Bridging the Skills Gap: A Course Model for Modern Generative AI Education

AI summary

Overview

Research area: AI education and curriculum design — specifically the teaching of generative AI tool use in higher education, placed in the arXiv AI Safety & Ethics category.

Technical level: Beginner-Friendly. The paper is about course design, pedagogy, and student perception rather than technical AI methods.

Scope (one sentence): The paper describes the context, implementation, and reported impact of a university course that teaches undergraduate and graduate Computer Science students how to apply existing generative AI tools in software development, drawing on two mixed-method surveys and reflections from the instructor and a graduate student co-author.

What This Paper Is About

Educators have been hesitant to teach generative AI tools in classrooms, and the authors identify two resulting disconnects: industry increasingly values generative AI competency while higher education does not teach it, and students are experimenting with these tools on their own without formal guidance. The paper's goal is to address that gap by presenting a course developed at a private research university that teaches students to use generative AI tools responsibly and expertly, and by offering a model other departments can replicate.

Key Contributions

1. Names and frames two disconnects — the industry-versus-higher-education gap in generative AI competency, and the gap between students' informal experimentation and formal instruction.
2. Presents a concrete course model developed at a private research university to teach applications of existing generative AI tools in software development to undergraduate and graduate Computer Science students.
3. Reports student feedback from two mixed-method surveys, which the authors say showed students overwhelmingly found the course valuable and effective.
4. Provides a dual-perspective account — co-authored by the instructor and one of the graduate students — combining data analysis with reflections from both viewpoints, plus recommendations for replication in and beyond Computer Science departments. This version is extended with technical appendices.

Main Findings

• A stated disconnect between industry and academia: Generative AI competency is described as increasingly valued in industry but not in higher education.
• Unsupervised student use: Students are reported to be experimenting with generative AI without formal guidance.
• Coverage gap in top departments: The authors observe that while consistently top-ranked U.S. Computer Science departments teach the mechanisms and frameworks underlying AI, few appear to offer courses on applications for existing generative AI tools. The abstract does not state how many departments were examined or how this was determined.
• Positive student reception: Two mixed-method surveys indicated students overwhelmingly found the course valuable and effective. The abstract does not report survey size, response rates, question wording, or any quantitative results.
• Computer Science is disproportionately affected: The abstract singles out Computer Science trajectories as particularly impacted by these gaps.
• A broader claim: The authors argue students across all fields — not just Computer Science — must be taught to responsibly and expertly harness AI tools to ensure job market readiness and positive outcomes.

Methodology in Plain English

The authors began from observed problems — educators avoiding the topic and students using the tools anyway — and responded by designing and running an actual course. The course was built at a private research university and targeted both undergraduate and graduate Computer Science students, focusing on how to apply existing generative AI tools in software development rather than on how those tools are built internally.

To assess the course, they ran two mixed-method surveys, meaning the surveys gathered both numerical responses and open-ended, qualitative ones. The paper then combines analysis of that survey data with personal reflections from two people in different roles: the instructor who taught the course and a graduate student who took it. Because only the abstract was available, the specific survey instruments, the number of participants, the analysis techniques, and the course's detailed structure (assignments, topics, duration) are not described here — the full paper, including its technical appendices, is where those details would appear.

Why This Matters

Impact on research: The paper contributes a documented, replicable example to the emerging literature on generative AI in education, moving the discussion from general debate about whether to teach these tools toward a concrete account of how one course did it and how students responded.

Real-world applications:

• Curriculum design: Departments can adapt the described course structure when building their own generative AI application courses.
• Cross-disciplinary teaching: The replication recommendations are explicitly framed as applying beyond Computer Science, so fields such as humanities, social sciences, or professional programs could use the model as a starting template.
• Responsible-use instruction: The emphasis on using AI tools "responsibly and expertly" speaks directly to academic integrity and AI ethics conversations on campuses.
• Reducing educator hesitancy: By presenting a working example rather than a theoretical argument, the paper gives hesitant instructors a reference point.

Industry relevance: The paper's framing centers on the mismatch between what employers value and what universities teach. If the authors' premise holds, courses like this one are positioned as a way to improve graduates' job market readiness in a labor market that increasingly expects fluency with generative AI tools.

Future Directions

• Replication at other institutions: The paper offers recommendations for reproducing the course, but whether the model transfers to different types of universities, class sizes, and resource levels is an open question.
• Expansion beyond Computer Science: The authors recommend replication beyond CS departments; how the course would need to change for non-programming disciplines is not resolved by the abstract.
• Curriculum currency: Generative AI tools change rapidly. The abstract does not address how a course built around specific existing tools stays current, which is a natural follow-up concern.
• Measuring longer-term outcomes: The reported evidence is student perception from two surveys. Whether the course produces durable competency, ethical judgment, or employment advantages would require follow-up study that the abstract does not describe.
• Institutional adoption: The abstract identifies educator hesitancy as a core problem but does not say what would move institutions to adopt such courses at scale.

Target Audience

• University faculty and instructors, especially in Computer Science, considering whether and how to teach generative AI tools.
• Curriculum designers and department administrators looking for a concrete, classroom-tested model to adapt.
• Education researchers studying the integration of AI into higher education and the industry-academia skills gap.
• AI ethics and safety practitioners interested in how responsible AI use is taught at the undergraduate and graduate level.
• Graduate students and teaching assistants who may be involved in designing or delivering similar courses, since the paper is partly written from a student's perspective.

Authors’ abstract

Research on how the popularization of generative Artificial Intelligence (AI) tools impacts learning environments has led to hesitancy among educators to teach these tools in classrooms, creating two observed disconnects. Generative AI competency is increasingly valued in industry but not in higher education, and students are experimenting with generative AI without formal guidance. The authors argue students across fields must be taught to responsibly and expertly harness the potential of AI tools to ensure job market readiness and positive outcomes. Computer Science trajectories are particularly impacted, and while consistently top ranked U.S. Computer Science departments teach the mechanisms and frameworks underlying AI, few appear to offer courses on applications for existing generative AI tools. A course was developed at a private research university to teach undergraduate and graduate Computer Science students applications for generative AI tools in software development. Two mixed method surveys indicated students overwhelmingly found the course valuable and effective. Co-authored by the instructor and one of the graduate students, this paper explores the context, implementation, and impact of the course through data analysis and reflections from both perspectives. It additionally offers recommendations for replication in and beyond Computer Science departments. This is the extended version of this paper to include technical appendices.