I spent a lot of time studying Nietzsche in college (while getting a degree in philosophy) and it is pretty annoying how the pop culture conception of his ideas has so little to do with what he actually wrote.
I think Nietzsche would find the “entrepreneurial” rebranding of his ideas to be irritating and frankly not the audience he wanted. He was writing for a very, very specific group of people, not a mass market in any sense. He doesn’t care about you losing your job because a robot took it, he is concerned with far more consequential and foundational issues. Nietzsche himself was very critical of “merchants” and technology more broadly, so I think he would find the idea of LLMs being treated as actual conscious entities to be a hilariously stupid joke, more indicative of how society’s standards have fallen than how high AI abilities have risen.
Anyway, rather than engage in a long comment on why Nietzsche would find this article annoying – I do think there is some value to be had in using AI tools as philosophical conversations.
Personally I’ve gotten a lot of value by proposing a certain book theme or argument to ChatGPT and critiquing it, exploring other books on the topic, and so on. Previously this required someone willing to sit and debate philosophical questions with you, which isn’t everyone’s favorite activity :)