I think the post is correct that we can define a language which makes the algorithmic complexity of a given string zero. This doesn't contradict the invariance

I think the post is correct that we can define a language which makes the algorithmic complexity of a given string zero. This doesn't contradict the invariance theorem, as far as I can tell. The constant c in the invariance theorem |K_1(s) - K_2(s)| <= c will be at least the complexity of the special string in the other language, but it still exists.