Dr. Sam Birks, a medical educator at the University of Sheffield and an emergency doctor, presented a research-based project where medical students utilized Large Language Models (LLMs) to generate their own exam and practice questions. The core of the workshop centered on the "double-checking" phase, where the true learning occurred as students critically verified the accuracy of AI-generated content. This process transformed assessment into a collaborative opportunity for peer learning and deep engagement with the material.
Dr. Birks shared the specific prompts used in the project and reflected on the lessons learned regarding the responsible use of AI in specialized fields like medical education. The session concluded by highlighting how "flipped assessment" not only improves authentic learning but also equips students with the critical thinking skills necessary to navigate an AI-augmented professional landscape.