4 ms·

HN would probably be against any Sarah Connor jokes, so I'll just observe that I think we are near the nexus where AI and robotics have a (hopefully) harmonious

• randallsquared 17y ago
  > And if not, I hope someone listens to Yudkowsky. I think the chances of that are really slim. EY's best hope is to be the creator of AI, but waiting until he has a full theory of FAI first (if I understand what he's doing correctly) seems to make it much more likely he'll be beat, even if he otherwise wouldn't be (which I have no opinion on).
  • jibiki 17y ago
    I think we could hope that AI advance will be gradual, so that an eye-opening crisis with non-friendly AI occurs before truly strong AI develops.
    • randallsquared 17y ago
      We can hope that. Another thing we could hope is that we don't really figure out what makes intelligence tick until after the (temporary) end of scarcity or until after uploading is possible. Given our problems with building large software systems, it seems possible to me that we'll be able to copy the brain before we understand enough to build intelligence without such copying. This puts me in the bizarre position of being very interested in languages and systems which directly address complexity, while hoping in some ways that we fail to figure out how to build significantly more complex systems from scratch. :)