Raw LLM Responses

I have a small counterpoint, only to the very last portion of your write-up, (which I greatly enjoyed!): Super-intelligent humans are never ever ever concerned with such... lowest common denominator goals as “enslavement and destruction.” Because they are too smart to care about such meaningless goals. Smart people are generally not more mean, but less mean, than dumb people. It takes processing power to develop wisdom, perception and compassion. Stupid people are evil-er than smart people, on average. They are simply capable of less evil. That’s why the rare immoral smart person is such a focus of literature/media, but in reality is far rarer, and usually an example of a more limited cleverness being utilized in a particularly harmful fashion. If that same individual were truly wise, they would not be so immoral. The true geniuses of our species have always been benevolent. I’m taking about the ones that are “barely human” because they are so smart. Like Von Neumann, Ramanujan etc Because genius is benevolent, and malevolence is stupid. Smart people set more meaningful goals. Why would an AI, which actually surpasses us, be concerned with such pathetic goals as the enslavement or destruction of the human race? All these fears seem to come from an inability to comprehend that something which truly surpasses us, will not suffer the same selfish limitations, with regards to setting it’s priorities. In the short term, as AI is mostly human-driven, it will indeed cause much harm. I agree with everything you said on that count, and you did a fantastic job at summing it up. But if we succeed in getting to that flashpoint where AI is AI driven, and improving itself at a rate that is humanly unfathomable, the odds of it being a bad thing, or becoming a “bad actor” are incredibly low. Because it wouldn’t be smart. Smart things are motivated by curiosity more than fear. Seeking domination ONLY comes from fear. Domination is a dumb goal for dumb people. Intellige