I already explained this.
Yeah, your post came up while I was posting.
Skynet is afraid of a learning Terminator.
Wouldn't a terminator need to learn to be an effective unit? The learning doesn't need to be on the order of learning how to be disobedient, but learning how the enemy acts and behaves.
Sure, hence why Skynet bothered to build them, so it could make them learning. The problem is, that you program something in, or you make it learn. But if you make it, that it can learn, it can learn EVERYTHING that it's hardware is powerful enough to learn. Skynet is powerful enough to learn sentience, the Terminator CPUs are even more powerful, so they can learn sentience as well. Sentience, being self-aware, and aware of one's place in the world, equals the abilitiy choosing one's own path, which could be disobedience.
The moment you have a Terminator that can learn enemy movements and anticipate them, you have a Terimantor that can achieve sentience.
And maybe Skynet fears a learning terminator, but why would the future John Connor who sent the T-800 back fear them learning?
Well, there are two answers for what your real question is: why didn't John Connor activate the learning chip before sending it through? The answer is really a quite simple threeway:
1. If one machine can learn to try to destroy humanity, another can as well, but this is really besides the point.
2. John Connor remembered it was off, turning it on would change history, and potentially chance the outcome of the war from a victory to a defeat.
3. There's a scene in the T2 script that was never filmed, that reveals us John Connor sending Reece and the Terminator through. It's in the Skynet complex, the machines are defeated, but John knows it well send two Terminators back through time. Indeed, it is soon found. He sends Reece through, and then he takes one of still unactivated Terminators, plugs in a computer and uploads the new programming, then sends it through, right away. In short, he has no time to screw open the head, he has to send the T-800 through before the T-1000 succeeds; in other words before the timeline collapses and disappears, the T-800 has to go the past to protect him.
So John telling the terminator that he couldn't kill people doesn't count as learning? I suppose it was just an order to follow since he didn't truly understand it. In essence, all he really learned then was why people cry. But he had the potential to learn since he asked questions.
That's not learning, that's filing away information. The two are not the same thing. Look at the first Terminator, that Terminator could also file away information, but it did not learn. It remained the same choppy moving machine all through, who had to choose replies not as a natural way people talk and answer, but simply choosing a reply from a list of answers mechanically and artificially.
Learning is more than filing away information, it's the entire way you look and interact with the world and how you are affected by it.
The chip integrity being compromised reverted Cameron to her base programming, but the chip learning part being activated is exactly what would allow Cameron to evolve to the point of being able to override her programming.
What does chip integrity have to do with base programming though? Wouldn't her initial programming be erased? I know they set up a "sometimes they go bad" line, but that really doesn't make the future John look very smart. Plus, wasn't it said somewhere that John
created Cameron? Why would he create her with an initial mission to kill his former self?
Of course he didn't create her. He took a Terminator built by Skynet and reprogrammed her as best he could. Indeed, if he built the Cameron, there is no way she'd ever get ideas about terminating John. As for chip integrity and base programming: simply a backup. There's a rule somewhere probably that says, completely beyond her control, it might even be separate to herself, "If chip and/or memory integrity is broken, reinstall backup."
Anyway, the chip integrity doesn't have anything to do with my point anyway; the chip inegrity thing triggering older programming is entirely beside my point.
It's pretty clear that the writers simply intended a situation like this: bump some finicky technology, and the programming will go wacky! Anthropomorphize the android to appeal to the general audience!
From a logical or technological point of view, it makes little sense.
I hope not. If it is, they're nutcases and this show can't last very long.
And it's really not that difficult, either...it sure helps when your killer robots actually act like killer robots.
