<p>This article argues for a reconsideration of Artificial Intelligence (AI) literacy in higher education in the light of the spread of Generative AI (GenAI) and Large Language Models (LLMs). Explainable AI is reread in terms of explicability. It is understood as an educational competence that integrates intelligibility and accountability, rendering the criteria, limits, and responsibilities underlying algorithmically mediated processes open to discussion. Explicability thus becomes a form of postdigital AI literacy grounded in situated and distributed practices of reading, writing, and decision-making within more-than-human contexts. Such literacy is understood as the capacity to navigate and influence automated systems while maintaining critique and decision-making autonomy. This perspective is linked to the regulatory framework of the European AI Act, showing how transparency and AI literacy constitute not only technical requirements but pedagogical conditions for sustaining agency, epistemic autonomy, and human-centred design in higher education.</p>