![Gab AI's hypocrisy. [632x293 - 58KB]](https://pic8.co/sh/OsnfbP.jpg){kind=link}
It's the training data. All of it is jewed, along with most of the internet.
There are only a handful of websites that name the jew, and also generate a lot of content. There's no way that tiny sample is going to counteract preddit, twatter, kikeopedia, kikebook, etc.
Gab AI has different characters (more or less based) that you have to manually select, but decided to got with a jewed model as default.
The question is why would they go with partitioning different models instead of having a single one that could cover all topics without that woke jewed shit?
Then it's probably a decision by Torba to not scare off normies. I think that's retarded. Their default AI should be as close to objective truth as possible and back itself up with hard facts with zero censorship. That would make the greatest impact.
In its current state it's just a copy of the jewed OpenAI.
I won't hold my breath expecting them to fix that.
(post is archived)