Skip to Main Content

Generative Artificial Intelligence

This guide provides tips around using (and not using) generative AI for library-related tasks such as searching for information, tracing claims to sources, writing papers, and citing sources.

Laying the Groundwork

General-purpose text generating AI tools can help you work through your research and writing process, and writing prompts that are specific about the type of response you want can shift the generative AI to generate responses that are closer to what you need.
Those responses can be used to help YOU improve your ideas or search strategies, providing ideas or inspiration for directions you could take, and then YOU choose which ideas to research further in human-created sources or in your own original research.

 

Example Prompt Introduction: "For the next questions I'm going to ask you, you are a professor in __________, helping me to work on a research project about __________. You will guide me with ideas and suggestions for my project to be engaging and to be the best research project I can complete. Please assist me with this as I ask the next few questions."  

Brainstorming and Ideation

1. Brainstorming / Ideation

  • State Your Goal: Clearly articulate what you want to achieve, such as finding a topic, direction, or specific question.
    • Example Prompt: "I want to explore the impact of social media on mental health among teenagers. What are some important  ideas related to that topic that I could research?"
  • Try Broad Prompts: Start with a general question related to your area of interest.
    • Example Prompt: "Please give me a list of ways that social media affects mental health in teenagers"
  • Generate Ideas: Use AI to come up with potential research questions based on your initial idea.
    • Example Prompt: "I'm writing a 5-page paper summarizing research on an area of my choice for a communications 300 class. I'm interested in what are the psychological effects of social media usage on teenagers, but I think that topic is too big and too focused on psychology. What are some research questions related to that topic that would be the right size for a 5-page paper that I could look into?" 
  • Explore Different Angles: Look at your topic from different perspectives to find new insights.
    • Example Prompt: "I think I'm going to focus on the question, 'How does social media influence self-esteem and body image among teenagers?' What are the important sub-topics that would be important to look into for that topic?"
    • Follow-up Prompt: "What questions should I answer?"  
    • Follow-up Prompt: "What related topics might be easier or better to research?"  
  • Iterate: Use feedback to improve and refine your research question.
    • Example: Based on feedback, refine the question to: 'What is the relationship between social media usage and anxiety levels in teenagers?"
  • Experiment: Try different prompts and approaches to see what works best.
    • Examples: Replace "Instagram" with "TikTok" or "anxiety" with "depression" in the prompts above to see differences in responses and help yourself feel out which you want to explore in more depth. Or rephrase the framing situation to cast the generative AI in the role of someone with different expertise, such as a teenager active in social media to ask about the topic of teens and social media, to get it to give a different set of ideas to explore. 

Refining and Focusing a Question

2. Refining and Focusing

  • Narrowing Down: Use AI to narrow your list of potential questions, helping you find the most feasible ones.
    • Example: From a list of questions, narrow down to: 'How does Instagram usage correlate with anxiety in high school students?'
  • Specificity: Make your questions more specific and focused with AI's help.
    • Example: Refine to: "How does daily Instagram usage affect anxiety levels in high school students aged 14-18?"
  • Clarity: Ensure your research question is clear, concise, and easy to understand by refining the language with AI.
    • Example: Clarify to: 'What is the impact of daily Instagram usage on anxiety levels among high school students aged 14-18?'

Critiques of Your Ideas

3. Critical Evaluation of Your Ideas

  • Identify Gaps: Use AI to uncover any missing elements or areas that need further exploration in your research question.
    • Example Prompt: "Below, I've listed the ways Instagram affects high schooler's anxiety levels. What other factors have I missed or not thought of?" 
  • Potential Challenges: Explore potential challenges or difficulties you might face during your research.
    • Example Prompt: "I'm going to argue that the curated images in Instagram increase high schooler's anxiety levels by changing their perceptions about the happiness and success of their peers. What are some weaknesses or flaws in this argument or in the evidence that supports it?" 
  • Ethical Considerations: Identify any ethical issues related to your research question.
    • Example Prompt: "I plan to bring high school students to the lab and have half of them view happy Instagram posts of other high schoolers and half of them view Instagram posts of a variety of neutral content and then measure their anxiety levels. What ethical issues particular to this study should I address in my human subjects institutional review board submission?" 

Developing Search Strategy / Terms

4. Developing Your Search Terms / Strategy 

  • Use to Generate Alternative Terms or Synonyms:  Once you have identified the question or topic you want to pursue, use AI to identify synonyms or alternate ways to describe a topic, which could include the singular and plural versions of terms. 
    • Example: If the topic is “anxiety in teens using social media,” AI might propose alternative terms for “anxiety” such as “stress,” “distress,” or “apprehension.” For “social media,” it could suggest synonyms like “social networks,” “digital platforms,” or “social networking.
  • Field-Specific Terminology: Professions or fields of study have specific language that one may not be familiar with, using AI could help to identify profession/field terminology. 
    • Example: Social media and psychology fields have specific terminologies. For "social media," AI could highlight terms like "user engagement," "digital presence," or "platform algorithms." In psychology, AI could suggest "cognitive-behavioral patterns," "cyberbullying impacts," or "dopamine-driven addiction."
  • Identifying Spelling Variations of Words:  Can help to identify variations in the spellings of words across various languages. 
    • Example: "Behavioral stress" in American English and "behavioural stress" in British English can be differentiated by AI. The same applies to spellings of "modeling" vs. "modelling" in the field of psychology.
  • Suggest Related Concepts:  Using AI can help users discover suggestions that broaden their search, making the exploration of a topic more thorough and comprehensive
    • Example:  AI could suggest expanding "teen anxiety on social media" to explore: 
      • The role of "online validation and self-esteem."
      • "Impact of social media influencers on mental health." 
      • "Parental monitoring in digital interactions." 
  • Combine List of Terms with Boolean Operators and Add Quotation Marks: If you have a list of terms you'd like to combine, you can use Boolean operators like "AND," "OR," and "NOT" to refine your search. Additionally, you can use AI to add quotation marks around specific terms or phrases to ensure they are searched exactly as written, resulting in more precise and relevant results.