In both law and ethics, we typically distinguish between persons, who are subjected to and the subjects of the law, from things, which are nothing more than objects or property owned by persons. This way of dividing up and categorizing all that exists has been useful and expedient, even if we have, at various times in our history, mistakenly and unfortunately categorized as things that which later came to be seen as other persons. The question that now confronts us in the face of artificial intelligence is whether this new kind of being is to be classified as a thing or instrument that we can use and even abuse as we see fit, whether it should be classified as another person or another kind of person with its own set of rights and responsibilities, or whether it might be something other that is both/and or neither/nor. This chapter will present and evaluate the costs and benefits of all three options and conclude by proposing a moral and legal ontology suitable for responding to and taking responsibility for artificial intelligence.