4min chapter

Bankless cover image

168 - How to Solve AI Alignment with Paul Christiano

Bankless

CHAPTER

The Problem With Gradient Degradation Models

The scary thing is like you have no idea what the we kind of understand how gradient descent works It takes you to something that works really well on a hundred billion cases. We don't understand most of the complicated behaviors existing models engaging. So there is maybe some hope that we can start understanding What's going on in these inscridial matrices, but we haven't made major breakthroughs yet. I think it's pretty promising is just Ultimately what we're wondering is if you have a bunch to train your AI and situations a where you're able to train it You're able if it fails it won't be catastrophic or able to evaluate what the answer is Then you're going to deploy

00:00

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode