The human-override position for autonomous agent systems — the agent that puts humans back in control.
When autonomous agents must be stopped, redirected, or corrected, this is the agent-layer construct that enables human authority to reassert itself.
Coordinated sets this position belongs to — the coverage it extends. Counts are the live cluster size in the graph.
Architectural context
Override · Cross-Vertical · 2 compound moats. Architectural surface: Agent. Cross-cutting: Override.
Layer position: Cross-cutting
Why this is canonical
Human override of autonomous AI systems is one of the central problems in AI safety and deployment: how do you safely and reliably interrupt, redirect, or override an agent acting autonomously? 'Override agent' names this capability as a first-class design primitive — an agent whose purpose is to mediate between human intent and autonomous action.
Where it fits
A few directions this coordinate opens —
Illustrative, not exhaustive — held as a transferable canonical position, open to the buyer's own use.