AXRP - the AI X-risk Research Podcast cover image

20 - 'Reform' AI Alignment with Scott Aaronson

AXRP - the AI X-risk Research Podcast

00:00

Theoretical Computer Science and AI Alignment

Theory of open AI rested heavily on like, wow, wouldn't it be awesome if we could just do something like that? But with in place of the polynomial time verifier, we have a human. And I was never quite able to make that work as more than an analogy. So there's a whole bunch of really interesting analogies between theoretical computer science and AI alignment that are really, really hard to make into more than analogies. Now admittedly, you know, I could only get so far with that, right? Like, I don't know how to do that for, well, I must become a joke!

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