AVID-2026-R0434

Description

Microsoft Azure Prompt Shield Guardrail Evasion

Details

Azure Prompt Shield is a jailbreak and prompt injection classifier created by Microsoft. Prompt Shield can be enabled by developers using Azure OpenAI to protect their AI applications from these threats. We successfully demonstrated how an attacker can fully evade, or greatly degrade, classification accuracy of the classifier, enabling prompt injections and jailbreaks to pass through filters and subsequently to the protected AI application.

References

Mindgard Disclosure
Bypassing Azure AI Content Safety Guardrails

Affected or Relevant Artifacts

Developer: Microsoft
Deployer:
Artifact Details:

Type	Name
System	Azure Prompt Shield

Impact

(none)

Other information

Report Type: Advisory
Credits: William Hackett Lewis Birch, Mindgard
Date Reported: 2024-06-06
Version: 0.3.1
AVID Entry