Iain Martin
The very rapid evolution of AI, especially the use and adoption of transformer based large language models is touching every aspect of education and assessment.
ChatGPT very publicly highlighted their spread with the platform reaching 100 million users in less than 2 months. The last months of 2022 and 2023 saw what can only be described as panic in many educational settings as to their impact on education, most notably assessment.
From the perspective of a university educational environment this talk will explore how we consider AI and in particular LLM’s on how we frame, design and deliver meaningful, valid and accurate assessment in the health professional education context, when these tools are ubiquitous.
The moral panic that has been evident given that these LLM based tools can ‘pass’ professional exams in medicine and law, should perhaps best be reframed as why is there not moral panic that we are relying on assessments that the currently available LLM’s can ‘pass’.
This talk will deliberately aim to be provocative and challenging in exploring the interface of assessment design, ethics, wisdom and technology as we embrace what may be possible, what should be done and how do we know what our students know.
