The point raised in the article is an interesting one. I wouldn't dismiss it as merely "nitpicking," but rather as a valuable reminder that the language we use matters. I agree that overusing personification ("AI thinks, reasons, plans...") can be misleading.