Techniqueresource-developmentATLAS

AML.T0060Publish Hallucinated Entities

What it is

Adversaries may create an entity they control, such as a software package, website, or email address to a source hallucinated by an LLM. The hallucinations may take the form of package names commands, URLs, company names, or email addresses that point the victim to the entity controlled by the adversary. When the victim interacts with the adversary-controlled entity, the attack can proceed.

References

  1. https://atlas.mitre.org/techniques/AML.T0060

Related by meaning· 6

Nearest entities by semantic similarity across the cs-graph corpus.

ATLAS
Discover LLM Hallucinations
ATLAS
Publish Poisoned Models
ATLAS
LLM Trusted Output Components Manipulation
ATLAS
Publish Poisoned Datasets
ATLAS
LLM Prompt Crafting
ATLAS
LLM Data Leakage
Sourced from MITRE ATLAS — Adversarial Threat Landscape for AI Systems. Curated by Adam Lundqvist, SQUR.