Was this forwarded to you by a friend? Sign up, and get your own copy of the news that matters sent to your inbox every week. Sign up for the On EdTech newsletter. Interested in additional analysis? Try with our 30-day free trial and Upgrade to the On EdTech+ newsletter.
It’s the most wonderful time of the year! By which I mean the latest season of The Great British Bake Off has finally landed here in the US. For the next couple of months, I’ll be writing these posts with my head full of tarts, Swiss meringue, and brioche. But while my sweet tooth is otherwise occupied, what did I read this week to feed the other part of my brain?
It’s complicated
Back at my old employer, I spent several years trying to dodge covering assessment technology, largely because, even before the arrival of generative AI, it was already clear that this was a highly complex set of tools applied in deeply context-specific ways.
Generative AI has only amplified that complexity. Since its emergence, the sector has been wrestling with difficult questions: How do we prevent and detect cheating? How do we design valid, authentic assessments that are less vulnerable to AI shortcuts? And can we leverage AI positively, for example, to scale assessment design or deliver faster, richer feedback?
In a new article, Thomas Corbin, Margaret Bearman, David Boud, and Philip Dawson argue that seeking neat technological or policy “solutions” to the challenge of AI in assessment is the wrong approach.
Drawing on in-depth interviews with twenty instructional faculty in Australia, Corbin et al. contend that assessment in the age of generative AI fits what Rittel and Weber famously described as a wicked problem.
Wicked problems are defined by ten characteristics. I won’t unpack them all here, but they all map clearly onto assessment challenges. For example.
Definition
Wicked problems typically cannot “be clearly or conclusively defined.”
Unlike technical problems where stakeholders can in theory agree on what needs fixing, wicked problems mean different things to different people and these varying definitions pull solutions in contradictory directions. Without agreement on what the problem is, a singular, cohesive response becomes impossible.
In assessment, there are deep divisions over how generative AI should be defined or understood in education. Some argue it should be embraced as a professional tool, one that students must learn to use effectively in order to thrive in the workforce. Others see it primarily as a shortcut, or worse, a vehicle for cheating that undermines genuine learning.
