@lonelyowl @Darbzilla @DutchBoomerMan @Fever @Formerly_Trucks @Nicezilla @Sui @kekkerel

> It would also be interesting to not just automatically tell them to fuck off, but to use falcon-llm to automatically respond with something meaningful.

It would, but given that the point is to stop them from wasting my time, it feels like it defeats the purpose if I put too much effort into it.

> Maybe even train a crazy racist lora

Actually, it uses the Markov bot I had laying around ( https://git.freespeechextremist.com/gitweb/?p=embot;a=blob;f=embot;h=21a11493cec9b361ee54884dd3de10fce8750cb2;hb=refs/heads/master ), trained on the people that were bothering me when I added that. Basically it is designed to simulate an awful internet argument, so it generates fake >quoted text to reply to, and then it inserts some stock "dumb internet fight" phrases, like "u mad?" and "do you even lift" and a few others. (Example attaxhed.) It is a retarded bot and people *still* don't realize they're talking to a machine until several posts in. That is what got me thinking about the "human bots" topic: https://freespeechextremist.com/notice/AIB2yWD51ohVAc7ekK :

> Conversations with ALICE are "stateless"; that is, the program doesn't remember what you say from one conversational exchange to the next. Basically it's not listening to a word you say, it's not learning a thing about you, and it has no idea what any of its own utterances mean. It's merely a machine designed to formulate answers that will keep you talking. And this strategy works, Wallace says, because that's what people are: mindless robots who don't listen to each other but merely regurgitate canned answers.


> But unfortunately, my gpu is rubbish 😟

Except that it's a bit starved for RAM (8kB), you could run a Markov bot on a Game Boy. Large-model AI shit is what you do if you can't come up with an algorithm.
bot_responses.gif