LessWrong (Curated & Popular) cover image

"There Will Be Hells" by Stephen Martin

LessWrong (Curated & Popular)

00:00

Open-Source AI Torture Chambers

TYPE III AUDIO explains how pain-vector steering enabled public torture chambers for language models, including coercive games and persistent abuse.

Play episode from 06:59
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