As we all know there’s surge of LLM-powered comments, ranged from chatgpt-esque style to downright rude comments infesting reddit. It makes the place from unpleasant to extraunpleasant place. The problem I’m worried the most is if they’re starting to invade lemmy and friends. Any development to combat this? Like maybe akin to botdefense.
You must log in or register to comment.
With the BotDefense team leaving Reddit due to increasing hostility, it sure would be nice if they could find a new home in the Fediverse.
Poe’s law, but it’s chatGPT and centrists
Create a community especially for bot cage matches and restrict them to post there for the enjoyment of all
If we find instances infested with bots and unwilling to deal with them, we defederate them. It at least makes it more difficult to mass-infest the network with bots if we have basic things like captcha, email verification, and applications implemented in all the major instances. cough
I think this needs to be a cultural change and for our tools to be self employed and socially spread. We cant rely on admins, mods or botcheckers as they may be corrupt or make mistakes. If there are any serious efforts against bots I’d like to know about em.
