TechniquepersistenceATLAS

AML.T0061LLM Prompt Self-Replication

What it is

An adversary may use a carefully crafted [LLM Prompt Injection](/techniques/AML.T0051) designed to cause the LLM to replicate the prompt as part of its output. This allows the prompt to propagate to other LLMs and persist on the system. The self-replicating prompt is typically paired with other malicious instructions (ex: [LLM Jailbreak](/techniques/AML.T0054), [LLM Data Leakage](/techniques/AML.T0057)).

References

  1. https://atlas.mitre.org/techniques/AML.T0061

Related by meaning· 6

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

ATLAS
LLM Prompt Injection
ATLAS
LLM Prompt Crafting
ATLAS
Extract LLM System Prompt
ATLAS
LLM Data Leakage
ATLAS
LLM Prompt Obfuscation
ATLAS
Prompt Infiltration via Public-Facing Application
Sourced from MITRE ATLAS — Adversarial Threat Landscape for AI Systems. Curated by Adam Lundqvist, SQUR.