Designing assessment tasks that are less vulnerable to AI

The growing accessibility and capability of generative AI platforms that can rapidly produce text, images, or other content in response to simple user prompts (e.g., ChatGPT, DALL-E, etc) pose a new and significant threat to assessment integrity. As such, many academics are wondering how they can redesign assessment to reduce academic misconduct. In response to these concerns, this page features guidance, practical strategies, case studies, and resources about good assessment practices that can enhance student learning while also reducing the risk of academic misconduct.

Seven practical strategies for improving assessment design and integrity

While there are various strategies to improve assessment design in your subject, we focus on seven key strategies below. While we provide examples of how to implement these strategies, how suitable a particular strategy is will depend on your teaching context, including the nature of your subject, year level of students and class size, among other considerations. We also acknowledge that redesigning assessments is not without its own challenges, especially in relation to workload and resourcing.

Case Studies