The rapid changes in the fields of artificial intelligence (AI) and big data analytics have transformed an algorithmic decision-making system (ADMS) with considerable uses in the government sector and the medical field. Nevertheless, there is growing interest in the literature on ADMS that is still scattered across disciplines and offers conflicting findings regarding its effectiveness. The socio-technical relationship of ADMS, the ethical issues, and organizational impact are complex, yet they are not comprehensively studied, even though the use of ADMS grows. The study will provide a thorough review of the literature that will bring together interdisciplinary viewpoints on the creation, adoption, and impacts of ADMS in these essential sectors. The paper integrates both empirical and theoretical sources to explain such key considerations that affect algorithmic judgments, including accuracy, fairness, transparency, human oversight, and governance systems, based on a comprehensive search and rigorous selection process. The findings provide inconsistency in efficiency gains, transfer of power, and dehumanization-anthropomorphism dialectic in the organizational setting, which presents significant contradictions between technological efficacy and socio-ethical issues. The paper focuses on the need to have balanced approaches to ensure that human agency is safeguarded as algorithmic abilities are used, extending theoretical concepts of fairness beyond measures of accuracy to consider equity and legitimacy. This synthesis is informative of future research activities focused on conducting empirical validation, assessing harm, adaptive governance, and inclusive design. Ultimately, the work contributes to the understanding of the interdisciplinary character of ADMS and gives essential guidance towards moral, accountable, and just algorithmic decision-making in high-stakes areas.
Almezoghy et al. (Sun,) studied this question.