The Indefensibility of Algorithmic Bias: A Human Dignity Perspective on Artificial Intelligence

Abstract

AI-based (social) identification technology, which relies on machine learning performance supported by algorithmic complexity to achieve desired outcomes, has permeated humans’ daily lives. On the one hand, this technology offers significant benefits, particularly in enhancing operational efficiency and facilitating the analysis of large volumes of data. On the other hand, however, it presents a severe ethical challenge, as AI technology operates through algorithms that have not been free from bias. This study critically examined the ethical issues of discrimination within the discourse on human dignity, particularly in the context of biased algorithmic applications from a (Catholic) moral theologian’s perspective. This article is intended to reveal that (a) algorithmic bias has the potential to disproportionately affect vulnerable and minority groups, and (b) discrimination resulting from algorithmic bias is exceedingly difficult to eliminate, thereby necessitating a discourse grounded in the infinite dignity of every human being.