That's a terrible take and its desperately trying to draw an equivalence where there isn't one.

I'd argue that the slop code creates more drudgery with having to constantly babysit the LLM. Never mind a new blog post every week about how your "agentic workflow" from last week is all wrong and you need even more infrastructure to wrangle the LLM. It's worse than the way the JavaScript ecosystem used to be!

Reading someone else's code is challenging, but at least with a person you can ask them questions or debate.

I guess I'm just someone who finds reviewing someone else's work tedious, though a necessary part of the job.