338. AI SciFi — Physicist, Science Fiction Author, and AI Expert David Brin on ChatGPT and Whether AI Poses an Existential Threat
The Michael Shermer Show
00:00
The Importance of Reciprocal Accountability
The one thing that has worked and the one thing that worked was reciprocal accountability. We have five well-tuned arenas within which adversarial accountability results in positive sum outcomes. In other words, if we design an arena for AIs to compete with each other and design incentives so that they can tattle on each other when they are about to do bad things, then the Elie Ezer's Euchowski's scenario doesn't happen. It's right there in front of us. So what I'm trying to say is we need a six arena. What are those five areas? Well, first off, if any of you in your audience don't understand the term positive sum, it
Transcript
Play full episode
