Beyond moral panic: the algorithmic self at the intersection of UDL, instructional automation, and agency in the post-digital era

Abstract

This article reframes educational technology in the post-digital era as a regime of decision-making rather than a set of tools. In a condition of infrastructural saturation, datafication and assessment platforms reshape the rules of recognition, producing an “algorithmic self” whose trajectories become legible or opaque according to what counts as evidence. Against the polarisation between techno-solutionism and moral panic, inclusion is treated as a procedural requirement rather than a rhetorical claim. Universal Design for Learning (UDL) is presented as both a pedagogical response and a governance framework for instructional automation, because it makes explicit, ex ante, the links between objectives, criteria, and commensurable forms of evidence, and it renders decisions transparent and contestable. Two design lenses, affordances and cognitive load, are used to show how digital environments can reduce or amplify cognitive barriers, and how standardised traces can generate epistemic injustice by devaluing legitimate forms of expression. A three-mode taxonomy of automation—assistive, augmentative, and delegative—is then outlined, with UDL operating respectively as enabler, coherence constraint, and veto criterion, thereby protecting agency and the curriculum from metric-driven “shadow curricula”. The analysis is finally connected to the European regulatory context (GDPR, Accessibility Act, AI Act), where transparency, human oversight, explainability, and contestability increasingly function as institutional obligations. Implications are discussed for assessment literacy and for governing automation by design.