I do agree. I still think that the article articulates a very interesting thought... the better the input for a problem, the better the output. This applies both to LLMs but also for colleagues.