A Discipline-Agnostic AI Literacy Course for Academic Research: Architecture, Pedagogy, and Implementation

AI literacy requires sustained, practice-based training beyond brief workshops: The paper identifies a critical gap between narrow technical courses and superficial general-literacy interventions, arguing that students need structured, ongoing instruction to develop the "critical judgment" necessary for responsible AI use in research—not just tool proficiency.

Literature review is an ideal entry point for teaching responsible AI use: The course focuses specifically on AI-assisted literature review as a foundational competency, breaking it into four sequential modules (comprehension, taxonomy construction, gap identification, and synthesis) that build students' ability to verify AI outputs and catch errors at each stage.

Explicit verification practices and attribution standards are essential safeguards: The curriculum embeds "an explicit verification discipline and standardized AI attribution practice" into every module, suggesting that responsible AI use requires systematic, built-in checks rather than relying on user judgment alone.

Hallucination detection shows the largest confidence gains: Pre- and post-course data indicate that students' self-reported confidence in detecting AI hallucinations improved substantially, suggesting this is both a learnable skill and a priority concern for academic researchers using generative AI.

Discipline-agnostic, prerequisite-free design increases accessibility: By serving upper-level undergraduates and graduate students across all fields with differentiated expectations, the course model demonstrates that responsible AI literacy can be taught as a universal research competency rather than a specialized technical skill.