A colleague recently shared a polite email from a student appealing their assessment grades. Every rubric criterion was defended and addressed in tremendous detail.
It felt optimised, and in an age of generative AI, maybe that’s exactly what it was.
We’re entering a new phase where students use AI not just to prepare assessments but to craft appeals, generating arguments perfectly shaped to align with criteria and maximise persuasive force.
To understand this development, we must first examine the role of rubrics in contemporary education. Assessment rubrics function as what Michel Foucault might recognise as disciplinary technologies, tools that standardise judgment and render subjective evaluation processes transparent and measurable. They represent institutional attempts to rationalise assessment, making explicit the criteria by which student work is evaluated, theoretically democratising access to success criteria.
Constructing appeals with unprecedented precision
As Pierre Bourdieu reminds us, institutions often reward not just knowledge but the ability to navigate codes and expectations. When rubrics and standardised criteria are coupled with AI-augmented optimisation, however, we risk shifting learning’s centre from transformative engagement to compliance engineering, undermining the outcomes we are attempting to measure.
Students with access to sophisticated AI tools can now systematically analyse rubric language, identify optimisation opportunities, and construct appeals with unprecedented precision. This development represents what Jürgen Habermas would likely classify as the colonisation of educational lifeworlds by instrumental rationality, the reduction of learning processes to technical problems requiring algorithmic solutions.
When academic feedback like “this section lacks depth” gets treated as a technical problem to solve, however, rather than expert judgment to engage with, we transform educational dialogue. The more “optimised” the process, the less space for generosity, nuance, or authentic learning’s messy back-and-forth.
Jacques Rancière’s work on pedagogy suggests that educational relationships depend on the assumption of human mutuality, a recognition that both student and teacher are capable of thought and interpretation. AI-mediated appeals disrupt this dynamic. When students rely on AI to process feedback, the algorithm does not engage with feedback as a thinking subject but processes it as information to be optimised against.
Recognition and validation
I expect that students using AI for appeals genuinely care. They want recognition and validation, in addition to graduating with their degree. Max Weber’s analysis of rationalisation processes, however, helps us understand how well-intentioned actions can contribute to broader structural changes that undermine their original purposes. Weber observed how the rationalisation of social life (the systematic organisation of action according to calculated rules) tends to displace value-rational action (action oriented toward ultimate values) with instrumental rationality (action oriented toward efficiency). When students optimise appeals against rubric criteria, they engage in precisely this type of instrumental calculation, even when their underlying motivations remain value-oriented.
Academic assessment involves what Aristotle called phronesis: practical wisdom that cannot be reduced to rule-following. When educators evaluate student work, they exercise judgment that draws on disciplinary expertise, pedagogical experience, and contextual understanding. This judgment necessarily involves interpretation and cannot be fully systematised. AI-optimised appeals attempt to bypass this judgmental dimension by reducing assessment to rule application. This reduction represents what Herbert Marcuse might recognise as one-dimensional thinking, the flattening of complex educational relationships into technical procedures.
The proliferation of AI-mediated appeals has broader implications for educational institutions. If Anthony Giddens is correct that modern institutions depend on trust relationships between expert systems and lay participants, then the mechanisation of student appeals may erode the trust relationships that sustain educational institutions.
Educators as algorithmic systems
When students systematically optimise against assessment criteria rather than engaging with feedback as developmental guidance, they effectively treat educators as algorithmic systems rather than professional practitioners. This shift may prompt educators to become more defensive in their assessment practices, potentially reducing the pedagogical risk-taking that often produces meaningful learning experiences.
If appeals processes become dominated by AI optimisation, institutions may respond by developing counter-measures: AI systems to evaluate AI-generated appeals. In Jean Baudrillard’s terms, a simulation replaces real interaction with its mechanised imitation.
This broader context helps explain why AI-optimised appeals feel unsettling even when students’ motivations appear legitimate. The optimisation process treats educational relationships as data to be manipulated rather than human connections involving care, judgment, and mutual recognition.
The messy middle
We live in the messy middle where human and machine shape one another. It is a zone of entanglement where our judgements, our values and our decisions are increasingly mediated, supported or even challenged by machine outputs. Machines, however, do not care. Education’s meaning is formed in relational and ethical spaces. We must protect them.
Jonathan Boymal is an associate professor of economics in the School of Economics, Finance and Marketing, RMIT University’s College of Business and Law. He has 25 years of higher education leadership experience at the undergraduate and postgraduate levels across Melbourne, Hong Kong, Singapore and Vietnam, in roles including Associate Deputy Vice-Chancellor, Learning Teaching and Quality and Academic Director, Quality and Learning and Teaching Futures. Jonathan holds a PhD in Economics.
This article was originally published on EduResearch Matters. Read the original article.
Got something on your mind? Go on then, engage. Submit your opinion piece, letter to the editor, or Quick Word now.