From principles to process: the principlist approach to AI ethics and lessons from Internet bills of rights

Abstract This article examines possible lessons the principlist approach to AI ethics could learn from the relatively developed initiatives for Internet bills of rights. From intergovernmental organisations, governments to technology companies and civil society groups, a series of guidelines that lay out principles for the ethical design, development and use of AI technologies have emerged. Despite the prominence of this ‘principlist’ approach to AI ethics in the literature, its interplay with similar approaches to technology governance has not received any attention. This article seeks to fill this void by exploring possible lessons that the more nascent principlist approach to AI ethics could learn from Internet bills of rights. It argues that the future of the principlist approach hinges on averting factors that undercut the potential of the Internet bill of rights project while at the same time drawing upon its progressive elements. The article thus explores imperatives of grounding the principlist approach in a sustained normative and institutional process that enjoys a level of legitimacy and normative novelty. That would shift the focus in AI ethics discourse from articulating principles to setting in motion a process that would clarify, sustain and legitimise them.