What is a Prompt?
The input text or instruction given to an AI model to guide its response.
Pronunciation: /prɒmpt/
Prompt explained in plain English
A prompt is what you type or provide to an AI model to tell it what you want. It can be a question, instruction, example, or combination of these. The quality of the prompt significantly affects the quality of the model's output.\n\nPrompt engineering — the skill of crafting effective prompts — has become a key discipline in working with AI systems.
Analogy
A prompt is like a brief you give to a contractor. The clearer and more detailed your brief, the more likely you are to get exactly what you wanted.
Example
Instead of prompting "Write about AI," a better prompt would be: "Write a 200-word explanation of machine learning for a 12-year-old, using an analogy about learning to ride a bicycle."
How is Prompt used?
Every interaction with ChatGPT, Claude, or any LLM starts with a prompt. Developers embed prompts in applications to automate tasks like summarisation, classification, and content generation.
Common misconceptions about Prompt
More words in a prompt do not always mean better results. Effective prompts are clear and purposeful, not necessarily long.
History
Prompting as a concept existed in early language models, but the term "prompt engineering" gained prominence with GPT-3's in-context learning capabilities (2020).
Related terms
People also read
- LLM
A type of AI model trained on vast amounts of text to understand and generate human language.
- Agent
An AI system that can perceive its environment, make decisions, and take actions to achieve goals autonomously.
- Chain-of-Thought Prompting
Asking an AI to show its reasoning step by step before giving a final answer, which often improves accuracy on complex tasks.
- Inference
The phase when a trained model is actually used — taking new input and producing a prediction or response.
- RAG
A technique that combines AI language models with external knowledge retrieval for more accurate answers.
- AI slop
Output from a generative AI system that favors quantity over quality.
- average precision at k
A metric for summarizing a model's performance on a single prompt that generates ranked results, such as a numbered list of book recommendations.
- black box model
A model whose "reasoning" is impossible or difficult for humans to understand.
- Confabulation
When an AI produces a confident, fluent answer that sounds true but is factually wrong — generating plausible language without a reliable link to reality.
- Context Window
The maximum amount of text a language model can consider at one time, measured in tokens.