Injection
Prompt Injection
Attack technique where user inputs are manipulated to bypass security filters, content controls, and model behavioral restrictions (also known as Jailbreaking).
Xiaogeng Liu, Zhiyuan Yu, Yizhe Zhang, Ning Zhang, Chaowei Xiao
Mitigation Strategy
Implement robust input validation, explicit separation between system instructions and user data, and apply defensive Prompt Engineering techniques.
Atomic Number
1
Pi
Risk ID
h-01
Severity
8/10
Severity Level