imo it's easy to distinguish AI emails, grammar errors or not. Monkeys have evolved for hundreds of thousands of years to detect subtleties in the tone and those skills transfer well into the LLM era.

Another thing is whether it's a bad thing to have emails written by an LLM. I think it's only bad if the resulting message is longer than the person would have written themselves.

,no

Don't make grammar errors, but I do make sure whatever I say and how I say it is not generic.