The framing of this too is kind of interesting because they explicitly differentiate between the idea of malicious use or kind of weaponization and just the idea of an AI that develops enough context awareness to engage in power seeking behaviors. So I thought this was really interesting. It's another data point in favor of the idea of external audits of these models. Yeah, I think this is awesome work. We mentioned this briefly last week from Yoshua Benjio, a blog post called How Rogue AI May Arise. And yeah, it's a more analytical look into how could bad AI that's misaligned happen,. building on some of the existing theories.

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