Skip to Main Content

Generative Artificial Intelligence: Generative AI Prompting

Resources on Generative AI

What are AI prompts?

An AI prompt is a question, instruction, or piece of text you give to an artificial intelligence system to guide its response. It's how you "talk" to the AI tool to get it to do something like answer a question, write a paragraph, or solve a problem.

A prompt can be just a few words, a sentence, or even a few paragraphs. The way you write your prompt has a big impact on how accurate and useful the AI’s response will be. The more clear instructions and context you give to a language model like ChatGPT, the better the response will match what you’re looking for.

The CLEAR framework for successful AI prompts

The CLEAR framework, authored by University of Mexico Librarian Leo S. Lo consists of five key components:

C-Concise: Prompts should be as direct and clear as possible. AI tools such as ChatGPT are most successful when given direct, specific prompts. 

L-Logical: Prompts should be structured logically with a coherent order of ideas. The more logical your prompt is, the stronger the response from the AI tool (ChatGPT etc.) will be. 

E-Explicit: Provide clear output instructions. Your prompt should be specific on the type and amount of information you are looking for. 

A-Adaptive: Flexible and customizable prompts. If your initial prompt was too general or too specific, experiment with different phrasing or instructions.

R-Reflective: Effective prompts are continuously evaluated and improved. Maintaining a reflective mindset when using AI tools allows for stronger evaluation of the responses.

Strong prompt examples:

In 2-3 paragraphs, define inflation, and explain its causes.

Provide a 500 word summary of the connection between Instagram filters and body dysmorphia.

Weak prompt examples: 

Tell me about the causes of inflation.

What is the connection between social media and mental health?

Lo, Leo S.. "The CLEAR Path: A Framework for Enhancing Information Literacy through Prompt Engineering." Journal of Academic Librarianship , 4 (2023). doi:https://doi.org/10.1016/j.acalib.2023.102720.