A Brief Overview of Machine Ethics

Ear­lier, I lamented that even though Eliezer named schol­ar­ship as one of the Twelve Virtues of Ra­tion­al­ity, there is sur­pris­ingly lit­tle in­ter­est in (or cit­ing of) the aca­demic liter­a­ture on some of Less Wrong’s cen­tral dis­cus­sion top­ics.

Pre­vi­ously, I pro­vided an overview of for­mal episte­mol­ogy, that field of philos­o­phy that deals with (1) math­e­mat­i­cally for­mal­iz­ing con­cepts re­lated to in­duc­tion, be­lief, choice, and ac­tion, and (2) ar­gu­ing about the foun­da­tions of prob­a­bil­ity, statis­tics, game the­ory, de­ci­sion the­ory, and al­gorith­mic learn­ing the­ory.

Now, I’ve writ­ten Ma­chine Ethics is the Fu­ture, an in­tro­duc­tion to ma­chine ethics, the aca­demic field that stud­ies the prob­lem of how to de­sign ar­tifi­cial moral agents that act eth­i­cally (along with a few re­lated prob­lems). There, you will find PDFs of a dozen pa­pers on the sub­ject.