I've been getting a small number of emails and message from people because of this. This is how it has been going:

Me: Oh, this is a nice email. Wait, something seems off. Let me check this in Pangram... Oh, 100% AI predicted.

I respond: Hey, thanks for the note. Just FYI, Pangram doesn't really like your writing style.

Them: Ahh, you got me. I use AI to help me write. Blah, blah...

---

I really don't know what to do here. I still don't know if I'm talking to a human. If you are messaging me because you like my writing and you're using AI to write these short messages for you, I don't know what to tell you. Please stop. It takes two minutes to write a quick message... I really don't care if there are typos or it's a bit awkward. But if you're using AI to facilitate this I'm not going to know if I'm ever talking to a person! And, there is a nontrivial chance I'm going to report your account for spam, which I really don't want to do if you're a human and a fan of my writing.

Using AI for these kinds of short messages is poisoning the well in ways that many people don't seem to understand.

It's the strangest feeling... getting a flattering note, and ending up not even knowing if anyone actually read my blog in the first place.