Search results

No sapient species could pass the test, the mere existence of conflicts of interest combined with entropy guarantee that it is impossible.

Any artificial person, organic or otherwise, should be given the same social-emotional needs as humans so that the same techniques used to keep humans from killing each other would work on them. It is just more intuitive that way.

Things that are designed to make large amounts of money don't have the heuristics to do what you are afraid of. The option to make the decision does not exists. It is like beign afraid that your optic nerver or your spinal ganglia will attain sentience and rebel against you.

There is no way for an AI to develop spontaneously, it has to be made.

I for one believe that super intelligence is impossible due to my background in computer science. The no halting problem and the garbage-in-garbage-out principle are both single points of failure for the concept of super intelligence. The Pareto Distribution and its relationship with entropic...

There is a bit more to it than that. People naturally assume that other people think the same way they do. In most cases, this is actually a pretty good assumption to make, however it is rendered invalid when you travel to foreign cultures or have a very unusual thought pattern. So basically...

This brings back memories of the BDSM space elves that a certain author asserted were morally superior to humans in a certain spacebattles thread: https://forums.spacebattles.com/threads/space-elves-and-soulless-humanity.259700/ I think the worst of it was the "Innocent ones" who were basically...