Prompt
In short
The input that tells an AI system what to produce or do, including a question, instructions, context, or supplied material.
A prompt gives an AI system a task and the information needed to respond. It might be a short question, a detailed instruction, or a request accompanied by a document or image.
Useful prompts specify the result you need. Context, audience, output format, and boundaries help when they change what a successful response looks like.
A practical example
Imagine you need descriptions for a new range of desk lamps. A vague request asks for attractive copy. A more useful prompt supplies the approved specifications, identifies the audience, and requests two sentences per item.
You could also require the response to flag missing dimensions and preserve the supplied warranty wording. These are illustrative instructions, not a required prompt formula. They make the output easier to check against the source information.
What a prompt does not guarantee
Clear wording reduces ambiguity, but it does not make a generated answer automatically correct. Review product claims before publishing them, particularly measurements, compatibility, prices, and delivery promises.
A prompt also does not grant access to data or permission to take actions. Those capabilities depend on the system, connected tools, and authorization. Asking an assistant to update stock does not establish that it can access your inventory.