I was recently approached in my capacity as module lead and asked if I’d be willing to trial software known as GrAIde as part of a pilot scheme using AI to mark formative assessments. As a trade unionist and a person who quite likes having a job and an income, I politely declined.

GrAIde bills itself as an AI powered marking and assessment feedback platform which learns from academics and provides feedback to students. According to the website, if the AI has high enough confidence in its predicted grade, it was fast track this, if confidence is low, the tutor will review it. It’s billed as a productivity tool and the marketing of the product seems focussed towards management as opposed to academics – which I think is significant in the context of the current state of the sector, where corporate phrases like “maximising productivity” and “reducing costs” have become more and more normalised. The benefits of the system focus on the micro-management aspects of the system, which would very likely undermine the marking and assessment tariffs across the University.

In my view, marking, alongside teaching and scholarly research, is one of the cornerstones of being an academic. It is a core part of our job descriptions, requiring expert subject knowledge as well as advanced knowledge of pedagogy and constructive alignment, the ability to formulate learning outcomes and assessments.

To relinquish this to a machine is to usher in the very thing that will make us redundant. I articulated my thoughts to my colleagues who had also been approached to trial this software and we fed this back to the person who had approached us.

To their credit, this person accepted our views and, on our behalf, relayed that the software wasn’t suitable for our needs, but I do worry that in the future this platform might be imposed by more senior echelons of the University. I also worry that overworked, burnt out staff, might welcome this without really considering the long-term ramifications. In the short-term, more free time sounds appealing, and if pitched in the right way, staff might focus on the so-called benefits.

I also worry about the impact this will have on our reputation and credibility: if I was a student, I would certainly feel short-changed if I knew my hard work was being marked by AI. From a recruitment perspective it’s probable that students would choose an HEI where their work is marked by a human, not a robot.

I know that AI is here to stay, but senior leadership should be considering productivity tools that help with the repetitive, time-consuming administrative tasks that burden academics: the tasks that don’t require specialist knowledge yet take up an increasingly large proportion of our workload. The focus should not be on streamlining marking and teaching to free us up for even more administration. If we welcome in AI to mark on our behalf, then we welcome in the very technology that will replace us in the classroom.

Stephen Marks (UCU Rep)