LLMs are generating horrible enough answers to questions as it is without deliberately biasing the model, but you do you China.

...Not that everybody else aren't doing it too, and without any external political prompting. Yeah, I'm looking at you Musk.