Scientific American covers cognitive scientist Selmer Bringsjord’s efforts to program a thoroughly evil artificial intelligence. As presented in the article, Bringsjord’s working definition of evil seems pretty confused.
To be truly evil, someone must have sought to do harm by planning to commit some morally wrong action with no prompting from others (whether this person successfully executes his or her plan is beside the point). The evil person must have tried to carry out this plan with the hope of “causing considerable harm to others,” Bringsjord says. Finally, “and most importantly,” he adds, if this evil person were willing to analyze his or her reasons for wanting to commit this morally wrong action, these reasons would either prove to be incoherent, or they would reveal that the evil person knew he or she was doing something wrong and regarded the harm caused as a good thing.
Parts of that paragraph read as describing a sadist, a psychopath, or someone who is badly confused. None of these things seem like a good stand-in for evil. But, then, evil is a notoriously difficult idea to define.
I wonder if this general approach– skip the rigorous definition, instead try to recreate the behavior– might appeal to experimental philosophers. Is there anything to be gained from trying to model confusing psychological phenomena like weakness of the will or self-deception? If we could program a computer to behave as if it were deceiving itself, could that possibly give us any insight into what’s going on when we deceive ourselves?
Neuroethics: Ethics and the Sciences of the Mind
By Neil Levy, University of Melbourne (December 2008)