Prompt
A prompt is the input to a generative model. A clear task helps but does not replace data connections or access permissions.
Give the model a specific task
A prompt is the input to which a generative AI model responds. It may contain a question, instructions, examples and material to process. The information the model actually receives also depends on the surrounding application.
In a fictional support example, the instruction is: “Summarise this enquiry and identify the missing order reference.” The attached message is material to process. It must be distinguished from the processing instructions.
A request such as “Use current delivery data” does not establish a connection to the order system. Likewise, “You may only read” does not replace technically restricted access rights. A prompt shapes the model's task; data connections, permissions and result checks remain responsibilities of the application.
Sources
Related terms
- Large language modelA large language model processes language using learned patterns. Fluent text proves neither factual accuracy nor access to current business data.
- Context windowThe context window limits how many tokens a language model can consider in one processing step. It is not a permanent knowledge archive.
- Model cardA model card documents an AI model’s purpose, foundations, evaluation conditions and known limitations for a particular version.
- AI readinessAI readiness concerns the foundations of an AI project: objectives, usable data, skills, responsibilities and an operating workflow.
- Context engineeringContext engineering prepares information for individual model calls: relevant instructions, messages, sources and current states.
- Anomaly detectionMethods that flag unusual observations relative to a defined reference or a learned pattern.
Articles on this topic
AI & Automation with ASES
AI assistants and automated workflows