Skip to Main Content

Generative Artificial Intelligence for Students

The previous page highlighted some of the limitations of ChatGPT and cautioned you against using it as a replacement for Library databases and search engines like Google. This page will focus on the situations where it can be beneficial to use an AI chatbot's assistance for college work. As always, be sure to confirm AI usage guidelines with your instructor before using AI assistance with your coursework. 

College Work

picture of a student working at a desk with the personification of an AI chatbot sitting next to him.Here are 4 scenarios where AI chatbots like ChatGPT can help you with your college research assignments:

1. Brainstorming ideas, topics, and research questions: When you have a new assignment, and don't know where to begin, asking ChatGPT to summarize the current state of research on a topic may be just what you need to begin the brainstorming process. Chatbots can be useful for providing entry-level knowledge on a topic in a conversational tone, but keep in mind that highly niche or emerging topics may not be covered.

Prompt example: "I’m writing a research paper for Sociology and I need help coming up with a topic. I’m interested in topics related to climate change. Please give me a list of 10 topic ideas related to climate change."

2. Gathering keywords to use in Library searches: When you are starting to research a new topic, it can be difficult to come up with search terms that are likely to appear in the sources you want to find. You can ask a chatbot for a list of keywords on specific topics and use those keywords in your searches.

Prompt example: "Give me a list of keywords related to innovation in biomedicine."

3. Suggesting synonyms for keywords: In addition to providing you with a strong set of keywords to begin your searching, chatbots can also act as a thesaurus and provide you with several synonyms to those keywords. Try asking a chatbot for a list of synonyms or similar concepts for select keywords you've identified as best for your research.

Prompt example: "Suggest 10 synonyms relevant to engineering that I can use for the following words: manufacture; properties; debris; bricks." 

4. Summarizing text: You can use chatbots like ChatGPT to summarize readings for you by pasting in the text and asking for a summary. You can give additional prompts to refine the summary, such as asking the chatbot to assume a certain level of knowledge on the part of the reader and to limit the summary to a certain amount of sentences. Keep in mind, most chatbots have a word limit that will determine how much text you can paste in at one time. 

Prompt example: "Summarize this text in five sentences or less. Summarize it for me as if I am a first year undergraduate student."

Sources: Choice and Brown University

Fact-Checking

When using generative AI chatbots for research purposes, honing your fact-checking abilities is crucial. First and foremost, don't take what ChatGPT tells you at face value.

1. Find better coverage: Look to see if other reliable sources contain the same information and can confirm what ChatGPT says. This could be as simple as searching for a Wikipedia entry on the topic or doing a Google search to see if a person ChatGPT mentions exists. The short video below [1:38] will show you an example of fact-checking in this way. 

 

2. Verify citations: If a generative AI tool provides a reference, try to confirm that the source actually exists. Try copying parts of the citation (like the title of an article) into a search tool like Google Scholar or the Library's OneSearch. Do a Google search for the lead author. Check for the academic journal in the Library's journal finder. The video below [2:36] will provide you with an example of checking a citation in this manner. 

If it turns out the source is real, you'll need to confirm that it actually contains what ChatGPT says it does. Read the original source or its abstract/summary. 

How to Write Better Prompts

Fact-checking is only half of the story. If you want to get the most out of generative AI tools, you'll also need to become adept at asking the "right" questions in the "right" manner. Prompt engineering -- the deliberate crafting and optimization of questions/requests given to AI models to elicit desired outputs -- is increasingly recognized as a crucial skill for maximizing the effectiveness of AI like ChatGPT.

So what are the best ways to craft your prompts? 

  1. Be as specific as possible in your request. ChatGPT can't read your mind. If its outputs are too long, ask for brief replies. If outputs are too simple, ask for expert-level writing. The less the AI has to guess at what you want, the more likely you’ll get it.
  2. One way to be specific is to give the AI a "persona" in your request and roleplay with it. For example, "You are an expert career advisor and I am a STEM student trying to pick a college major..." 
  3. Tell the AI what your goal is so it can help tailor the response. 
  4. Provide examples and ask for additional suggestions. Giving the tool a framework can improve its answers. 
  5. Provide one side of a debate and ask the tool to provide other perspectives. 

Further reading: Prompt engineering (OpenAI)

Sources: OpenAI and Pace University


In the video below [11:41], AI experts discuss how to effectively prompt AI like Midjourney, ChatGPT, and Microsoft’s Bing (now Copilot), as well as how to take the lead, weaving your own expertise into the interaction.

Accused of Cheating?

Some instructors might use tools that claim to be able to detect whether a text was written by generative AI. However, those tools are "neither accurate nor reliable." They give both false negatives and false positives. Additionally, AI detectors have been found to be more likely to label text written by non-native English speakers as AI-written

If you are falsely accused of cheating with generative AI*, please refer to the following article and video, which give helpful recommendations for disproving those allegations:

*Please note: these recommendations assume that you have used AI within the guidelines provided by your instructor. Using AI in a manner explicitly prohibited by your instructor will, most likely, negate these recommendations. 

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.