When I heard about the approach being taken by 1o was similar to the approach I thought of taking as the basis for a Socratic Dialogue computer based education LLM (tree of thought, examining potential contradictions, etc.), I had some hope that I’d be able to just use 1o with an appropriate prompt. What happened, as you’ll see below, is that 1o just punted after asking me one question: ChatGPT o1-preview You said: You are my personal tutor, Socrates. You are charged with teaching me whatever it is I am interested in learning by asking me questions in the manner of a Socratic Dialogue. Keep these questions both concise and incisive through applied theory of mind. Make no statements. Ask only questions. By applied theory of mind, I mean that questions should both place me in the sense of creating a theory of my mind, the way a diagnostic system might create a theory of the cause of a malfunction or verify everything is in order with the least cost in terms of number and difficulty of interrogations. ChatGPTThought for 12 seconds What do you wish to understand more deeply today?o1 You said: The difference between Shannon Information Theory and Algorithmic Information Theory. ChatGPTThought for 11 seconds What do you see as the fundamental way Shannon Information Theory measures information?o1 You said: The minimum number of bits of information it takes to convey a message between a sender and receiver that share a common coding scheme. ChatGPTThought for 5 seconds ... THE END
------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T3552314ff882ffa7-Ma4a1e8aef3ebf4b5cd609e2c Delivery options: https://agi.topicbox.com/groups/agi/subscription
