Indirect Prompt Injection Attacks Against LLM Assistants (schneier.com) 6 points by lootsauce 1y ago ↗ HN
[–] lootsauce 1y ago ↗ The recent DEFCON talk for the referenced paper https://www.youtube.com/live/pleLhJRW9FwThe risks in LLM powered systems seems like an opened pandora's box the more I look into mitigating it.
1 comment of 2
[ 4.8 ms ] story [ 15.0 ms ] threadThe risks in LLM powered systems seems like an opened pandora's box the more I look into mitigating it.