I’m reminded of the Star Trek (classic) episode where Spock is playing 3-d chess with the Enterprise computer, and beating it. He tells McCoy that since the computer was programmed by Spock to play chess, he should not be able to beat it because it would never make a mistake, unlike any of us.
That would require 100% perfect coding, which is impossible: the more complex you design something, the more likely unintended consequences are going to occur. All the computer aids in the world cannot overcome a dumb person trying to sound smart, or a smart person who knows better than the computer what s/he wants to do.
I wrote a program to solve Cracker Barrel’s peg game. The program solved the game better than I did - found every possible solution.
The hardest starting point is to put the empty peg in the middle.
I’ll agree 3-D chess is another story.