Beyond persons and things: the legal status of artificial intelligence in the EU

Mocanu, Diana [UCL]

Technological progress seems to not only herald the adoption of appropriate regulatory measures in the case of AI, but also a serious reconsideration of the philosophical underpinnings of legal conceptions about personhood and thing-hood. These uncanny entities, which increasingly straddle the border between human and artifact, will likely be involved in hard cases before European courts, requiring the latter to answer some difficult questions. What those answers would ultimately amount to is an outline of the legal status of AI systems, determining therefore what legislation is applicable and enforceable in relation to the practical uses and the ensuing consequences of those uses of AI. There are three possible answers to this. The law can treat AIs as things, it could consider them legal persons or it could create a new, sui generis legal status, specific to these entities. The proposed research therefore contends that to systematize the arguments in favor and against each of these courses of action in turn and to contribute new ones is what is needed in order to reach the ultimate goal of outlining a ‘portrait-robot’ of what the legal status of AI could look like in the EU.