Comparative judgement of co-intelligence products

In one of our courses, we use an innovative assignment to develop students' basic AI literacythrough a creative task: designing a reintroduction marketing plan for a reintroduction of adiscontinued brand (band revival), to be presented in a PowerPoint report. Students are asked touse AI throughout the assignment, provided they do so responsibly and in line with courseguidelines. For example, they must make role plays (interviews) with self-defined persona. A keychallenge is that traditional assignment assessment criteria do not adequately capture theintended learning outcomes. In particular, they fall short in evaluating creativity and students'critical reflection on their use of AI. To address this, we explore comparative judgement (CJ) as analternative assessment approach. Instead of relying on predefined rubrics, threeexaminers compare student products and decide which differs most. By explaining why oneproduct is different from others, examiners externalize more valid and nuanced criteria,especially regarding creativity and critical reflection with AI. This approach also offers greaterflexibility in justifying judgments. While CJ is often technically complex, we simplify the process tomake it more accessible for classroom use. On base of the CJ a new rubric is designed which is"AI integrated".
Feedback questions:
1. How does comparative judgement support the assessment of creativity in AI-integratedassignments?
2. How does comparative judgement help capture students' responsible and critical use of AI?