MITRE ATLAS 2026 Top AI Detection: LLM System Prompt Extraction Attempt [AML.T00
Detects attempts by users to extract the underlying system prompt from an LLM via prompt injection techniques or identifies successful leaks where the model's response matches a pre-calculated system prompt hash. Exposing system prompts can lead to the discovery of proprietary business logic, safety guidelines, and facilitate further adversarial jailbreaking.
Sigma
