What is a prompt in AI?
A prompt is the input, usually text, you give an AI model to tell it what you want, such as a question, instruction or request. The model generates its response based on the prompt, so how a prompt is phrased strongly influences the quality and relevance of the output.
In AI systems, the prompt is your side of the conversation, the instruction or question the model responds to. It can be a simple question ('summarise this') or a detailed set of instructions with context and examples.
Because models are sensitive to wording, prompt design affects results. General-purpose AI often requires careful 'prompt engineering' to get reliable output, and even then a chatbot with no access to your data can only answer from its training, not from your actual numbers.
iDBQuery is built so your prompt can be plain and natural, and still produce a trustworthy answer about *your* data. When you ask something like 'what were our top five products by margin last quarter?', iDBQuery interprets that prompt, maps it onto your connected sources using its semantic layer, writes the SQL, runs it against your live data, and returns a cited answer that traces back to source rows. You do not need to craft elaborate prompts or feed it context by hand, the understanding of your schema is already built in. And because it grounds every answer in your data and shows its working, you avoid the guesswork and made-up numbers that a bare prompt to a generic chatbot can produce.
Updated 2026-06-22