AI Already Knows Us Too Well

AI Already Knows Us Too Well

A few weeks ago, GPT-4 prompted me when I logged in. “Would you like to see my description of you, based on our chats, to share on social media?” the chatbot asked me. Being an AI ethicist, I wearily answered “yes” to see what it was up to. It then generated a flashy paragraph about my personality traits. I did not share it. But days later, after a quick web search, I could see on platforms like Reddit and LinkedIn that numerous users had enthus…