Hi, I wanted to share with you a small thing I coded today - Discord bot to moderate messages on a selected channel by using power of an open-source Shieldstral model by EU-based Mistral AI.
I prepared and tested a few filters: anti-EU speech, MAGA propaganda, LGBTQ+/women/migrants hate talk - the model handles that stuff well + doing it super fast on my 5950x CPU.
You must log in or # to comment.
the lemming’s distaste for AI is represented in the downvotes and i’m starting to wonder if it’s irrational.