MITRE ATLASmitigation
AML.M0024 AI Telemetry Logging
- Created
- 2025-03-12
Description
Implement logging of inputs and outputs of deployed AI models. When deploying AI agents, implement logging of the intermediate steps of agentic actions and decisions, data access and tool use, installation commands, and identity of the agent. Monitoring logs can help to detect security threats and mitigate impacts.
Additionally, having logging enabled can discourage adversaries who want to remain undetected from utilizing AI resources.
Related
Mitigates · MITRE ATLAS (17)
- AML.T0101 Data Destruction via AI Agent Tool Invocation technique
- AML.T0051.002 Triggered sub-technique
- AML.T0085 Data from AI Services technique
- AML.T0085.000 RAG Databases sub-technique
- AML.T0085.001 AI Agent Tools sub-technique
- AML.T0086 Exfiltration via AI Agent Tool Invocation technique
- AML.T0051 LLM Prompt Injection technique
- AML.T0051.000 Direct sub-technique
- AML.T0051.001 Indirect sub-technique
- AML.T0053 AI Agent Tool Invocation technique
- AML.T0005.001 Train Proxy via Replication sub-technique
- AML.T0024 Exfiltration via AI Inference API technique
- AML.T0024.000 Infer Training Data Membership sub-technique
- AML.T0024.001 Invert AI Model sub-technique
- AML.T0024.002 Extract AI Model sub-technique
- AML.T0040 AI Model Inference API Access technique
- AML.T0047 AI-Enabled Product or Service technique