As AI models go rogue, do you still trust OpenAI and Anthropic to stop them?theguardian.com3 points·jimnotgym··0 commentsOpen articleSaveView on HN