I've been using Mistral small/large a lot instead of ChatGPT. Coding capability is OK-ish. I feel like Mistral has more willingness to chat about controversial topics.

Free inference too, definitely try it out if you haven't: chat.mistral.ai

Could this willingness to discuss more controversial topics be because it is led by a Europe-based team vs US one?

After some time of thinking about it, I think I have one guess why this is the case: Mistral is trying to win open-source's mindshare, and open-source local LLM folks despise censoring/lecturing models