AXRP - the AI X-risk Research Podcast cover image

1 - Adversarial Policies with Adam Gleave

AXRP - the AI X-risk Research Podcast

00:00

Are There Any Misconceptions About Security?

i think one misconception people ist, you know, because it has wod adversary alome a title, this is all about security. Security is a good way of doing a kind of worse case testing, having that security mind set. It can be relevant generally, when you care about robustness or policy. And even if a policy seems to be completely reasonable when you'e test in e bunch of different ways. Even ven it can harbour fse kimes i like really surprising failure modes.

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