Rethinking Human Rights Protection in the Age of AI: From Ex-Post Remedies to Rights-by-Design

Abstract

As artificial intelligence rapidly becomes a mediating but often invisible factor in decision-making across many areas that are central to the realisation of human rights, ranging from social welfare to the judiciary, legal protection still predominantly operates through an ex-post logic, reacting only after harm has occurred. This article identifies the functional defects of that model in contemporary algorithmic systems, where decision-making opacity, automation, and fragmented responsibility complicate the identification of the responsible actor, proof of causation, and the effective contestation of outcomes.