Research question
How safe are agents across diverse environments and failure modes?
Key findings
- 349 environments and 2,000 test cases across eight risk categories - No tested agent scored above 60% safety
A broad agent safety benchmark from Tsinghua showing all sixteen tested agents fell short across interaction environments.
How safe are agents across diverse environments and failure modes?
Spot an error, or know a record we should add? Members can propose additions and submit corrections.