Artificial intelligence value alignment principles the state of art review from information systems research

Abstract

Information and communication technologies (ICTs) must be designed and used for humane ends. The rapid adoption of Artificial Intelligence (AI) has raised the critical question of whether we can ensure AI's alignment with human values to guide its design and use. We perform a selective literature review with the specific search terms of the papers published in the top information systems (basket of 8 journals and 5 AI journals in IS) from 2000-2020 to answer this question. The findings indicate that IS research has contributed insufficiently to a deeper understanding of human values and AI value alignment principles. Moreover, the mainstream IS research on AI is mostly dominated from its technical and managerial aspects. Thus, the future research agendas are proposed accordingly. The paper provides some food for thoughts in studying human values and AI alignment within the context of IS research.