March 14, 2012
How to make ethical robots
From PhysOrg:
In the future according to robotics researchers, robots will likely fight our wars, care for our elderly, babysit our children, and serve and entertain us in a wide variety of situations. But as robotic development continues to grow, one subfield of robotics research is lagging behind other areas: roboethics, or ensuring that robot behavior adheres to certain moral standards. In a new paper that provides a broad overview of ethical behavior in robots, researchers emphasize the importance of being proactive rather than reactive in this area.
The big question, according to the researchers, is how we can ensure that future robotic technology preserves our humanity and our societies’ values. They explain that, while there is no simple answer, a few techniques could be useful for enforcing ethical behavior in robots. One method involves an “ethical governor,” a name inspired by the mechanical governor for the steam engine, which ensured that the powerful engines behaved safely and within predefined bounds of performance. Similarly, an ethical governor would ensure that robot behavior would stay within predefined ethical bounds. For example, for autonomous military robots, these bounds would include principles derived from the Geneva Conventions and other rules of engagement that humans use. Civilian robots would have different sets of bounds specific to their purposes.
More here.
Posted by Azra Raza at 08:47 AM | Permalink
Comments
Emotions, those directly related to morality and otherwise, do not and cannot apply to robots and talk of "ethical behavior" with regard to robots is utterly nonsensical and a categorial mistake of the egregious sort. At the very least, this reveals an inability to understand how we (i.e., human beings) come to behave morally and how ethical principles are dependent on judgment, the capacity for which is distinctively human. A sentient android or robot is the stuff of science fiction and psychological predicates and ethical grammar that apply to human beings do not apply to robots.
Posted by: Patrick S. O'Donnell | Mar 14, 2012 9:42:44 AM
Indeed, it's not just the role of emotions in ethics but their role in *cognition* that needs to be taken into account in any possible successful AI project which aims to create anything like human intelligence. (Hence scifi tropes like the android Commander Data's optional emotion chip in "Star Trek: TNG" are fundamentally misconceived, as is the idea that Vulcans are superior to others in their cognition because they suppress their emotions.) As always, see Damasio's "Descartes' Error" and his subsequent books on the subject.
One of the deficiencies in both deontological and consquentialist ethics is their presumption that an ethical code can be summed up in an algorithmic formulation, with no explicit role for the normative function of our emotional reactions. Indeed, the use of the phrase "ethical *code*" reveals this presumption. (At least the Aristotelean-derived field of Virtue Theory, and Virtue Ethics, from Anscombe onward, tries to address the role of emotions in ethics.)
It may be possible, of course, to create an emotionless intelligence, *if* one could solve the motivator problem for logically ambiguous situations, but it would be truly alien to us.
Posted by: Kai Matthews | Mar 14, 2012 11:41:58 AM
I'd also add that, as much as I admire Isaac Asimov, his Three Laws of Robotics formula commits the same algorithmic fallacy.
Posted by: Kai Matthews | Mar 14, 2012 11:45:51 AM
@Patrick: emotions "... do not and cannot apply to robots" ... and neither does the notion of judgement. Quite a categorical statement.
It's not that I expect there to be robots that people could justifiably ascribe emotions or judgement to in my lifetime, but for the life of me I can't figure out what kind of metaphysical principle could rule it out. Care to enlighten us?
Posted by: Matt_M | Mar 15, 2012 5:25:00 AM
Post a comment