We need to talk about AI moderation

There are tons of posts right now about people running into issues with an overzealous automod. Of course any model will need to be trained and this itself isn’t a big issue.

What stuck out to me was that the community mod themselves could not approve their own post in their own community. To me, this breaks the contract of being a community curator. If I can be overruled by the AI with no recourse other than to email admins, why would I spend the time creating content on the site?

Again, I have no issue with strict filters. It just feels bad that we as humans aren’t trusted to overrule them in our own communities.

Is this AI-first hierarchy part of the long-term roadmap of the site, or is mods not being able to approve their own posts in their own communities just a missing feature?