Generative Artificial Intelligence (AI) holds promise in improving higher education, offering both opportunities and challenges for universities. AI-powered tools can provide new learning modes, personalize learning experiences, and enable students to receive tailored support and feedback. Additionally, AI can streamline administrative tasks, enhancing efficiency and potentially reducing operational costs. However, the integration of AI also raises important questions about data privacy, ethics, algorithmic bias, and the need for faculty and staff preparation. Balancing AI’s potential and addressing these challenges is important, as the advent of these tools is already reshaping aspects of higher education.
AI and Academic Integrity
The UHM student code of conduct (IV.B.1.a) addresses “Cheating, plagiarism, or other forms of academic dishonesty.” It gives the instructor authority over defining unauthorized assistance, authorized sources, and specifically prohibited behavior in classes. For this reason, instructors are strongly encouraged to:
- Be specific about expectations and limitations on student use of AI in assignments,
- Hold students responsible for the accuracy of facts and sources used in assignments, and to
- Talk through scenarios with classes to provide clarity on expectations
Syllabi and class discussions should make instructor expectations clear with respect to use of AI tools. UHOIC maintains sample statements for adaptation in a syllabus or specific assignments.
Assignment and Assessment Redesign
The widespread availability of AI tools creates both challenges and opportunities for instructors. Similar to the reframing required at the onset of digital calculators (at least for math and science instruction) and internet search engines, assignments and assessments may need redesign to achieve our learning objectives.
Some useful redesign strategies are:
- Chunk assignments with due dates for individual elements that precede final submission: an outline, notes on research articles, and drafts.
- Instructors might consider Google Docs version history and/or draftback, a tool to record the writing process, to review students’ writing progression. Versions of a document can be renamed to show stages of an assignment.
- Replace a writing assignment with an audio file, podcast, video, speech, drawing, diagram, or multimedia project. (make running to AI more work than it’s worth)
- Incorporate AI. Ask students to generate a ChatGPT response to a question of their own choosing, then write an analysis of the strengths and weaknesses of the response.
- Reference class materials or sources that are not available on the internet
- Include visuals — images or videos that students need to respond to
- Connect to current events or conversations in your field
- Ask for application of personal knowledge/experience
(Khan Academy’s, AI for Education)
Additional strategies include:
- Employ authentic and contextualized assessments. Design to require critical thinking, analysis, and application of knowledge in real-world contexts. By tailoring assessments to specific scenarios or case studies, it is challenging for students to rely solely on AI for complete solutions
- Use open-ended questions that prompt students to demonstrate their understanding, creativity, and ability to articulate ideas. These are less likely to have direct answers generated by AI tools, encouraging engagement in original thinking and reflection
- Incorporate elements that evaluate the process students followed to arrive at solutions. This could include written explanations, justifications, or reflections alongside final submissions, demonstrating the individual thought processes and learning journeys
(UH Recommendations on Assessment)
Many resources can be found in this annotated listing maintained by UHOIC, which includes sections on Accessibility, Assessment, References, Copyright, Educational Considerations, Ethics, Pedagogy, Classroom Policies, Prompting, Tools, and Supplemental Resources.
Learning About Generative AI
Experts agree that generative AI tools are here to stay. Teaching and learning about AI is one piece of a need to assist students in learning appropriate 21st century digital skills. Like other new technologies, there are challenges and concerns that come along. Faculty and students are encouraged to experiment with AI tools while staying within legal and ethical parameters to the best of their ability. Proliferation of AI-integrated tools is predicted, and university personnel should be aware of policies and procedures that relate to the adoption of new technologies, including: