AXRP - the AI X-risk Research Podcast cover image

1 - Adversarial Policies with Adam Gleave

AXRP - the AI X-risk Research Podcast

00:00

How to Train a Kicker to Block a Bull

Researchers used adversarial palsies to train and defeat random policies. The results were surprising, as the attacks looked very similar to real-life adversaries. But they had no effect on victim's ability to defend themselves from attack.

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