AML.CS0022
ChatGPT Package Hallucination
- Type
- Exercise
- When
- 2024-06
- Actor, as ATLAS names it
- Not stated
- Target
- ChatGPT users
- ATLAS attributes it to Tenable (security evaluation).
- ATLAS attributes it to Lasso Security (security evaluation).
- On ShieldMap: ChatGPT (matched by name in the target).
Techniques, in order
- 1AI Model Inference API AccessAI Model Access · AML.T0040
- 2Discover LLM HallucinationsDiscovery · AML.T0062
- 3Publish Hallucinated EntitiesResource Development · AML.T0060
- 4Supply Chain Compromise: SoftwareInitial Access · AML.T0010.001
- 5User Execution: Malicious PackageExecution · AML.T0011.001
- 6External Harms: User HarmImpact · AML.T0048.003
How these attacks work
References
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.