Ethical Artificial Intelligence

Citation data:

CONFERENCE: Research & Exhibition

Publication Year:
2017
Usage 85
Abstract Views 85
Repository URL:
http://digitalcommons.lmu.edu/honors-research-and-exhibition/2017/section-03/6
Author(s):
Drewe, Mackenzie
Tags:
Film and Media Studies
conference paper description
In an age where technology is rapidly advancing and artificial intelligence is coming increasingly into focus, new questions are starting to arise. Researchers are trying to discover the best way for artificial intelligence to learn how to complete tasks with less help from humans. Currently, the most effective way for machines to learn is adaptive machine learning, which gives machines minimal instructions for a task and lets them learn on their own through repeated trials. It is unknown everything a machine could accomplish using this type of learning, including if they could learn how to be ethical. If it were possible for machines to learn ethics using adaptive machine learning, it would create unknown consequences for humans that are commonly explored in films and television. Combining the fictional portrayal of ethical machines in film with the reality of teaching machines ethics through adaptive machine learning, there is a lot to be uncovered before technology gets too ahead of us.