Between Efficiency and Empathy: An Exploratory Study of Albanian Students' Attitudes Towards AI in Ethical Decision-Making

This article presents an exploratory study of how university students in Albania (N = 146) perceive the role of artificial intelligence (AI) in ethical decision-making. While AI is increasingly integrated into everyday choices, little is known about how individuals in non-Western, digitally emerging contexts negotiate trust, empathy, and moral responsibility in relation to algorithmic systems. Drawing on survey data, we find that respondents maintain a strong attachment to empathy as a moral reference point (M = 3.12) and express significant reservations about delegating emotionally consequential decisions to AI (M = 3.41 for routine vs. emotional delegation). Trust in AI's ethical competence is moderate and conditional, strongly associated with perceived safety (r =.54) and transparency. We interpret these findings through the lens of what we term the "residual moral subject"—a concept describing the persistence of human moral agency in technologically mediated environments, albeit in a reconfigured form. The study contributes empirical nuance to debates on algorithmic governance and highlights the need for culturally situated research on AI ethics.