UTC

AML.T0112.000

Machine Compromise: Local AI Agent

Adversaries may achieve full system compromise by abusing AI agents running locally on a host, such as computer-use agents or AI-driven browsers.

  • Tactic Impact
  • Maturity: Demonstrated
  • Agentic AI
Full entry on MITRE ATLAS →
Mitigations
  • Generative AI GuardrailsAML.M0020

    Guardrails are safety controls placed between users, tools, and generative AI models to evaluate prompts, retrieved context, model outputs, and agent actions before they are accepted, executed, or shown to a user.

    Technical - AI

  • AI Telemetry Logging and MonitoringAML.M0024

    Implement logging and monitoring across the AI lifecycle to detect suspicious activity and support investigation. Capture security-relevant access, changes, and operations across data pipelines, model development, deployment, and connected systems.

    Technical - Cyber

From MITRE ATLAS 2026.10, Copyright 2021-2026 MITRE, Apache License 2.0.