Academics and AI marking: a welcome aid for overworked staff or a sloppy shortcut?
Educational institutions find themselves suddenly confronted with a fork in the road Get our breaking news email , free app or daily news podcast Armin Alimardani has blue hair, is fresh faced, wears a smart watch and describes himself as “a tech enthusiast”. He is an unlikely sentinel, perhaps, to warn that a drift towards academics using generative artificial intelligence to help mark students…
Western Sydney University senior lecturer Armin Alimardani warns that the use of generative artificial intelligence (AI) to mark students' assignments could pose an existential threat to higher education. Alimardani, who has been studying and experimenting with AI for years, finds the idea of using AI to lighten the burden of grading "extremely concerning".
He emphasizes the need to be "really careful" where they are stepping, as there is a risk that higher education could collapse if professors replace human judgment with algorithms.
Several Australian universities, including Western Sydney University, the University of Newcastle, Deakin University, RMIT, and the University of Adelaide, have allowed staff to use AI in marking assessments and providing students with feedback. However, these institutions maintain a qualified green light for AI, with students having the option to opt out or with staff using AI to improve efficiency and support assessment and administrative activities, but not to assign grades.
Alimardani argues that AI could potentially improve the quality of marking by functioning as a "second pair of eyes" to detect inconsistencies in human assessment. He is working on creating models that can pick out such inconsistencies. However, he notes that the potential benefits of AI in marking have not been thoroughly tested, while the risks it poses are significant enough to warrant a "really slow" and considered evaluation of those claims.
One potential issue, Alimardani points out, is "verification drift": if 100 academics are given an AI marking tool but told to verify its assessments, some may not do so. He calls this "verification drift," and warns that after a few AI-generated assessments appear fine, humans overseeing the verification may become less vigilant, allowing AI hallucinations to slip through as facts.
Such scandals, as have been seen in courtrooms, would be almost inevitable, tarnishing university reputations and creating a sense of unfairness among students that could lead them to question the value of higher education.
Written by urgent.news from The Guardian Australia's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
