20 - 'Reform' AI Alignment with Scott Aaronson
AXRP - the AI X-risk Research Podcast
00:00
The Future of Machine Learning and Cryptography
To be effective, to be robust against that kind of attack, whatever behavior is back stored in should be something that the AI would have considered doing in the normal course of its operation. You could also say, if the AI knows that it would never want to shut itself down in any circumstance, then it could just make a trivial modification to itself and not do that. I'm now fairly confident that that's going to be part of the future of both machine learning and cryptography.
Transcript
Play full episode
