{"version":1,"type":"rich","provider_name":"Libsyn","provider_url":"https:\/\/www.libsyn.com","height":90,"width":600,"title":"Shameless Guesses, Not Hallucinations","description":"   I hate the term \u201challucinations\u201d for when AIs say false things. It\u2019s perfectly calculated to mislead the reader - to make them think AIs are crazy, or maybe just have incomprehensible failure modes. AIs say false things for the same reason you do. At least, I did. In school, I would take multiple choice tests. When I didn\u2019t know the answer to a question, I would guess. Schoolchild urban legend said that \u201cC\u201d was the best bet, so I would fill in bubble C. It was fine. Probably got a couple extra points that way, maybe raised my GPA by 0.1 over the counterfactual. Some kids never guessed. They thought it was dishonest. I had trouble understanding them, but when I think back on it, I had limits too. I would guess on multiple choice questions, but never the short answer section. \u201cWho invented the cotton gin?\u201d For any \u201cwho invented\u201d question in US History, there\u2019s a 10% chance it\u2019s Thomas Edison. Still, I never put down his name. \u201cWho negotiated the purchase of southern Arizona from Mexico?\u201d The most common name in the United States has long been \u201cJohn Smith\u201d, applying to 1\/10,000 individuals. An 0.01% chance of getting a question right is better than zero, right? If I\u2019d guessed \u201cJohn Smith\u201d for every short answer question I didn\u2019t know, I might have gotten ~1 extra point in my school career, with no downside. You can go further. https:\/\/www.astralcodexten.com\/p\/shameless-guesses-not-hallucinations ","author_name":"Astral Codex Ten Podcast","author_url":"http:\/\/sscpodcast.libsyn.com\/website","html":"