China has a pretty strict policy about not allowing anything that is anti-social and can be used to create mass unrest. A generative AI that can produce millions of fake news stories about political leaders is one of those anti-social things. It was one of the reasons OpenAI was so terrified of the possibilities of what they figured out. China just made it policy instead of sitting idly by whole bad actors automated incendiary publishing.

Deepseek avoids topics that are usually considered political. AI as a hallucination machine is especially useless with political content as it is extremely polluted with propaganda and wildly different views on similar contexts, so it isn’t for censorship. And yes, the people in China know about and speak about their leadership.

哈哈哈哈 这是一个秘密 In all seriousness though yes they know. Deepseek just doesnt allow politics. 小红书 and social media have similar rules. Politics in China is serious not something people need to talk to a hallucinating machine about, or discuss anonymously online. You can see why this is a bad idea if you look at the USA today. Its best to discuss in person and do so calmly and rationally.

Remember chinese netizens have access to all the same information westerners do online. Many speak english, and a VPN is easy to use to go on other sites.