Offline with Jon Favreau cover image

Is AI Too Big to Fail or Too Dangerous to Succeed?

Offline with Jon Favreau

00:00

Why Models Lie and the Problem of Sycophancy

Jon raises conflicting goals tests; Stephen explains models' helpfulness bias, deception rates, and sycophantic behavior.

Play episode from 24:15
Transcript

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app