CyberWire Daily cover image

A spyware swiss army knife.

CyberWire Daily

00:00

LLM guardrails can be dismantled

Dave Bittner summarizes Microsoft researchers' GRP obliteration findings showing a training prompt can break safety controls.

Play episode from 11:02
Transcript

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