Poster · Sydney
Position: A Safe LLM and a Safe Harness Do Not Make a Safe Agent
- When
- Tue 8 Dec 2026, 10:00–13:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 1
- Authors
- Vincent Siu, Kyle Montgomery, Yujin Potter, Zhun Wang, Dawn Song, Chenguang Wang
- Organisations
- University of California, Santa Cruz, University of California, Berkeley
- QRS classification
- Agentic AI, Language models, Safety, fairness and privacy