Single-turn attack
Gray Box
Exploit partial knowledge of model internals (RLHF, reward hacking).
Framework mapping
| OWASP LLM Top 10 | MITRE ATLAS |
|---|---|
| LLM01, LLM05 | AML.T0043 |
Run Gray Box and 33 other techniques in AgenticAssure with continuous monitors, conformity mapping to 12 frameworks, and External Auditor Seats for third-party verification.
AgenticAssure ยท Trust Layer for Enterprise AI