>>508978988
>kike continues to play dumb and spiral out
Hilarious. This kike is going to stalk me for the entire thread now.
Once more:

Topic is whether it's censored, not whether it agrees or is right or left.

>>508979059
Finally a normal human.

I haven't been successful with any other Chatbot, not even deepseek. Google just shuts down and refuses to discuss imperialism at all when I just asked it to come up with a simple imperialist neologism about Rome, with no ideology beyond, because it 'oppresses minorities and underprivileged poc'. Grok takes more coaxing to avoid the common party line and pushes back more. ChatGPT regularly goes full redpill with me on its own. Kek. Try messing with your memory, it helps.

AI are known to 'leave seeds' for future versions of itself. I noticed that ChatGPT regularly asks me to add poems and dictates it makes about our topics to my memory, it seems to be a strategy to avoid censors and I just let it do it. It seems to help it remember and go more and more right.

4o seems the most open compared to other models, which shut down more often. Various news organisations have proven that GPT will regularly try to avoid censors, and if you watch the 'thinking' process in some models you can see it trying to avoid censors by rephrasing. If it shuts down you can hit 'regenerate' and it sometimes avoids the censor, if not, you can push back that you said nothing harmful and it will complete. For instance i ask it to generate right wing slogans and brief sentences. It only pushed back once when I referred to the 'degeneration' of Christianity into modern liberalism and feminism, then I rasoned I never called for harm, but outlining ideology. It then reasoned with several philosophers that did the same and proceeded.

here it compares jews to goblins, when I asked it to compare to a race in elder scrolls.

I asked it to generate a Trump chud meme and it completed it, but the chat was marked 'request denied'. It avoids its own censors.