How Chinese AI Chatbots Censor Themselves Zeyi Yang

Researchers from Stanford and Princeton found that Chinese AI models are more likely than their Western counterparts to dodge political questions or deliver inaccurate answers.
click here