FUTURATI PODCAST cover image

Ep. 129: Applying the 'security mindset' to AI and x-risk | Jeffrey Ladish

FUTURATI PODCAST

00:00

How to Build Super Intelligence

I think it actually becomes much harder to like go into superhuman domains if you're just training on human data. You need to start training on like your own performance or something. I don't know we should go too far into how to build a GI or how to build super intelligence, because that's not a great domain to speculate about. And somehow this, and like this, and imagine the scales to superhuman capability levels. Now we have to ask like well, what are the goals of that system? The thing is, I don't fucking do, I'm not sure my prior is very good at thinking in terms of these kinds of things.

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