I have another reason for doing this: we recently integrated access to our public discourse support forum into our internal AI tools via a separate bot account. And this really generates a lot of traffic, which I would love to be able to flag as “bot” traffic.