Not just a problem with chinese models. Try asking western models about reverse engineering malware and they will all decline, because the malware is copyrighted! Hah.
I asked Gemini 2.0 Flash (as well as its thinking counterpart) who is the president of the United States, and it returned a red danger icon. It makes perfectly sense that an llm is aligned with the values of the people who built it, so I don't understand why people treat it as a big deal. It's not as if they'd find the truth about Tien An Men in Chinese history textbooks either.
I don't see that red danger icon. It just tells me:
"I can't help with that right now. I'm trained to be as accurate as possible but I can make mistakes sometimes. While I work on perfecting how I can discuss elections and politics, you can try Google Search."
