A pragmatic protocol for stress testing assessment formats against GenAI

This practical protocol provides educators with a structured approach to evaluating the resilience of current assessment formats against the capabilities of generative AI tools such as ChatGPT. It outlines a step-by-step process for planning, conducting, analyzing, and reporting AI stress tests, helping educators understand how GenAI may perform on their existing assessments. The protocol aims to support informed reflection, foster secure assessment design, and guide future reform by integrating or safeguarding against GenAI use. Developed from the authors’ extensive experience and literature, this protocol is designed to be adaptable and replicable, offering a valuable resource in the rapidly evolving landscape of AI in education.