Training humans to treat AI chatbots as more than what they are is far more dangerous than what Anthropic is trying to prevent.
Comment