In addition, LLMs make excellent _critics_.

Posted Jun 26, 2025 16:27 UTC (Thu) by davecb (subscriber, #1574)
Parent article: Supporting kernel development with large language models

Back in the October 2024 Communications of the ACM, Advait Sarkar wrote "AI Should Challenge, Not Obey.
Let’s transform our robot secretaries into Socratic gadflies" , at
https://cacm.acm.org/opinion/ai-should-challenge-not-obey/

I've used them successfully as "linters" for both code and prose.