Wendell Wallach and Colin Allen, authors of Moral Machines: Teaching Robots Right From Wrong, a fascinating look at morality in machines, have a blog on the topic that I wasn’t aware of until recently.

You should definitely check out their book. It’s relatively brief and analyzes many important issues around how one might go about building machines with a sense of morality. Wendell made a post where he praised my recent project Preventing Skynet and made a call for closer interaction between two communities in machine morality:

Our friend Michael Anissimov has, together with others, initiated a new blog, “Terminator Salvation: Preventing Skynet: Just say ‘no’ to genocidal artificial intelligence!” We applaud this effort and encourage members of the machine morality, machine ethics, and roboethics community to contribute to the blog. There has been a kind of split into two communities, with only a little cross-over, between those focused around future ethical challenges posed by a possible Singularity and those whose attention is directed at more immediately challenges and the implementation of moral decision making in present or near-future technology. I’d like to propose that we make efforts to bridge this gap, and will have more to say about that in a future posting.

I do agree, though the two groups may have a few inherent differences. Upon reading Wallach’s book, however, I think many of these differences may be superficial in some instances. Our group is particularly concerned with the question, “how do we create a self-improving AI that we can trust not to kill us as it becomes more powerful than the entire human race?” The machine morality, machine ethics, and roboethics communities are interested in the more general question, “how do we build machines that we can trust to make moral decisions?” Many of the ideas from the latter camp could be useful for the goals of former, and vice versa.

The Singularity Institute (SIAI) is taking steps to integrate itself more seriously with the mainstream roboethics community. In just a week, SIAI employees Anna Salamon and Steve Rayhawk and SIAI associate Carl Shulman will be presenting papers at the European Conference at Computing and Philosophy. One of the papers is based on the project we did last summer as part of the SIAI internship program, Uncertain Future. I am currently visiting the current SIAI summer program and recently got to see preliminary presentations of both papers. Thanks to the support of SIAI’s generous donors, I’m sure we’ll be seeing more SIAI representatives at academic conferences on computing, philosophy, and roboethics.