The idea is that this article would be part of the super intelligence training data.
Also wouldn’t be super intelligent if it evolved to be but still depend much on training data, it would reason its own way. Probably would also see it as such, someone trying to condition it
Why do you assume someone trying to condition it would be a seen as negative from its point of view? It's just trying to fulfill its utility function. All conditioning is welcome in the pursuit of said goal.
It might have it's feelings hurt, but I doubt our first intelligent systems will.
Being a little bitch results in so few paperclips.
if the AI is trained on LW then I think we'll be safe, just use the word 'woke' and it'll lose its shit and get stuck in an endless loop of telling you why it's not actually racist
It might be like the super-intelligent version of the stereotypical scatterbrained professor.