Search Box

Wednesday, February 17, 2016

Quixote: A New System To Teach AIs "Value Alignment"

Storytelling May Be the Secret to Creating Ethical Artificial Intelligence

Aaron Krumins | Feburary 16, 2016



One of the more disturbing trends in robotics is how often some researchers gloss over the moral complexities of AI by suggesting Isaac Asimov’s “Three Laws of Robotics” will be sufficient to handle any ambiguities robots encounter. This only serves to demonstrate how unfamiliar many technologists are with the depth of the issue, for as a closer reading of Asimov’s work reveals, the three laws of robotics leave plenty of room for disastrous outcomes.
There are encouraging signs, however, that at least some other researchers are taking the problem seriously. Two figures leading the charge in this direction are Mark Riedl and Brent Harrison of Georgia Institute of Technology. They are pioneering a system called Quixote, by which an artificial intelligence learns “value alignment” by reading stories from different cultures.

Scheherazade is a storytelling AI that writes crowdsourced choose-your-own-adventure games (http://t.co/VUTZXaRz9L). Source: http://www.scoopnest.com/user/CNET/643383448200179712

<more at http://www.extremetech.com/extreme/223054-storytelling-may-be-the-secret-to-creating-ethical-artificial-intelligence; related links: http://www.news.gatech.edu/2016/02/12/using-stories-teach-human-values-artificial-agents (Using Stories to Teach Human Values to Artificial Agents. February 12, 2016) and http://www.cc.gatech.edu/~riedl/pubs/aaai-ethics16.pdf (Using Stories to Teach Human Values to Artificial Agents. Mark O. Riedl and Brent Harrison. 2015)>

No comments:

Post a Comment