Oversight-Centered Metrology and Control for Agentic Systems: Costly Interrupt Channels, Claim Margins, and Deployment-Relevant Evaluation

Abstract

This preprint develops an oversight-centered metrology and control theory for agentic systems in real workflows, yielding deployment-relevant evaluation criteria that treat human review, automated checks, delayed labels, and external auditing as costly interrupt channels rather than privileged oracles. It formalizes workflow-level estimands, claim-justification margins, transport mismatch, congestion, routing error, audit gaming, redundancy, and safe control under delay, irreversibility, and ambiguity budgets.

Keywords

• agentic systems evaluation
• oversight-centered metrology
• costly interrupt channels
• deployment-relevant evaluation
• human-AI oversight
• workflow-level estimands
• claim margins
• review congestion
• audit gaming
• safe control
• transportability
• post-deployment monitoring

Identifiers and source records

• DOI
• Works entry
• ORCID