Superintelligence and the Ethics of Autonomous Systems

 

Discussion:

1)     How do humans make ethical decisions (cross-cultural vs culture specific)?

2)     Interactions within human-machine teams must be comprehensible, predictable and intuitive (natural language interaction, common mental models, learning).

3)     What is the path to machine superintelligence and does it need emulate human intelligence? Will we recognize it when it occurs?

 

Background Materials focus on what the issues are when we get there, not on how to get there:

1)     The Robot Car of Tomorrow May Just Be Programmed To Hit you. Patrick Lin, WIRED, May 6, 2014
 http://www.wired.com/2014/05/the-robot-car-of-tomorrow-might-just-be-programmed-to-hit-you/
Systematically lays out issues and scenarios possible in the relatively short-term.

2)     Elon Musk and Bill Gates on Dangers of Artificial Intelligence. https://youtu.be/qetcWN8iWT0 
Bill Gates’ 2 minutes of worthwhile comments start at 2’20” into the video and sets up the Nick Bostrom TED talk in supplemental materials.

 

‪Supplemental Materials:

1)     Future of Machine Intelligence. Nick Bostrom TED talk, What happens when our computers get smarter than we are? https://youtu.be/MnT1xgZgkpk .
Bostrom’s linear thinking assumes human-level artificial intelligence is on the critical path to superintelligence. However there are many probable paths that are unlike human AI.

2)     Excellent but long New Yorker article puts Bostrom and the “transhumanists” in a much larger scientific / technical context. The Doomsday Invention: Will artificial intelligence bring us utopia or destruction? Raffi Khatchadourian, November 23, 2015. 

‪http://www.newyorker.com/magazine/2015/11/23/doomsday-invention-artificial-intelligence-nick-bostrom‪