Prompt
A prompt is the input or instruction given to an AI system to guide what it does next.
In modern generative AI tools, a prompt often takes the form of a question, request, instruction, example, or block of context sent to a model.
What it does
A prompt tells the model what kind of output is wanted and what context it should use.
It can be used to:
- Ask a question
- Request a summary or rewrite
- Generate text or code
- Provide examples or constraints
- Set tone, format, or task requirements
Core concepts
Input to the model
A prompt is the starting input that shapes how the model responds.
For an LLM, the prompt usually includes the text the model should interpret before generating output.
Context and instructions
Prompts are often more than one sentence.
They can include:
- Background context
- Task instructions
- Constraints
- Examples
- Desired format
Prompt quality
Better prompts usually produce more useful outputs.
Clear instructions, enough context, and a well-defined goal tend to help more than vague requests.
Common use cases
- Asking a chatbot a question
- Requesting code generation
- Summarizing long text
- Rewriting content in a specific style
- Extracting structured information from input
Practical notes
- A prompt does not guarantee a correct answer; it only guides the model.
- Prompting is often iterative. People refine prompts after seeing the first result.
- Tools such as ChatGPT make prompting feel conversational, but the same basic idea applies across many AI interfaces.
- Prompt quality matters most when the task needs precision, structure, or domain-specific context.
Frequently Asked Questions
Is a prompt just a question?
Not always. A prompt can be a question, but it can also be a multi-part instruction, a block of context, or a structured task description.
Why do prompts matter?
Because the model’s output depends heavily on the instructions and context it receives.
Are prompts only for ChatGPT?
No. Prompts are used across many AI systems, especially LLMs and tools such as ChatGPT.