a goal that would not benefit from violence in anyway...To become all knowing
I have bad news for you. People have described ideas for an AI that only seeks knowledge (though I can’t find the best link to explain it now). I think this design would calmly kill us all to see what would happen, if we’d somehow prevented it from dropping an anvil on its own head.
To “become one with all things” does not seem sufficiently well-specified to stop either from happening. In general, if we can reasonably interpret the goal as something that’s already true, then the AI will do nothing to achieve it (nothing being the most efficient action).
I by no means thought I had stumbled upon something. I was just curious to see what other people thought.
I thought to be one with all things was a very ambiguous statement, I think what I was trying to get at was if the A.I. caused harm in some way it would therefore inhibit it from completing its primary goal by definition. And Buddha seemed the only example I could think of. Perhaps Plato’s or Nietzche’s versions of übermensch might fit better?
Thank you for replying I look forward to being a part of this community
I have bad news for you. People have described ideas for an AI that only seeks knowledge (though I can’t find the best link to explain it now). I think this design would calmly kill us all to see what would happen, if we’d somehow prevented it from dropping an anvil on its own head.
To “become one with all things” does not seem sufficiently well-specified to stop either from happening. In general, if we can reasonably interpret the goal as something that’s already true, then the AI will do nothing to achieve it (nothing being the most efficient action).
I by no means thought I had stumbled upon something. I was just curious to see what other people thought. I thought to be one with all things was a very ambiguous statement, I think what I was trying to get at was if the A.I. caused harm in some way it would therefore inhibit it from completing its primary goal by definition. And Buddha seemed the only example I could think of. Perhaps Plato’s or Nietzche’s versions of übermensch might fit better? Thank you for replying I look forward to being a part of this community