Artificial Intelligence: Prompting Generative AI

Prompts are the instructions or inputs given by a user to guide generative AI (GenAI) in creating content. This generated content, also called an “output,” can come in the form of text, images, music, or any other form of digital media.

Why are prompts important?

Prompts define the task for the GenAI and influence the outputs in terms of style, content, and detail. The more specific the instructions given to the tool, the better the outcome. Prompting GenAI at each step keeps the AI tool on topic and more likely to provide the most helpful output.

What is prompt engineering?

Prompt engineering is the systematic process of designing clear, contextually relevant, and actionable prompts or inputs for GenAI models. These prompts serve as cues or instructions that guide the behaviour of GenAI models, influencing the generation of outputs or responses. Formulating clear and effective prompts is crucial to obtaining the best possible output.

Writing effective prompts involves understanding both the GenAI model's capabilities and limitations and using this knowledge to construct prompts that guide it toward producing the most relevant and high-quality results.

A good prompt is clear and specific, and it provides some context, making it easier for the AI model to generate a useful response.

A poor prompt is too broad and unclear, and it provides insufficient context, making it challenging for the AI model to create a useful response.

When constructing a prompt, it might be helpful to remember this age-old adage "garbage in, garbage out"; in other words, the quality of the output is dependent on the quality of the input. Vague or poorly constructed prompts will lead to unhelpful, irrelevant, or disappointing outputs.

Note:  Inclusive, non-discriminatory prompts avoid biased outputs.

CLEAR (2023) is an acronym for a set of guidelines intended to improve the creation and refinement of prompts for GenAI tools. It emphasizes five key principles to optimize interactions and achieve better outcomes when using GenAI:

• Concise: Be specific in your prompt. 
• Logical: Maintain a logical flow and order of ideas within your prompt.
• Explicit: Provide the GenAI tool with precise instructions on your desired output format, content, or scope to receive a better output. 
• Adaptive: Experiment with various prompt formulations and phrasing to try different ways of framing an issue and get alternative answers.
• Reflective: Adjust and improve your prompt by evaluating the performance of the GenAI tool based on your own assessments of the answers it gives. 

For examples, see: Maximizing learning with effective GenAI prompt writing: An instructor’s resource for students. 

Source: Lo, L. S. (2023). The CLEAR path: A framework for enhancing information literacy through prompt engineering. The Journal of Academic Librarianship, 49(4), https://doi.org/10.1016/j.acalib.2023.102720 

Guide Disclaimer