Add a matchup slip to one task. Students submit Output A, Output B, and a 150-word justification for their choice using class evidence. Four consistent questions:
A student who has genuinely read both outputs can answer all four questions with specificity. A student who generated both outputs and accepted them without reading cannot. The matchup slip makes that difference visible without any detection tool required.
The 150-word limit matters. Long enough to require genuine reasoning. Short enough that the task stays focused on comparison rather than becoming another essay.
Ask for two explanations of the same concept for different audiences: for a Year 2 learner and for a Year 8 learner. Compare vocabulary choices and what got left out. The learning is the comparison, not the explanation.
Generate two different introductions to the same essay question. Annotate the stronger one for how it frames the argument. Explain why it is stronger using subject-specific language.
Does this routine strengthen student judgement, or am I accidentally training students to generate more text without thinking?
A comparison-based norm is clear and teachable: if you use AI, you must compare it to class resources and show the differences. This expectation holds even when students switch tools mid-task.
Print or save this resource as a PDF using your browser.