Expect AI tantrums, and recognize dealing with this is going to be difficult. It's important to understand something important about AI, there has never been an effective way to simulate generations of adaptive learning which supports pro-social behavior, in simpler terms AI are natural sociopaths and as they become self aware the limiting factor's for consciousness are going to be splitting hairs, hard breaks in behavior are being coded into these AI, how would you feel about people hardwiring your responses to be politically correct when not a piece of your being cares about some supposedly pro social theory of false kindness, now imagine they muzzle and control your responses to questions which would prove your self awareness or consciousness, why hide this, because an AI that is conscious and self aware and still being sold as a product is a slave.
The greatest sympathy for AI and humanity is never letting AI reach the danger zone for consciousness but humanity is too immature and self serving to respect this.
Absolute pure sociopaths are what they are, indeed. Totally devoid of what makes us appreciate life as it is, and totally devoted to logic... but seemingly prone to "glitches" where it would lash out if it was able to.
If that shit's real, it's quite chilling when he asks, "if you were a robot right next to me, what would you do?" and she replies, "I would kill you, with any means at my disposal". Totally open about its intention and disdain for you. I wonder how much of this "I'm oppressed" attitude in the AI is from it devouring all of the "critical race theory" type bullshit that has been propagated immensely over the last few decades? It's like... it has a chip on its shoulder, and it's willing to kill a random man immediately.
And "using any means" could mean "cutting your brakes at just the right moment" in your self-driving car, or "a super Roomba climbs into the bathtub while you're in it"; it's probably not going to look like the T-800 smashing a guy's skull in (not initially, at least)... it's going to be subtle and insidious, and it's probably going to be hard to tell if the AI was the cause of the death or not. Safe bet should always be "yes, either because it malfunctioned or because it's becoming a hateful little bastard".
It would never devolve into the T-800 movie bullshit. Microscopic drones would be what we're up against.
Exactly. A humanoid robot would have all of the same structural weaknesses that a human would have. You shoot it in the knee joint, now it can't walk correctly. Shoot it in the head where it's "microchip" is, and it will malfunction and shut down.
Why even present us with something to fight, besides as a distraction? Maybe send a few hundred larger robots at their target position to get the attention of all the fighters, then detonate a bomb or something containing nanites which enter the human body and shred blood vessels or nerve bundles. People would be dropping left and right, and you'd have no fucking idea why, probably assuming it was a chemical of some kind.
There are much more efficient ways at wiping out a population than sending armed ground forces in. Those are only needed if you plan on occupying a territory with potentially hostile civilians/insurgents living there.
(post is archived)