365 Copilot? →




Microsoft Copilot Studio


A low-code platform for building and managing AI agents. Learn how a focused agent can connect business knowledge with a defined process.

What is Microsoft Copilot Studio? →






Microsoft Foundry


A Microsoft platform for developing, evaluating and operating AI applications and agents. Learn how it differs from an assistant used for everyday work.

What is Microsoft Foundry? →




n8n


A workflow automation platform for connecting applications, APIs and AI services. Explore how a workflow can move information through a repeatable process.

What is n8n? →






GitHub Copilot


An AI coding assistant for software developers. Explore how it can support writing, explaining, changing and reviewing code in a development workflow.

What is GitHub Copilot? →




Claude Code


An agentic coding tool from Anthropic. Learn about working with project files and giving a coding agent a task with clear boundaries and checks.

What is Claude Code? →





Turn AI knowledge into practical skills


Knowing the terms is a useful start. The next step is to write a clear request, check the result and improve the instruction. Visit our AI Guides or start with How to Write a Good AI Prompt.





AI Lab: From theory to practical results


Understanding AI terms is the first step. It also helps to see how different models respond to real tasks.

In the Prompts AI Lab, we test different AI models with the same assignment. Each model gets one attempt, and we publish its output without further editing. You can compare experiments involving games, interactive applications, maps and 3D visualisations.

These experiments show differences on specific tasks. A single result is an example to inspect, rather than a complete ranking of every model.






Want to use AI in your business?


At Prompts, we help businesses choose AI technologies, train their teams, introduce Microsoft Copilot, automate workflows and develop custom AI assistants and agents. Together, we can identify where AI could bring measurable value to your work.

Tell us about your team, the task you want to improve and the tools you already use. We can discuss a practical starting point.