Beyond Citation: Describing AI Use in Your Work

We advise authors to acknowledge substantive uses of AI in order to clearly indicate to their audience where the AI ends and the human input begins (for a definition of substantive use, see Resnick and Hosseini). Acknowledgment of the use of AI is slightly different from, and more comprehensive than, citation. Citation points to an exact source of information, and, when possible, it might also provide access to a source (some sources, such as a manuscript in an archive or a personal communication, might not be accessible to a writer’s audience, but they still must be cited so the author can vouch for them). But generative AI output is not a source in the traditional sense. 

While a generative AI tool may direct you to a source and, in the case of some AI tools, even include snippets or excerpts from a source, the AI output itself is not a source. It assembles information from sources based on predictive computation. And AI output can be flawed, omitting important context or surrounding text and containing errors. Further, AI tools may not always provide information with transparent or reliable provenance and thus the information may not be credible—or, rather, it is not always possible to determine its credibility. Thus, when you’re working with such tools and their output, you should click through to the source text, read the full content, and cite the source directly.   

We’ve provided guidance on the mechanics of documenting use of generative AI using MLA citation format. To fully describe use of generative AI tools, however, you may need to provide additional information that goes beyond citation (for example, the educator and researcher Anna Mills uses notes to indicate her use of AI and prompts, and links to or appends AI chat transcripts in her work). This is because finding information and citing it only scratches the surface of how writers are using generative AI tools. People are experimenting with using these tools in various ways—to begin research, to ideate, to help with drafting, and simply to learn about the tools’ capabilities. 

Generative AI and the way writers use it are still evolving, and the way instructors ask students to engage with it—and acknowledge or cite that engagement—is evolving too. Indicating where and how AI has touched your research and writing is just as much a part of academic integrity as citing your research sources is.

Works Cited

Mills, Anna. “AI and College Writing: An Orientation for Students (Draft).” docs.google.com/document/d/1cfs_ubpIEOUpSfSBTxQLU54jyOtMYpDYyRQLcLw4Kn4. Accessed 24 June 2025.

Resnick, David B., and Mohammad Hosseini. “Disclosing Artificial Intelligence Use in Scientific Research and Publication: When Should Disclosure be Mandatory, Optional, or Unnecessary?” Accountability in Research: Ethics, Integrity and Policy, 24 Mar. 2025, https://doi.org/10.1080/08989621.2025.2481949.