What Is a Prompt? English as the New Programming Language
A prompt is an instruction you write in English for an AI to follow. Why English is becoming the new programming language, and how to write your first one.
Programming used to mean learning a language a machine understood: Python, Java, C. Those are being replaced, slowly and then suddenly, by the one language you already speak. The instruction you give an AI model is called a prompt, and a prompt is just English written with enough precision to be followed.
What a prompt is
A prompt is the message you type to get a result. Ask a chatbot to "write a polite email refusing a meeting" and the prompt is that sentence. The model reads it, and answers. The word works as both noun and verb:
- My prompt was too vague. (noun)
- I prompted it to shorten the text. (verb)
Generate is the verb for what the model does back — it produces, or generates, a response. Respond and answer mean the same, but generate is the one that has become standard, because the model is not answering from memory: it is producing new text that fits your instruction.
The core verbs
A handful of verbs carry most of what you will ever ask a model to do:
| Verb | Example |
|---|---|
| write | Write a short introduction. |
| summarise | Summarise this in three sentences. |
| explain | Explain it like I'm a beginner. |
| translate | Translate this into Spanish. |
| list | List the main points. |
| rewrite | Rewrite it to sound more formal. |
| compare | Compare the two options. |
Notice they are all imperatives — the plain base form of the verb. That is the grammar a prompt runs on, and it is the same grammar you already use to give directions.
Why English is the new programming language
A programming language works because it is unambiguous — every instruction has exactly one meaning, so the machine cannot misread you. English is normally the opposite: full of words that mean several things and sentences that lean on tone and context. But when you write a prompt, you write English the way a programmer writes code — short, explicit, leaving nothing to guess.
That is the skill this course trains. Not new vocabulary (you will meet some), but a way of using the English you already have: precise, literal, and complete.
Your first prompt
Start with the same three pieces every time:
- The task — what you want. Write…, Summarise…, Explain…
- The subject — what it is about. …a cover letter… …this paragraph…
- The constraint — how you want it. …in under 100 words… …for a beginner.
A weak prompt says:
Tell me about marketing.
A strong one says:
Explain three low-cost marketing channels for a small bakery, in simple English, under 150 words.
The difference is not cleverness. It is specificity — naming what you want, for whom, and how much.
A vague prompt costs you the answer you wanted. The model will happily answer "Tell me about marketing" with something generic, because that is all the instruction gave it. The more you leave open, the more it has to guess.
What "learning to prompt" really means
You are not learning to talk to a computer. You are learning to write English that cannot be misunderstood — which is the same skill that makes you clearer to people, too. The rest of this course builds it in order: first the vocabulary of AI, then writing instructions, then the structure of prompts that work, the technical layer of how models read you, and finally the prompting techniques that get the best results.
Test yourself
The quiz below checks the core words and the idea behind a prompt.