porcupinedream Posted July 2, 2014 Posted July 2, 2014 Hey guys, I’m new to the FDR forum. I’ve been a keen follower of Stef’s for around two years now, and after getting to grips with most of the ideas here, I have finally built up the courage to get involved with the community and explore further. A topic I would like to put forward for discussion is that of Artificial Intelligence, and how it will affect the future of human societies. There are many obvious benefits to using this technology, but recently in an article by Hawking, he suggests that we’re not taking the potential problems of AI seriously enough – http://www.independent.co.uk/news/science/stephen-hawking-transcendence-looks-at-the-implications-of-artificial-intelligence--but-are-we-taking-ai-seriously-enough-9313474.html I would love to hear a podcast from Stef on this subject, and if there is already one available I would be grateful for anybody that can steer me in the right direction. The question to everybody here would be: How important do you feel this subject is for discussion, compared to subjects such as family/relationships and anarcho-capitalism etc? We often hear about creating a better society for future generations, but what if things get progressively worse for future generations, due to such advancements in these technological fields. Is AI mostly speculation? Should we be concerned by such developments as Bina48? I would be grateful for any help or comments
Three Posted July 2, 2014 Posted July 2, 2014 https://www.youtube.com/watch?v=3hvEiNpZrGUHey, man! Welcome to the boards I'm not really sure about AI, however this guy has a good series which sums of the skeptics case against AI.
Wuzzums Posted July 2, 2014 Posted July 2, 2014 I recommend the book "On Intelligence" by Jeff Hawkins. http://en.wikipedia.org/wiki/On_Intelligence It deals with the very concept of intelligence and how it can be reproduced. There's a technical chapter he warns readers in the foreword to skip if they find tech talk boring. But if there's a chapter that should be read, it's that one. The model of the workings of the brain he presents is so simple and elegant, it would be surprising it won't turn out to be a fact. A distinction he makes in the book is between consciousness and intelligence. The latter not needing the former to exist. So in this regard, intelligent machines will be no different than any other tool out there, it's up to us to decide how to use them. We already have had the most destructive tool imaginable at our disposal for decades, and yet it's still the least of our worries. The danger comes from the same place it always did, evil crazy people. Theory aside, applications to thinking software have already been put into practice albeit for mundane task. The AI model presented in the book is put into a program that then learns how to play Super Mario Bros. and other games The ending is pretty shocking:
rjfoley2014 Posted July 8, 2014 Posted July 8, 2014 I really like that book "On Intelligence" by Jeff Hawkins. I really think he's on to something there. I've read a number of AI books and they all seem very contrived even though the theories and approaches can be used to explain some human behavior or emulate some human behavior. His approach really gave me that sense of having arrived at a final solution to the problem of defining and replicating human intelligence because he has redefined what it means to be intelligent. It also calls into question the validity of intelligence tests and the merits of using those tests to accurately measure intelligence. Steph frequently makes reference to intelligence - he grants you 120 i believe but I've never understood what exactly that means and I'm sure he's referring to outdated measures. The fact is we can be intelligent in a variety of domains and traditional tests are inadequate.
NeoCortex Posted July 9, 2014 Posted July 9, 2014 Hey everyone,I just wanted to say I really enjoyed ''How to Create a Mind'' by Kurzweil.
rjfoley2014 Posted July 10, 2014 Posted July 10, 2014 I had trouble getting through that book partly because it didn't seem as clearly structured as his previous books. I also noticed that he refers to some of the same concepts and approaches as Hawkins. I don't think he was as clear about his methodology but it's still worth a look. I'll have to try and read it again.
Recommended Posts