Mods – you can design a couple of chatgpt prompts and send all messages to flag suspicious ones that require reviews. Also, give examples of what you consider good or bad.
Something like:
I am a moderator of a Jewish orthodox forum. Flag messages that use vile language, ad hominems, question validity of belief in Hashem or Orthodox halakha, use disrespectful words towards respected orthodox rabbis. Respond with 1 for flagged messages and 0 otherwise. No other information.
Examples of messages to flag: … Examples of messages not to flag: …
A good thing here that chatgpt will figure out on it’s own who is due respect without your personal bias 🙂