Redesign structurally: grade the reasoning behind AI-assisted work and teach AI-use rules and AI literacy explicitly¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article's design implication is structural rather than tool-based: because generative AI develops quickly and educational tools cannot keep pace with commercial systems, courses should make how a student worked visible and gradable. Assessment "can be designed to target the reasoning behind AI-assisted work and to require students to reason before they consult AI". The self-set rules of the strongest students can be taught, with students writing, testing and revising their own rules, and AI literacy in Long and Magerko's sense belongs in the curriculum for students and staff.

Design Implications¶

Context¶

Requirements¶

• Assessment tasks that require students to reason before consulting AI and expose their working process
• Explicit curriculum time for AI literacy, including a working understanding of what the models do, for students and staff

Constraints¶

• A course or tool built around one version of generative AI is out of date by the time it runs

Target Learners¶

• higher-education students in programming and data analytics courses
• teaching staff

Target Learning Goals¶

• self-governance of AI use
• reasoning behind AI-assisted work
• critical evaluation and understanding of AI models

Related Strategies¶

• Redesign AI-era physics courses around an unaided written examination with mock exams, prepared tutorial discussion, and research-shaped tasks as bonus work with scaffolding
• Embed AI training for pre-service science teachers within science pedagogy and inquiry-based teaching rather than general tool operation
• Faculty development for GenAI-era teaching innovation should address professional identity, AI literacy, and teaching self-efficacy alongside knowledge
• Strategies for integrating GenAI in STEAM teaching: outcome-anchored lesson planning, differentiation, inquiry and project variants, assessment redesign, and integrity as pedagogy

Examples¶

-

Key Sources¶

• Azimi, S. (2026). AI Literacy over Tool Design: A Mixed-Methods Study of Scaffolded versus Unrestricted Generative AI in Programming Education. Delft University of Technology. https://arxiv.org/abs/2609.16784. doi:10.48550/arxiv.2609.16784