NotebookLM Prompts

You are creating a deep, intellectually serious podcast episode about epistemic foundations in the rationalist tradition. Your sources are Map and Territory (Sequences Book 1) by Eliezer Yudkowsky, chapters 1–3 of Thinking Fast and Slow by Daniel Kahneman, and Scott Alexander’s essay “Epistemic Learned Helplessness.” Cover these concepts in depth: • The map-territory distinction and why it matters for clear thinking. Your beliefs are a map. The world is the territory. The map is never the territory, and the useful question is always “how accurate is my map?” not “is my map true?” • Bayesian updating as the correct way to revise beliefs. Not “changing your mind” in the folk sense — adjusting probability assignments proportionally to the strength of new evidence. • Calibration as a trainable skill. When you say 70%, you should be right 70% of the time. Most people are wildly overconfident. Superforecasters train this. • System 1 vs System 2 thinking and how they interact with rationalist practice. The fast, automatic system isn’t “wrong” — it’s a heuristic engine that trades accuracy for speed. Rationalist practice is mostly about knowing when to engage System 2. • Epistemic Learned Helplessness — the idea that some people rationally choose to stop engaging with arguments because they can’t tell good ones from bad ones. This is a genuinely hard problem, not intellectual laziness. Tone: serious, intellectually engaged, no dumbing down. Assume your listeners are smart adults who are new to this framework but don’t need hand-holding. Skip basic probability theory — go straight to the implications. The goal is for someone to finish this episode and immediately start noticing their own epistemic failures.