TwinBreak: Jailbreaking LLM Security Alignments based on Twin Prompts

Name: TwinBreak: Jailbreaking LLM Security Alignments based on Twin Prompts
Duration: 15 min
Description: Talk from 34th USENIX Security Symposium (USENIX Security '25).

Torsten Krauß

34th USENIX Security Symposium (USENIX Security '25) · Day 2 · LLM Security 2: Jailbreaking and Prompt Stealing