> Each block of Python or Applescript that they generate for me is a much better learning tool than a book - I'm going through the code line by line and researching everything.
I have been doing something similar. In my case, I prefer reading reference documentation (more to the point, more accurate), but I can never figure out where to start. These LLMs allow me to dive in and direct my own learning, by guiding my readings of that documentation (i.e. the authoritative source).
I think there has been too much emphasis (from both the hypesters and doomsayers) on AI doing the work, rather than looking at how we can use it as a learning tool.
Couldn't agree more. On a large and open ended feature I sometimes struggle with where to start and end up researching something tangential. Cool learning, but not efficient.
Claude Code gives me a directory, usually something that works, and then I research the heck out of it. In that way I am more of an editor, which seems to be my stronger skill.