Unverified50% confidenceFactExact time
The indirect attack evaluator specifically detects whether an agent can resist malicious instructions embedded in content from external tools or documents when processing such content.
1
Sources
50%
Confidence
Long-term
Relevance
9/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedA safety probe is a defense mechanism that detects malicious prompts by reading vectors from the model's internal hidden layers, without relying on the model's output.76% similarUnverifiedVICE actively simulates attack surface probing of an application's externally exposed interactive interfaces, with a methodology derived from penetration testing and red team exercises rather than traditional passive compliance checklists.72% similarUnverifiedAgentScope's tool review mechanism classifies tools by risk level, and the system automatically performs security checks when sensitive tools are invoked.72% similarUnverifiedCybersecurity evaluation refers to the process of systematically testing an AI model's capabilities in offensive and defensive cybersecurity scenarios, typically including tasks such as having the model attempt to discover vulnerabilities, write exploit code, and simulate social engineering attacks.70% similarUnverifiedThe paper notes that interpretability techniques can be used for auditing—that is, verifying whether dangerous capabilities have been genuinely removed rather than merely suppressed on the surface.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/925594API
curl https://kongchang.com/api/v1/knowledge/claims/925594MCP
get_claim(id=925594)