Assessment design levers that make learning visible and reduce the payoff from generic generated text¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article recommends concrete levers: staged submissions (proposal, outline, annotated bibliography, draft, final) that create a record of development; short oral or in-class verification with standardised prompts and preparation time; and personalised, critique-based tasks such as 'evaluating an AI-generated answer and correcting its authorities'. These levers allow institutions to permit legitimate language support while still detecting substitution through construct-relevant evidence.

Design Implications¶

Context¶

Requirements¶

• Verification components must not become a language performance test; prompts should be standardised with preparation time provided
• Staff need calibration through exemplars, short training modules, and moderation guidance

Constraints¶

• Inconsistency between staged work and explanation should trigger further inquiry rather than an automatic misconduct finding

Target Learners¶

• EAL and multilingual students in higher education

Target Learning Goals¶

• Evidencing disciplinary reasoning and judgement over time rather than fluent text production

Related Strategies¶

• Replace detector-centred assessment with a four-stage process-oriented model: declared starting point, versioned drafting, process artefacts, and targeted oral verification
• Ongoing Feedback
• Annotated Bibliographies
• Stage GAI use in assessment: permit exploration, restrict independent reasoning, require justification
• Teach AI-mediated judgement and make process, judgement and transfer visible in assessment
• Provide explicit instructional boundary cues, process-oriented assessment, and peer modeling of verification to frame responsible GenAI use
• Require process transparency artefacts: process logs, prompt records, reflective commentaries and oral defences
• Prevent AI-assisted plagiarism through student education, signed declarations, draft submissions, clear guidelines, and close monitoring

Examples¶

-

Key Sources¶

• Grace Li. (2026). GenAI Assessment and Language Equity: Drawing the Line between Support and Substitution. Journal of Academic Ethics. https://doi.org/10.1007/s10805-026-09760-1