Abstract:

Instructors across many disciplines are encountering a new instructional dilemma: students have immediate access to generative artificial intelligence tools that can produce technical solutions, yet those same courses are intended to develop the underlying knowledge and reasoning required to produce those solutions independently. When AI systems can generate working code or written responses, traditional assignments risk becoming exercises in tool use rather than opportunities for disciplinary learning. At the same time, ignoring these technologies prevents students from developing the AI literacy that will be increasingly expected in professional environments. This tension requires instructional approaches that integrate AI while preserving the learning processes that courses are designed to develop.

This paper describes the redesign of a database programming assignment in an undergraduate web development course to intentionally incorporate generative AI as part of the learning process. Rather than allowing unrestricted AI use, the assignment structures student interaction with AI systems through a sequence of design constraints. Students first complete foundational database tasks independently to establish core understanding. They are then required to use generative AI to extend their work by prompting the system to generate a new database table structure and associated SQL statements.

Students must document their prompts, evaluate the generated output, test the SQL statements in a live database environment, and revise the code when errors or inconsistencies occur. This structured approach shifts AI use from simple answer generation toward an interactive process of verification and revision. By requiring documentation of prompts and testing of generated outputs, the assignment makes student–AI interaction visible and assessable. Students must demonstrate how they interpret AI-generated code, identify limitations, and apply disciplinary knowledge to correct or refine the results. In this way, the activity encourages students to treat AI as a tool that supports reasoning rather than as a substitute for it.

The paper discusses the instructional design principles guiding this implementation and examines how structured prompting, documentation, and verification can support meaningful engagement with generative AI tools. Although the case example comes from a technical programming course, the approach is broadly applicable to other disciplines where students interact with AI-generated outputs. Embedding AI literacy directly within course assignments allows instructors to support responsible AI use while maintaining the rigor of disciplinary learning.

Keywords:

AI literacy, instructional design, prompt engineering, technical education, generative artificial intelligence.