However, I also write emails to reduce time spent in conversation, so while I have learned to be more terse, when I'm not, it's to provide references so that I can shorten or preclude a conversation.
An ML email shortener is a fantastic idea for business comms, with the caveat that short email culture can also reward vagueness.
I would pay money for a tool that checks my technical writing for reading level and warns me when passages go above Nth grade. Most people in the US graduate high school with an 8th grade reading level, and some people may not be able to read above a 5th grade level. That means tiny words and short sentences are king.
There are also metrics for clarity and simplicity that proxy on number of syllables. Those are really useful and don't require a complex ML model.
Actually, now that I think about it... this might be a good source for bootstrapping further training data.
[1] https://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readabi...