Risk · Model & Application Security

Excessive agency in autonomous agents

An agent is granted more permission, tool access, or autonomy than the task requires.

high
agentic-ai
autonomy
owasp-llm

How it happens

An agent is granted broader permissions, tool access, or autonomy than the task requires, so a single bad decision or manipulated instruction can cascade into a high-consequence action.

Why it matters

The blast radius of a failure scales with the agent's permissions, not with the intent behind the failure.

Mitigation

Mitigating controls

The controls that address this risk, ranked by effectiveness.

primary

Human-in-the-loop review

Mandatory human approval gate for high-consequence or irreversible AI-generated actions before they take effect.

supporting

AI system monitoring and logging

Continuous telemetry, drift detection, and audit logging of AI system inputs, outputs, and overrides.

Regulatory context

Framework and clause references

Framework	Clause	Title
OWASP Top 10 for LLM Applications	LLM06	Excessive Agency

Back to the risk library