Two modes an AI agent can operate in. Autonomous agents act without approval inside their authority boundary. Semi-autonomous agents recommend and wait for a human to confirm. Most real-world agents are semi-autonomous for high-stakes actions and autonomous for low-stakes ones.
Why it matters
Mismatched mode is the most common cause of agent disasters. Autonomous on a high-stakes action means runaway damage. Semi-autonomous on a low-stakes action means the human becomes the bottleneck.
Related terms
Authority Boundary
A clear line that defines what an AI agent is allowed to do and what it must not do. Includes tool access, decision rights, dol...
Escalation Over Autonomy
A design principle where agents flag and recommend rather than act unilaterally when outside their authority boundary. The defa...
Build with this on OTP
OTP encodes coordination intelligence so AI agent teams can run on it. If this term shows up in your team's playbook, it belongs in your OOS.
Found an issue with this definition? Tell us and we'll fix it.