Here it is: The explanation why the link for an appeal in Youtube's emails almost never works.

One user had direct contact with Youtube Partner Support. This case also confirms that Youtube has made mistakes and admits them (upon request).

There really does seem to be a battle going on here between "bot out of bounds" and employees who are "straightening everything out". Is this the future that artificial intelligence is supposed to bring us? Has the battle against AI already begun behind the scenes at YouTube?

Imagine you find yourself imprisoned at the slightest suspicion from an AI, and you only get out if you prove your innocence. Isn't it usually the case that one is presumed innocent until proven guilty? Thanks to AI, this paradigm is now being flipped.

YouTube essentially considers you guilty first, only to say later: Oh, sorry, actually not.