Standards
MITRE ATLAS AML.T0080 — AI Agent Context Poisoning (Memory AML.T0080.000)
Adversarial-technique catalog entry naming memory/context poisoning; the taxonomy counterpart to OWASP ASI06.
Specification
What it is
Adversarial-technique catalog entry naming memory/context poisoning; the taxonomy counterpart to OWASP ASI06.
At a glance
- Business model
- Specification
- Governing body
- MITRE (ATLAS)
- Maturity
- live technique in ATLAS data release v2026.09
- What it says about memory
- Sub-technique AML.T0080.000 covers Memory specifically: changes that persist across future chat sessions; AML.T0080.001 covers Thread.
Links and sources
- Sourceatlas.mitre.org
- Sourcebrandefense.io/knowledge-hub/agentic-security/memory-and-context-poisoning-in-ai-agents
Last checked nov 15, 2026; last updated nov 15, 2026; facts verified nov 15, 2026. Vendor facts come from the vendor’s own pages; benchmark claims are self-reported unless we say otherwise.