arXiv:2604.15579v2 Announce Type: replace-cross Abstract: There is increasing interest in integrating AI agents that invoke tools into domain-specific commercial software, where unintended tool calls can cause serious security and safety incidents. This has drawn growing research attention, and many agent security and safety benchmarks have emerged. They implicitly shape how the community approaches security and safety. Yet existing work exhibits a blind spot: it emphasizes training-based methods and neural guardrails, which reduce the likelihood of insecure or unsafe actions but cannot guaran
Source: arXiv cs.AI — read the full report at the original publisher.