Lessons from Designing a Graduate-Level AI Course for Research and Teaching

Abstract

This session presents a faculty experience designing a graduate-level course that prepares students and faculty to use generative AI critically, ethically, and practically in academic research and teaching. Co-designed by Zubair Barkat and Sean Johnson, AI Essentials for Researchers / AI for Aggies is a planned 6100-level course in Utah State University’s College of Arts and Sciences that aims to help make both the College and USU more AI-ready. The course builds directly on the bootcamp-style Sociology of Artificial Intelligence course delivered in Spring 2026, extending its hands-on, in-class model from undergraduate sociological inquiry to advanced graduate research training across disciplines. We envision a 25-person learning community that includes graduate students alongside approximately 10–12 faculty members, allowing participants at different career stages to experiment together, compare AI tools, and reflect on disciplinary differences in AI adoption. One design tension we are navigating is whether to offer the course for two or three credits, balancing depth, rigor, and sustained practice with graduate students’ course loads and faculty time commitments. The syllabus is designed as a living document, adaptable to the rapid evolution of AI tools and debates. Across the semester, participants will practice prompt engineering, build custom research assistants, use retrieval-augmented generation, design research instruments, explore AI-assisted coding and Python workflows, and examine ethics, bias, transparency, intellectual property, climate change, employment, visualization, and academic writing. A final capstone project asks students to leverage AI to address a real-world challenge while demonstrating practical functionality, societal relevance, and critical awareness of limitations, risks, and safeguards.

Presenters