What to Do about Artificial Consciousness

Chapter Summary

One of the primary ethical issues related to creating robust artificial intelligences is how to engineer them to ensure that they will behave morally — i.e. that they will consider and treat us appropriately. A much less commonly discussed issue is what their moral status will be — i.e. how we ought to consider and treat them. In this chapter, John Basl takes up the issue of the moral status of artificial consciousnesses. He defends a capacity-based account of moral status, on which an entity’s moral status is determined by the capacities it has, rather than its origins or material composition. An implication of this is that if a machine intelligence has cognitive and psychological capacities like ours, then it would have comparable moral status to us. However, Basl argues that it is highly unlikely that machines will have capacities (and so interests) like ours, and that in fact it will be very difficult to know whether they are conscious and, if they are, what capacities and interests they have.

See also Patrick Lin et al. (Ch. 23) and Wendell Wallach (Ch. 24) in this textbook.