Just as the US and UK refused to sign an international statement about AI safety at the AI Action Summit earlier this week, an AI study out of China revealed that AI models have reached a "red line" ...