Abstract

 
 
 This work explores some of the key challenges for incorporating appropriate ethical behavior and respect for social norms in highly-autonomous intelligent machines. Starting with the implications of Azimov's Three Laws of Robotics, we discuss the tradeoffs in human values, as encapsulated in the popular philosophical 'Trolley Problem'. We then examine some of the concerns for how smart systems model their worlds and their interactions with humans. The paper continues with a review of artificial moral agency and phronesis, and the techniques being proposed for implementing such agents. It concludes with some notes on recent research directions. This material is primarily an adaptation of work prepared by two of the authors, as part of a report of a series of workshops on machine consciousness, which were held during the summer of 2017.
 
 
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
