Generative AI Graduate School Statement

Graduate students, GTAs, and Graduate Program Directors:

Some of you may have concerns about generative AI tools like ChatGPT, and how these technologies might impact your learning, teaching, and research. Much attention has been paid in the media to the potential use of this technology by students to submit AI-generated text as their own work.

While there are legitimate concerns about the potential misuse of new AI technologies that we need to thoughtfully address together, there is also tremendous potential for these tools to positively augment human capabilities and productivity across many disciplines. The full extent of this impact remains to be seen, but it seems likely that most fields will see changes in what it means to be an expert practitioner who leverages AI judiciously. As learners and educators, we have a responsibility to carefully consider how best to prepare students to thrive in a world where AI is increasingly ubiquitous. This does not mean an abandonment of human-centered pedagogies, but some changes may be prudent to ensure students develop the skills needed to evaluate information, think critically, and apply knowledge, while also learning to utilize AI technologies in a responsible, ethical manner.

This cautiously optimistic position must be balanced by a concerted effort to address concerns about academic integrity. The core principles and spirit of the Graduate Honor Code and Undergraduate Honor Code remain true and relevant. As outlined in Article 1 of the Graduate Honor Code, we expect graduate students to exercise honesty and ethical behavior in all academic pursuits, whether coursework, research, or teaching. Using ChatGPT or other AI to generate work and misrepresenting it as one’s own original work (replacing the student’s effort expected by the course instructor with AI or using AI to replace rather than augment learning) would constitute cheating or plagiarism under the Code. The extent to which AI is appropriate for any given coursework is a decision faculty will need to make on a course-by-course or assignment-by-assignment basis. Whether the decision is made to prohibit or encourage use, the guidelines and principles governing the appropriate use in a given course should be clearly communicated in course documentation and via direct discussions with students.

As you consider how to use AI in your own efforts or address these issues in your classes, keep the following guidelines in mind:

1. Familiarize yourself with the technology and its implications.
2. Have open discussions in classes and labs on responsible AI use and citation expectations. Ground your conversation in the principles contained in the Undergraduate and/or Graduate Honor Code to establish that this is a matter of academic integrity.
3. Ensure that syllabi directly address the use of generative AI and frequently communicate AI-use expectations and limitations in courses and research.
4. Focus assignments on applying skills and knowledge over reciting facts. Emphasize critical analysis, individual perspectives, and personal reflection where appropriate.
5. Consider formative assessment strategies that emphasize the learning process over summative assessment strategies that emphasize the finished product.

TLOS is offering a series of workshops and engagements to support instructors, including graduate student instructors, as they consider how to approach these issues. Instructional specialists are also available for individual consultations.

With vigilance and some adaptation in our teaching methods, we are confident we can maintain the integrity of student work. Please let us know if you have any concerns or ideas for best practices. We are happy to work with you and provide training opportunities as needed.

Aimée M. Surprenant
Dean of the Graduate School

Dale Pike
Associate Vice Provost for Technology-enhanced Learning
TLOS