Algoritma dan Diskriminasi

Siapa yang bertanggung jawab atas pelanggaran HAM oleh AI

Abstract

The development of artificial intelligence (AI) has brought significant benefits to various sectors of life. However, despite the efficiency offered, AI algorithms also have the potential to violate human rights through discriminatory decisions against certain groups based on race, gender, religion, or socioeconomic status. This article addresses the central question: who is legally and ethically responsible for human rights violations caused by algorithms? Using a normative approach and analyzing case studies of algorithmic discrimination (such as recruitment systems, credit scoring, and facial recognition), this study identifies five parties with varying levels of responsibility: algorithm developers, AI service providers, governments/regulators, end users, and international standard-setters. The analysis shows that responsibility cannot be assigned solely to non-human entities (AI), but rather is multifaceted and layered (shared and layered accountability). This article concludes that a comprehensive regulatory framework, redress mechanisms for victims, and the principle of meaningful human control are needed to ensure accountability for any human rights violations resulting from algorithmic systems.