The emergence of the human right to safe artificial intelligence

Abstract

Several international agreements on artificial intelligence were adopted last year, first and foremost the UN General Assembly resolution on safe, secure, and trustworthy AI systems for sustainable development (A/78/L.49). There has also been a growing focus on the connections between new technologies and the legal sphere, evidenced by many publications that approach comprehensively the issue of AI governance and establishing a legal framework for AI applications. We may assume that specific rights related to artificial intelligence are emerging alongside digital rights, such as trustworthy AI, secure AI, reliable AI, and safe AI. This article presents the author’s perspective on the right to safe AI, guaranteeing that no harm will be caused and no undesirable consequences will occur due to such use. Minimizing risks and ensuring that humans maintain control over technology lie at the root of this right. The author explores the various human needs connected to AI systems and upon which AI-related human rights depend, including the need for control. Furthermore, he investigates how the right to safe AI may differ from existing human rights established in international law (rights to information and labor). The article contains an attempt to search for the meaning of the right to safe AI, primarily through examining some AI ethical guidelines. The proposed new right aims to ensure that the risks associated with developing and using AI systems are expected to be kept to an acceptable minimum. The findings of this study could be valuable for further exploration of human rights issues stemming from the rise of new technologies. The article briefly defines the human right to safe AI as “the right to safe use of AI systems” and stresses that developing AI systems should not threaten humanity.