On 10/29/2015 9:57 PM, Jason Resch wrote:
On Thu, Oct 29, 2015 at 10:37 PM, Brent Meeker <[email protected] <mailto:[email protected]>> wrote:On 10/29/2015 6:44 PM, Jason Resch wrote:On Wed, Oct 28, 2015 at 11:18 PM, Russell Standish <[email protected] <mailto:[email protected]>> wrote: On Wed, Oct 28, 2015 at 09:11:34PM -0500, Jason Resch wrote: > At some level, an algorithm cannot be held responsible for its actions > because it was doing the only thing it could do, what it was programmed to > do. At some point between a simplistic algorithm and a human level AI, > however, we seem able to assigning responsibility/culpability. What does an > algorithm minimally have to have before it reaches this point? > > The ability to learn? > Understanding of the consequences of its actions? > Rights that it cares about? > Personhood? > None of those things are required to assign legal responsibility. For example, a company can be held legally responsible, but a company is not a person, nor is it conscious, nor need it learn (although companies can learn). I think all that is required is a sense of agency, that holding something responsible is sufficient to affect the actions of that agent. If a robot can process the notion of responsibility such that its actions will be affected by it, then yes it can be held responsible regardless of whether any conscious understanding exists. What are the minimum requirements to program agency? It seems to be, if a program cannot learn/alter itself, it cannot be held responsible, for it is doing only what it was programmed to do.Suppose the program is designed to learn from experience. Then when it learns from experience, does that mean it's altering itself or just doing what it was programmed to do?Both, but I would argue a program that cannot change itself cannot be held responsible for its actions. If, every time it is given the same input, it gives the same output (e.g. if it is memoryless) then punishment can have no effect, it is what it is.
I agree, except that judicial punishment is meant for the rest of society to, not just the miscreant.
I just watched episode 3 of "The Brain with Dave Eagleman". He tells of a Canadian man who fell asleep in front of his TV one night, then got up and drove the home of his in-laws, enter their house and murdered his mothe in-law and strangled (but did not kill) his father in-law, and then drove to police station and said, "I think I've killed someone." He didn't remember any of it. He did it all sleepwalking. The jury acquitted him after being presented evidence that most members of his family had sleep disorders and had done things while sleepwalking.
They were convince that he wasn't responsible. But I think they made a mistake. The conscious part of him might not be responsible, but his subconscious was. And if his subconscious was aware enough of the world to develop an animus toward his in-laws then it could also be influenced by judicial punishment.
Brent -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
