Your AI red team may be finding the same prompt injection thousands of times

AI agents retrained on a larger pile of prompt injection attacks came out easier to hijack in three of four tests run by researchers at the University of Waterloo, Zhejiang University and KTH Royal Institute of Technology. The study found that the automated red-teaming tool that found more working attacks produced the weaker defence.

Source: www.helpnetsecurity.com

Curated Permalink