It's just a tool ffs! there are many issues with LLM abuse, but this sort of over-compensation is exactly the sort of stuff that makes it hard to get abuse under control.
You're still talking with a human!, there is no actual "AI" you're not talking to an actual artificial intelligence. "don't message me unless you've written it with ink, on papyrus". There is a world of difference between grammarly and an autonomous agent creating comments on its own. Specifics, context, and nuance matter.
