Contestable Agents for Data Governance: Recording Human Stewardship of AI-Generated Metadata Across Structured and Unstructured Systems

Posted on Oct 24 • Originally published at mech.app

Organizations deploying LLMs to classify operational data face a governance gap. Agents tag telemetry points, link warranty claims to service bulletins, and propose metadata for unstructured documents. When a human overrides an agent decision, most systems discard the agent's reasoning and store only the final value. You lose the audit trail, the correction context, and the signal needed to prevent classification drift.

A new ArXiv paper (2610.31553v1) introduces contestable governance: a pattern where agents produce scored, cited arguments for metadata decisions, stewards can accept or modify those arguments before commit, and a governance record retains the evidence, interventions, and outcome. The plumbing treats human corrections as first-class events, not silent overrides.

Industrial systems hold operational knowledge in two forms:

Governance programs manage these separately. Linking a telemetry point to its manual section, classifying a warranty failure mode, or tagging an HVAC zone requires domain specialists. Agents can propose classifications, but unreviewed proposals spread errors. A perfunctory approval leaves no record of what the reviewer actually checked.

The failure mode: agent-generated metadata drifts as equipment changes, terminology evolves, or edge cases accumulate. You discover the drift when downstream analytics break or compliance audits fail.

The pattern introduces three primitives:

Agent Proposal → Triage → [High Confidence → Auto-Commit] → [Uncertain → Steward Review → Governance Record] Enter fullscreen mode Exit fullscreen mode High-confidence proposals (score above threshold) commit automatically but still generate a governance record. Uncertain proposals enter a review queue. The steward sees:

@dataclass class GovernanceRecord: record_id: str timestamp: datetime target_entity: str # telemetry point ID, claim number, document hash agent_proposals: List [ ArgumentProposal ] steward_decision: StewardDecision final_metadata: Dict [ str, Any ] precedent_flag: bool version: int @dataclass class ArgumentProposal: agent_id: str proposed_value: Any confidence_score: float citations: List [ Citation ] reasoning: str @dataclass class StewardDecision: steward_id: str action: Literal [ " accept ", " reject ", " modify ", " defer " ] justification: str added_citations: List [ Citation ] Enter fullscreen mode Exit fullscreen mode The record lives in a separate governance store, not inline with the operational data. You query it for audit trails, drift detection, and agent retraining.

Agents reference a versioned glossary that supplies:

When the glossary updates, you can replay governance records to detect which decisions would change under new definitions. This surfaces semantic drift before it propagates.

The paper presents a reference implementation for classifying HVAC telemetry points using Brick Schema labels.

Input: A telemetry point with raw metadata:

{ "point_id": "AHU-3-SAT", "description": "Supply air temp sensor zone 2", "unit": "degF", "equipment_parent": "AHU-3" } Enter fullscreen mode Exit fullscreen mode Agent task: Propose a Brick Schema class (e.g., brick:Supply_Air_Temperature_Sensor ).

{ "point_id": "RTU-7-OAT", "description": "Outside temp", "unit": "degC", "equipment_parent": "RTU-7" } Enter fullscreen mode Exit fullscreen mode Agent proposes brick:Outside_Air_Temperature_Sensor with 0.68 confidence (ambiguous: could be brick:Outdoor_Air_Temperature_Sensor ). Steward reviews, confirms the correct class based on equipment manual, and adds a citation. The governance record captures both the agent's uncertainty and the steward's resolution.

The triage logic routes proposals based on confidence score and similarity to prior decisions:

The precedent gate controls which verified decisions become training examples. Only decisions where:

This prevents auto-committed high-confidence decisions from polluting the precedent store with obvious cases.

You need metrics to detect when agent metadata quality degrades:

-- Detect classification drift by equipment type SELECT equipment_type, COUNT ( * ) FILTER ( WHERE action = 'reject' ) AS rejections, COUNT ( * ) AS total_proposals, AVG ( confidence_score ) AS mean_confidence FROM governance_records WHERE timestamp > NOW () - INTERVAL '30 days' GROUP BY equipment_type HAVING COUNT ( * ) FILTER ( WHERE action = 'reject' ) / COUNT ( * ) > 0. 15; Enter fullscreen mode Exit fullscreen mode Warranty Failure-Mode Decision Example The paper's second use case links warranty claims to service bulletins and telemetry patterns.

{ "claim_id": "WC-2024-8832", "failure_description": "Compressor short cycling", "equipment_serial": "RTU-2024-4471", "claim_date": "2024-09-15" } Enter fullscreen mode Exit fullscreen mode Agent tasks:

The governance record now links the claim, the telemetry evidence, and the correct bulletin version. When SB-2024-02 is revised again, you can query governance records to find all claims that referenced it.

Retraining trigger: When override rate exceeds threshold, you can:

The key: you have structured data (governance records) instead of unstructured feedback (emails, tickets, manual corrections in spreadsheets).

The paper recommends separate governance store with periodic metadata snapshots. You keep the operational database clean and query the governance store only for audits and retraining.

Citation integrity: All citations include document hashes or version identifiers. If a cited document changes, the governance record flags the citation as stale.

Steward identity: Each governance record logs the steward's identity and timestamp. You can audit who approved which decisions and detect patterns (e.g., one steward consistently overriding a specific agent).

Steward fatigue: If too many proposals land in the review queue, stewards rubber-stamp decisions. Mitigation: tune confidence thresholds, improve agent prompts, surface precedent matches to reduce cognitive load.

Glossary staleness: Agents cite outdated terms, stewards correct them, but glossary never updates. Mitigation: automated glossary update proposals when override rate for specific terms exceeds threshold.

Citation rot: Linked documents move, get revised, or deleted. Mitigation: periodic citation validation job, flag stale citations in governance records.

Precedent pollution: Auto-committed decisions leak into precedent store. Mitigation: strict precedent gate (only steward-reviewed, initially-uncertain decisions qualify).

Conflicting precedents: Two stewards resolve similar cases differently. Mitigation: precedent conflict detection, escalation to domain lead for reconciliation.

The pattern shines in regulated industries (energy, manufacturing, healthcare) where metadata errors propagate into compliance failures. It adds complexity: you maintain a separate governance store, build a steward UI, and tune confidence thresholds. The payoff is structured feedback that improves agent performance and prevents silent drift.

If you are already logging agent decisions, contestable governance is the next step: capture not just the outcome but the reasoning, the human intervention, and the evidence. You turn corrections into training data instead of lost context.