Legal Personhood for AI?

Abstract

This chapter considers legal personhood for artificial agents. It engages with the legal issues of autonomous systems, asking the question whether (and if so, under what conditions) such systems should be given the status of a legal subject, capable of acting in law and/or being held liable in law. The main reason for considering this option is the rise of semi-autonomous systems that display unpredictable behaviour, causing harm not foreseeable by those who developed, sold, or deployed them. Under current law it might be difficult to establish liability for such harm. To investigate these issues, the chapter explains the concepts of legal subjectivity and legal agency, before inquiring into the nature of artificial agency. Finally, the chapter assesses whether attributing legal personhood to artificial agents would solve the problem of private law liability for harm caused by semi-autonomous systems.