I'm not saying "LLM's are only capable of AI slop."

We're specifically talking about the context of 'LLM-aided human output' not about general capability of these systems.

An interesting example was the neomacs post on the frontpage yesterday.

Some buttons on the front page don't even work.

And then you have Joachim Breitner using Claude Code to produce Soundness proofs in Lean4, mostly vibe coded.

It really is the QB in charge methinks

	▲	evalexec 13 hours ago \| parent \| next [-]
		Thank you for the feedback, I'm busy on the development of neomacs, so I didn't well polish the neomacs's landing page. I have updated the neomacs's landing page, make the "work in progress" indicator more prominent. Also, I'm working on using neomacs's wasm build as neomacs's landing page: https://eval-exec.github.io/neomacs/
	▲	evalexec 13 hours ago \| parent \| prev [-]
		neomacs is still in it's very early of development。 I believe that, after continuous multiple iterations of development/test， neomacs will undoubtedly become excellent software。