While generative AI programs are incredibly innovative, they also come with some serious concerns. We'll only cover two here.

• They may cite made up resources. ChatGPT and other large language models (LLMs) will readily create reading lists and work cited pages for your papers, but it is limited to the data they have been trained on and depending on the dataset, my not contain up-to-date information. To get around this, LLMs will completely make up citations to "support" the text it generates. Here, in our library, we've encountered instances of faculty and students looking for books and articles that don't exist.
   
• They may plagiarize resources. So where does ChatGPT, and other AI chatbots get their data? Generative AI programs require text and images to be "scraped" from existing resources and entered into their datasets. In some cases, companies are transparent as to where they get their work. In many cases though, we don't know and evidence shows that scraped resources may come from copyrighted works without the original creator's permission or knowledge. And it's not just major works. Anything freely found online is targeted, including product reviews and public comments and posts. Further, these programs also lack any way for authors or artists to have their works removed from the program. In fact, when you use these programs, they often are scraping your inquiries into their database as well.

You may be wondering why plagiarism is a problem if generative AI is supposed to be reshaping scraped text and images into something new. The problem is that it's easy to purposefully or accidentally prompt these programs into reusing other people's works, including creating art in another person's copyrighted style or pulling large blocks of text from copyrighted sources, such as entire pages of books. This means that the "generative" and "new" works that these AI programs create may actually be someone else's work. In addition, even when AI generative works don't directly plagiarize, they often reuse ideas and steal core concepts from copyrighted works without crediting the original creators. This is considered plagiarism and academic dishonesty.

Adapted from Arkansas State University Plagiarism libguide

APA

Consult your professor before you use AI in your research and writing. If you've used ChatGPT or another AI tool in your research, describe how you've used it in the method section or a comparable section in your paper.

For APA, use the reference template in section 10.10 in the manual.

This is an example for ChatGPT:

In-text (OpenAI, 2025)

Reference list: OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat

For a longer explanation, see APA style blog entry for Creating a reference to ChatGPT or other AI models and software.

MLA

Consult your professor before using AI in your research and writing. 

MLA does not recommend using the AI tool as an author, rather, cite it whenever you use a tool to paraphrase or summarize content. 

Acknowledge all functional uses of the tool (like editing your prose or translating words) in a note, your text, or another suitable location.

Example of citing a tool that summarizes text

Paraphrased in Your Prose

In Mansfield Park, physical locations like Mansfield Park and Sotherton reflect the morality and choices of the people who live in them (“Describe the theme”).

Works-Cited-List Entry

“Describe the theme of nature in Jane Austen’s Mansfield Park” prompt. ChatGPT, model GPT-4o, OpenAI, 23 Sept. 2024, chatgpt.com/share/66f1b0a0-d704-8000-be9a-85f53c850607.

Taken from "How do I Cite Generative AI in MLA Style?" https://style.mla.org/citing-generative-ai-updated-revised/