The Gradient: Perspectives on AI cover image

Connor Leahy on EleutherAI, Replicating GPT-2/GPT-3, AI Risk and Alignment

The Gradient: Perspectives on AI

00:00

The Radiant Idea of AI Alignment

It's basically the most radical idea about AI alignment is just maybe this won't go okay by default, which isn't that radical of an idea, really. It continues to shock and surprise me how often I talk to people like, you know, and these are not just like random off the street but professors of AI at reputable institutions. Usually, it's still hard to get like full on funding necessarily to do,you know, full time alignment where he is getting better though.

Transcript
Play full episode

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