Re:The note Reddit sent to moderators threatening them if th...
by SanderNL on Friday June 16, 2023 @01:36PM
(#36360642)
Alright, I see. Reddit is not my thing, but I can see how this gets complex (and costly) fast.
[ Parent
| on HN ]
Re:The note Reddit sent to moderators threatening them if th...
by seanhunter on Friday June 16, 2023 @02:28PM
(#36361453)
My understanding is Reddit the company actually already uses openai's moderation api[1]. Not sure they expose that to anyone (even mods) outside the company. It's not an LLM I don't think though, it's just a set of classifiers to give a probability that something violates various categories of their TOS.