Research Integrity and Safety

FAQs: AI in Research

FAQs regarding the use of AI in your research

Can I trust the content generated by GenAI tools, and what if it is inaccurate?

AI-generated content can be inaccurate and biased, introducing potentially plagiarized, false, and even fabricated/imagined content. It is the researcher’s responsibility to verify the accuracy, validity, and appropriateness of all output. GenAI tools cannot be held accountable for the inaccurate material they generate. Researchers must confirm the accuracy of all AI-generated materials and could be held responsible for research misconduct should fabrication, falsification, or plagiarism be contained within their materials.

Should I document that I used GenAI in my research?

Yes. The reproducibility of research is what enables us all to build on prior work. Transparency and completeness of reporting is essential to this. Any use of GenAI tools should be clearly cited, identifying the GenAI tool and describing how it was used to support the research and publication. Check with your publisher and with the AI tool source as some have specific policies regarding use and stock language for appropriate citation.

Can a GenAI tool be listed as an author?

A GenAI tool cannot be listed as an author. AI models “cannot meet the requirements for authorship as they cannot take responsibility for submitted work. As non-legal entities, they cannot assert the presence or absence of conflicts of interest nor manage copyright and license agreements” (Committee on Publication Ethics [COPE], Authorship and AI tools | COPE: Committee on Publication Ethics, 2023; Zielinski et al., 2023; Flanagin et al., 2023) and also “because any attribution of authorship carries with it accountability for the work, and AI tools cannot take such responsibility” (Nature, 2023, p 612; Hosseini, Rasmussen & Resnik, 2023).

Should I enter information into AI models or AI-Powered Search Engines?

For public data or internal data that you generate, there are no protection requirements under UGA EITS Policy , though be aware of potential consequences noted in this guidance. Entry of sensitive or restricted data as defined in the UGA EITS policy is prohibited without express authorization of the university. A contractual agreement outlining data processing, privacy, and security responsibilities in a form approved by the Office of Legal Affairs and Office of Information Security must be in place prior to any sensitive or restricted data entry to these systems.

Can I use GenAI to develop my thesis or dissertation?

Per UGA Graduate School Policy , use of GenAI in theses and dissertations is considered unauthorized assistance according to the Academic Code of Honesty and is prohibited unless specifically authorized by members of the advisory committee for use within the approved scope. If approved by the advisory committee, the extent of GenAI usage should be disclosed in a statement within the thesis or dissertation.

Can I use Turnitin to screen research papers for plagiarism?

The Turnitin suite of tools is designed for use in the classroom to support students in learning how to engage in appropriate citation and referencing techniques and support instructors in evaluating the academic integrity of student work. Turnitin is sometimes used to screen trainee work in non-classroom settings, but users should keep in mind that this tool was specifically designed for classroom use. It is not recommended that this or any other tool be used for any other purpose for which it was not designed. It is important you fully understand tool capabilities and limitations and the implications of its use in research paper assessment, including ways in which content entered may affect the risk of false positives when subsequent drafts are tested. Additionally, the potential negative impact on intellectual property of uploading what may be high-impact research into such an engine must be understood. Such considerations associated with research have resulted in vendors specifically designing tools for this particular application.

Questions and other feedback

 Please submit any questions or other feedback on this guidance to the Office of Research here: oris@uga.edu.

 References

Thanks to multiple peer institutions for informing the content of this page, including the University of Michigan, Virginia Polytechnic Institute and State University, and Cornell University.