Ran into some more AI-generated posts today. Not even by sig campaigners, lol.

I'm going to open a thread in Reputation which acts as a reference for reports on AI posts. Before doing that, I'd like to get any final feedback on how to proceed. Note this is simply for attaching references to reports, not for any kind of tagging (although if several of a user's posts are deleted as spam then a neutral indicating they are an AI spammer may be warranted).

After much deliberation and testing, here are the criteria I will be using to report posts as AI spam:

A. These 3 AI content detectors must all give a result of the post having a strong likelihood of containing AI-written material:

1. Hive Moderation - 80% likely to contain AI generated text or above
2. Writefull GPT Detector - 80% likely to come from GPT-3, GPT-4 or ChatGPT or above
3. Writer AI Detector - 20% human-generated content or below

B. There must be at least 3 posts by the account that meet the requirements specified in A.

So, if its just a one-off or even two-off, I think the poster deserves some leniency, but if they make a habit of it, I strongly think they should be named and shamed and their posts reported.

C. Clues to help identify an AI poster before putting their posts through the detection process include


and if their posts are frequently riddled with phrases mentioned here.

Anyone who thinks this shouldn't be done, or should be done in a different way, please feel free to chime in now or forever hold your peace. Thanks.