AML.CS0057
Storm-2139 Azure OpenAI Guardrail Bypass
- Type
- Incident
- When
- 2024-07 to 2024-12
- Actor, as ATLAS names it
- Not stated
- Target
- Microsoft Azure OpenAI Service
- ATLAS attributes it to Storm-2139 (adversarial operation).
Techniques, in order
- 1Develop CapabilitiesResource Development · AML.T0017
- 2Acquire InfrastructureResource Development · AML.T0008
- 3Stage CapabilitiesResource Development · AML.T0079
- 4Valid AccountsInitial Access · AML.T0012
- 5AI Model Inference API AccessAI Model Access · AML.T0040
- 6LLM JailbreakDefense Evasion · AML.T0054
- 7Generate DeepfakesAI Attack Adaptation · AML.T0088
- 8External Harms: Societal HarmImpact · AML.T0048.002
- 9External Harms: Financial HarmImpact · AML.T0048.000
How these attacks work
References
- Taking legal action to protect the public from abusive AI-generated content
- Disrupting a global cybercrime network abusing generative AI
- How Microsoft is taking down AI hackers who create harmful images of celebrities and others
- AIID - Incident 955: Global Cybercrime Network Storm-2139 Allegedly Exploits AI to Generate Deepfake Content
From MITRE ATLAS 2026.10 (Apache-2.0). ShieldMap lists who and what was involved and the techniques used, and links to ATLAS for the write-up; it does not copy the steps. Attribution is ATLAS's, not ShieldPi's.