Engineering Design Division

Generative AI, like ChatGPT, are large language models. They are artificial intelligence technology designed to understand and generate human-like text based on vast amounts of information. ChatGPT operates using a technique called machine learning, where it learns patterns and information from extensive datasets to produce coherent and contextually relevant responses.

If you wish to use ChatGPT or another generative AI tool for your design project, it's important to keep in mind their limitations. They may sometimes provide inaccurate or biased information. It's important to verify the information with other sources. ChatGPT currently has some strengths and weakness when it comes to its use in research.

What is ChatGPT good for and not good for?

What is it good for?

• Brainstorming ideas
• Narrowing your topic ideas for a research paper, and keywords for searching in library databases.
• Explaining information in ways that are easy to understand
• Summarizing and outlining
• Asking questions (be sure to fact check the results) You can ask a million questions without fear of being judged.
• Translating text to different languages (not completely fluent in every language)
• Helping write or debug computing code

What is it not so good for?

• Library research (not yet). For now, it's best to use Library search, Library databases, or Google Scholar. This may change in the future with more specialized search tools based on generative AI.
   
• Asking for any information that would have dire consequences if it was incorrect (such as health, financial, legal advice, and so on). This is because of its tendency to sometimes make up answers, but still sound very confident.

Prompts are what you choose to type in the chatbox to query ChatGPT. It's important to consider creating well-crafted prompts as it affects the output given by AI models.

The CLEAR framework for prompt engineering is a structured approach used to design effective prompts for language models. It stands for Concise, Logical, Explicit, Adaptive, and Reflective.

Concise: brevity and clarity

A concise prompt allows AI language models to focus on the most important aspects of the task, which results in more pertinent and precise responses. Clarity is also crucial in order for the AI-generated content to meet your needs.

Example:

Use a more concise and explicit prompt such as “Explain the process of photosynthesis and its significance” instead of “Can you provide me with a detailed explanation of the process of photosynthesis and its significance?”

Logical: structured and coherent prompts

When crafting prompts, consider the logical flow and order of ideas. This helps AI models comprehend the context and relationships between ideas.

Example:

List the steps to write a research paper, beginning with selecting a topic and ending with proofreading the final draft” is a logically structured question

Explicit: clear output specifications

Provide precise instructions regarding the desired output and format. This helps reduce irrelevant and unanticipated results. Be specific about the type of information you seek and how it should be conveyed.

Example:

Rather than prompting, “What are some renewable energy sources?”, opt for a more explicit version like, “Identify five renewable energy sources and explain how each works.”

Adaptive: flexibility and customization in prompts

This component is the first that more so builds off the others listed above. If you receive results that do not match your information need, try rephrasing. It's important to be flexible and willing to try new approaches.

Example:

If an initial prompt such as “Discuss the impact of green space on mental health” does not give enough information, consider a more focused and adaptable prompt such as “Examine the relationship between exposure to the outdoors and anxiety in college students.”

Reflective: continuous evaluation and improvement of prompts

This component also builds off the others. Take the time to assess the prompt results and evaluate the performance of the AI model. Apply your insights to future prompts. This mindset is helpful to adapting to the ever-changing field of AI.

Example:

After acquiring an AI-generated list of strategies for effective time management, evaluate the relevance and applicability of each strategy. Consider the target audience's needs, and use this information to tailor future prompts to generate content that better addresses specific challenges or contexts.

Again, it's important to verify information produced by generative AI tools as they can make mistakes and hallucinate ("make-up") responses. A way to do this is to practice lateral reading. Watch the video below to learn how to perform lateral reading.

The original source for the CLEAR prompt engineering framework can be found in the following paper:

Lo, L. S. (2023). The CLEAR path: A framework for enhancing information literacy through prompt engineering. The Journal of Academic Librarianship, 49(4), 102720. https://doi.org/10.1016/j.acalib.2023.102720

Some components of this page were adapted from "Student Guide to Chat GBT", by the University of Arizona Libraries licensed under a Creative Commons Attribution 4.0 International License.”