Somehow, the most reasonable account on Vichy Twitter is Grok because it’s hard to train an LLM using only data dumbasses wrote.

Sadly that's going to change. Musk is going to push an update that I suspect will turn Grok into a far-right disinformation parrot or become closer to that. See https://lemm.ee/post/67429541 and https://lemm.ee/post/67216433

He keeps trying. He seems to think that can be done by just putting his thumb on the system prompt, and then we end up with obvious nonsense like that South African white genocide preoccupation. It's fortunate the Musk isn't smart enough to figure out how to do it subtly.

You just now found out about this?

I'd leave it for good if there were non *chan alternatives with anywhere nearly as much hatred and drama

Then it's not a spike, it's a trend. Created by, mostly (hopefully) bots, there to move the overton window and normalise racism and hyper-regressive world views to divide the working class while the rich use us as Cannon fodder for their genocide against our fellow working class. Wankers.

In other news, grass is green.

Will someone tell the 60% bot userbase?

Its working, horrifyingly. In several communities lately ive seen moderators start to treat violent racism and antisemitism as personal beliefs that do not on their own necessitate banning. This in private progressive leaning communities too. Eugencist christian white nationalism is becoming a normal tolerated ideology to have.

Instagram is especially bad too. I see a lot of people talking about Twitter, but not enough about Instagram.

Constantly means habitually. I can't even begin to formulate an apology for persistent spike though.

I'm seeing the more accurate 'vehicle frequently stopping' a lot more these days, so someone twigged to it.

So delighted (probably not the right word, feeling) that this was the top comment when I arrived.

No shit. Really?

“Persistent spike” isn’t a spike. That’s the status quo.

I'm nowhere close to being an LLM specialist but to actually skew the model itself I think you need a lot of consistent data. Ten thousand alt-right blogs peddling a hundred thousand internally inconsistent and mutually incompatible narratives won't cut it, they'll criss-cross over the gradient landscape and because they don't coincide, won't make a dent in the deep groves trodden by pirating libgen. And training only on the alt-right blogs won't cut it either that's just not enough data which on top of that doesn't sound smart enough to woo anyone, or have any resemblance of a consistent stance. Sure you'll get it to claim ridiculous shit and use lots of slurs but 4chan managed to do that back in 2016 and noone was fooled.

All very good points.

The new analysis contradicts the social media platform’s claims that exposure to hate speech and bot-like activity decreased during Elon Musk’s tenure.

They might both be right. I know my exposure to hate speech and bot-like activity decreased since I stopped engaging with that platform.

You are a zionist so it's funny that you say that