In this episode we dig into why ChatGPT sometimes gives you garbage outputs and what you can do about it. From hallucinations and drift to bad prompts and long chats, we’ll cover the common pitfalls, what to do when it sucks, and how the imperfections of AI might actually be keeping your own skills sharp.



Main Topics Covered




WTF moments with bad ChatGPT outputs



Why it’s math, not magic (probability, tokens, predictions)



Hallucinations and why “AI slop” happens



Improving your prompts



Memory checks and correcting drift



When to start a new chat



The importance of knowing your own stuff



Skills-of-operation vs skills-of-understanding



Historical parallels: calculators, smartphones, cameras



A solution to the concern about outsourcing thinking and expression



Weekly use case: Joe’s meal planning with ChatGPT




Links & Resources for This Episode




Listen to Ep. 1: What You Actually Need to Know about ChatGPT



Listen to Ep. 8: Getting ChatGPT to Do What You Want



Listen to Ep. 9: Helpful ChatGPT Features You Might Not Know About



Subscribe to the ChatGPT Curious Newsletter: The Curious Companion



Submit a Question



Visit the Website



Feeling curious AND generous? Click here to support the podcast.

Chapters
(00:00:00) - ChatGPT Curious(00:00:38) - Why Chat GPT SUCKS & What to Do About It(00:05:07) - How to Train ChatGPT to Stop Sucking(00:08:14) - ChatGPT: What You Can Do to Make It Less Stupid(00:13:17) - Talking to a project in a chat(00:13:38) - GPT: Know Your Right(00:14:37) - ChatGPT: Its Stupidity has Value(00:15:14) - Skills of Operations vs Skills of Understandability(00:19:58) - The Problem With Out-Of-Touch Thinking(00:21:57) - How I Use Chat GPT(00:23:48) - ChatGPT: Curious

Ver todos los episodios