You keep on saying that " as you learned" which is just wrong.
Well, I'm assuming you can.
I have always maintained my position that the designers of the AI know what's going on, or that if they made a mistake and not sure what's going on, if they wish to know, given time they will know what's going on.
Which means that in these experiments, the programmer sometimes doesn't understand how the output occurred. In the case of some of these, it's not software-determined, but is the result of subtle effects in the hardware itself, some of which remain not understood.
Could we in principle, find out what they are? Yep. Does that mean that the programmer knew beforehand? No.
Any attempt to say that we don't know or hint that it is not possible for us to know the so called evolution process in computers is a lie
As you learned the man who programmed the stop/go system didn't know. On the other hand, it's probably true that we could eventually figure it out. So, I'm wondering if you realize what you were doing in characterizing both as lies.
and is unscientific since they are mixing science with voodoo.
Nonsense. For example, microwave antennas had always exhibited a persistent hiss at a particular wavelength. Bell scientists struggled to eliminate it, and thought they had excluded external signals because the hiss occurred wherever the antenna might be aimed.
Then they realized that the hiss was the predicted microwave background remnant of the Big Bang. The two engineers got a Nobel for their discovery. It wasn't voodoo when they didn't know what it was. IT was just outside of their understanding at the time.
Emergent systems are a real phenomenon, and can be understood, but a different scientific tool box is necessary and strict reductionism will often fail. It's not voodoo; it's just not simple.
And this is one of our fundamental differences, you keep claiming that the machines evoluted process that no one understands
Rather, no one understands what it's doing. The stop/go computer uses less components than any programmer can use for the same problem. The designer thinks that perhaps the machine uses subtle effects from nearby components. Far as I know, no one's actually figured it out. As you learned from genetic algorithms, evolutionary processes are often able to use effects that no one knows about.
Likewise, the genetic algorithms that found a better diesel engine, apparently utilized some yet-to be understood change in gas flow that made the final result better.
which is basically held voodoo as science
and at the same time knowingly or unknowing, think God does not know what might come
Here, you're underestimating God. As Aquinas pointed out, an omnipotent Creator can use contingency as readily as He can use necessity. He's greater than you suppose.
and that He has to create in-perfect creations to be evolved (to something better instead of from perfection mutate to imperfection).
Or since evolutionary processes are more efficient than design, perhaps He just chose the best way. If you spend some time studying nature, you'll find that for all its apparent complexity, the principles by which it works are quite elegant (in the proper sense, i.e. simple).
I am interested to know which part of the above do you disagree?
I don't agree that things we are yet to understand, amount to "voodoo." I don't agree that God is not capable enough to use contingency. I don't agree that God is limited to what humans can do. I accept the fact of emergent systems.
I think that covers it.