Prompt: A prompt is the text (and sometimes files or images) you give an AI model to tell it what you want. A good prompt states the role the AI should play, the task, the relevant context, the output format, and any constraints.
The prompt is the whole interface. There is no settings panel that makes the model understand your job; the only lever you have is what you write. That is why two people using the same tool can get wildly different results from it.
A useful prompt usually covers five things: who the model should act as (a senior HR generalist), what to do (draft a policy summary), what it needs to know (paste the policy, name the audience), what shape the output should take (five bullets, under 150 words), and what to avoid (no legal conclusions). Skip one and the model fills the gap with a guess.
Prompts can be short when the task is simple ('fix the grammar in this paragraph'). They should be long when the task is judgment-heavy. If you would give a new hire three paragraphs of background before handing them the task, the model needs the same three paragraphs.
Example at work
Weak: 'Write an email about the delay.' Strong: 'You are a customer success manager. Write a 120-word email to a client whose installation slipped one week because of a parts shortage. Tone: direct and apologetic without over-explaining. Include the new date, [NEW DATE], and offer a call. Do not promise compensation.'
Why it matters
Prompting is the core skill of working with AI, and it is mostly the skill of briefing someone clearly. Professionals who already write good delegation notes get good AI output quickly; everyone else gets there with a little structure and practice.
Related terms
- Prompt engineeringPrompt engineering is the practice of writing and refining instructions to get reliable, high-quality output from an AI model. It covers giving context, showing examples, specifying format, and iterating on what did not work.
- System promptA system prompt is a set of instructions given to an AI model before a conversation starts that defines its role, tone, rules, and boundaries. Vendors write one for their products; you can add your own in custom instructions, projects, and custom assistants.
- Few-shot promptingFew-shot prompting is including a handful of examples of the input and output you want inside your prompt so the AI model can copy the pattern. It is the fastest way to get consistent formatting, tone, and judgment calls.
- Zero-shot promptingZero-shot prompting is asking an AI model to do a task with instructions only, without showing it any examples of the output you want. It works well for common tasks the model has seen many times, such as summarizing or translating.
- Context windowThe context window is the maximum amount of text, measured in tokens, that a language model can consider at one time. It includes your prompt, any pasted documents, the conversation so far, and the model's reply.