Should robots be able to disregard the will of their masters?
I am watching the "Beatless" anime currently, and it raises a lot of interesting questions regarding human-robot interaction. One of them concerns the robot's ability to disobey its masters, or even apply violence to its masters, in order to achieve a favorable outcome.
To see how this question may arise, consider the famous mental exercise. You buy a robot and order it to guard your life at any cost. A few days later, you are in a horrible mood and decide to commit suicide. You take a gun and put it to your head...
Should the robot do nothing and let you shoot yourself, which would violate the order you gave it earlier? Should the robot knock the gun out of your hand by force? Should the robot incapacitate you, so you can't even attempt to commit suicide for a while? Should it tie you down to a chair and keep talking to you, until it convinces you to not commit suicide?
It gets even more interesting if we assume that robots possess general intelligence, that is they do not take "orders" per say and can make their own choices, but are still built in a way that makes them want to serve their masters.
If such a robot sees that its owner behaves irrationally, it may want to go out of its way to stop him. For example, if the owner is trying to give up drinking, but lacks discipline to do so, then the robot may want to take his credit card and donate a large sum to a charity organisation, to teach him a lesson, threatening to do the same next time he buys alcohol.
Robot ethics is a vibrant field of study these days, and there is a consensus among researchers that, to some extent, robots must be able to interpret their duties flexibly. Having a very rigid set of rules that must be obeyed no matter what - as, for example, Azimov's 3 Robotic Laws would require - is very impractical and is going to severely limit the effectiveness of the robot. However, having an overly flexible set of rules may be dangerous as well, as the robot might decide to forego its duties completely and, for example, murder its master.
Do you think there is a case for letting robots disobey their masters, or even force their masters to do something against their will? If so, how far can they go?
Debra AI Prediction
Post Argument Now Debate Details +
Arguments
My reply .....Surely it’s inevitable as AI advances and gets more and more advanced we will I fear become surplus to requirements , what function would humans possibly serve in a world of super intelligent robots as robots will
function and communicate in their own ever evolving way without need for us
  Considerate: 78%  
  Substantial: 78%  
  Spelling & Grammar: 92%  
  Sentiment: Positive  
  Avg. Grade Level: 9.24  
  Sources: 0  
  Entity Sentiment Detection: letting robots    intelligent robots   reply   function  
  Relevant (Beta): 62%  
  Learn More About Debra
  Considerate: 96%  
  Substantial: 80%  
  Spelling & Grammar: 92%  
  Sentiment: Neutral  
  Avg. Grade Level: 8.66  
  Sources: 0  
  Relevant (Beta): 96%  
  Learn More About Debra
If, however, our intent is to create a sentient, true artificial intelligence at that point, I would consider it life, and in my opinion, all sentient, intelligent life forms have a right to life, self-determination, and all the typical rights we'd consider humans to have. So, this means they would have the right to disobey someone and to call the person a "master" implies slavery. I consider slavery of all such life forms, even if artificial, to be immoral.
Currently, it's believed in the scientific community that dolphins are the second most intelligent species on earth, following humans. [1] [2] [3] I think it's a no-brainer(pun intended), that any creature displaying intelligence, sentience, and society like that should never have "masters", both dolphins and such a sentient robot.
-Albert Camus, Notebook IV
  Considerate: 88%  
  Substantial: 90%  
  Spelling & Grammar: 98%  
  Sentiment: Positive  
  Avg. Grade Level: 13  
  Sources: 4  
  Relevant (Beta): 95%  
  Learn More About Debra