LMs as evaluators and failure modes

The Don’t Worry About the Vase Podcast is a listener-supported podcast. To receive new posts and support the cost of creation, consider becoming a free or paid subscriber.

* 00:00 - Introduction

* 01:47 - Table of Contents

* 03:55 - Language Models Offer Mundane Utility

* 05:06 - Refine Your Paper

* 07:18 - Language Models Don’t Offer Mundane Utility

* 09:03 - Huh, Upgrades

* 09:10 - On Your Marks

* 13:05 - Get My Agent On The Line

* 14:56 - Deepfaketown and Botpocalypse Soon

* 17:18 - Fun With Media Generation

* 18:55 - Greetings From The Torment Nexus

* 19:12 - A Young Lady’s Illustrated Primer

* 22:25 - You Drive Me Crazy

* 22:45 - They Took Our Jobs

* 33:58 - They Are Hiring

* 35:31 - Levels of Friction

* 36:32 - In Other AI News

* 37:47 - Show Me the Money

* 46:33 - Quickly, There’s No Time

* 47:32 - The Week in Audio

* 49:12 - eighty thousand Interviews About AI

* 56:40 - Chapter 140

* 56:41 - The Lighter Side

https://open.substack.com/pub/thezvi/p/ai-161-part-1-80000-interviews?utm_campaign=post-expanded-share&utm_medium=web