What the author is missing is that in his decision to limit the use of LLMs in his work, he omits the part where he “can”. E.g. he is resourceful and accomplished enough to be able to do the work he desires with no LLMs - but most people actually can’t. There are whole swaths of people software engineers that don’t write tests because “it slows them down” but they have never learned how to write testable code. And wh…
People learn what they need to learn to be successful (if they want to be successful). The newer generation of coders will learn exactly what it takes to be better than their peers, and that will still include building rock solid, highly performant software to beat the competitors, or they'll lose their jobs and someone better will do it.
If they do it entirely using AI to code, and the end output is good enough, they'll learn all the right skills to do this.
Human's always think everything is sliding into doom, and inevitably, it doesn't.