Skip to main content
COMPEL Body of Knowledge
Get Certified Platform

COMPEL Glossary / human-intervention-rate

Human-intervention rate

An agent SLI measuring the rate at which humans must intervene on agent actions — via escalation, override, or correction.

What this means in practice

Inversely correlates with autonomy and with trust; tracking intervention rate by action class reveals where the agent is weakest.

Synonyms

agent HITL rate , intervention frequency

Related Terms Network

This diagram shows how Human-intervention rate connects to other concepts in the COMPEL body of knowledge. Each linked node is a term you can explore in turn.

Related terms for Human-intervention rate Network graph showing 3 related glossary terms connected to Human-intervention rate. Human-interven… Goal-achie… Escalation… Human-in-t…
3 related concepts connected to Human-intervention rate.

The COMPEL Glossary Graph visualizes relationships between framework terminology, showing how concepts interconnect across domains, stages, and pillars. Term nodes cluster by pillar affiliation while cross-references reveal semantic dependencies — for example, how risk appetite connects to control effectiveness, model governance, and assurance requirements. This network representation helps practitioners navigate the framework vocabulary and understand that COMPEL terminology forms a coherent conceptual system rather than isolated definitions.

Accessible list of related terms
  • Goal-achievement rate — An agent service-level indicator measuring the fraction of tasks that reach their intended outcome without human intervention.
  • Escalation protocol — Deterministic rules that pause the agent and escalate to a human for decision — triggered by low confidence, high-consequence tool calls, out-of-distribution inputs, or explicit agent request.
  • Human-in-the-Loop — Human-in-the-loop (HITL) is an AI system design pattern where a human must actively review, approve, and authorize each AI decision or action before it is executed, providing maximum human oversight at the cost of reduced speed and scalability.

See also

  • Goal-achievement rate — An agent service-level indicator measuring the fraction of tasks that reach their intended outcome without human intervention.
  • Escalation protocol — Deterministic rules that pause the agent and escalate to a human for decision — triggered by low confidence, high-consequence tool calls, out-of-distribution inputs, or explicit agent request.
  • Human-in-the-Loop — Human-in-the-loop (HITL) is an AI system design pattern where a human must actively review, approve, and authorize each AI decision or action before it is executed, providing maximum human oversight at the cost of reduced speed and scalability.