# LEN-0290 — Variance / Entropy Lens

## Ontology Metadata

Code: LEN-0290
Version: LEN-0290@0.1.0
Ontology release: 0.1.0
Updated: 2026-05-10T00:00:00Z
Canonical URL: /workbench/lenses/variance-entropy-lens/

## Summary

Measures structural variability across repeated or comparable evaluations and identifies divergence beyond expected bounds.

## Primary Pattern Matches

- PAT-0110 — Asymmetric Structure
  - URL: /workbench/patterns/asymmetric-structure/
  - Summary: A structural condition where comparable regions receive unequal rule, constraint, or authority application without a declared differentiation rule.
- PAT-0270 — Density Spike
  - URL: /workbench/patterns/density-spike/
  - Summary: A structural condition where nodes, edges, dependencies, decisions, or effects concentrate sharply within a localized region beyond declared thresholds.
- PAT-0290 — Divergent Outputs
  - URL: /workbench/patterns/divergent-outputs/
  - Summary: A structural condition where parallel evaluations under comparable scope and shared authority produce non-equivalent outputs.
- PAT-0300 — Escalation Growth
  - URL: /workbench/patterns/escalation-growth/
  - Summary: A structural condition where impact, authority, scope, or consequence increases across sequential states without a declared limiting mechanism.
- PAT-0210 — Non-Deterministic Execution
  - URL: /workbench/patterns/non-deterministic-execution/
  - Summary: A structural condition where equivalent inputs and declared constraints produce divergent outputs across executions.

## Secondary Pattern Matches

- PAT-0170 — Constraints Underspecified
  - URL: /workbench/patterns/constraints-underspecified/
  - Summary: A structural condition where declared constraints are insufficient to eliminate ambiguity or multiple admissible states.
- PAT-0320 — Convergence Failure
  - URL: /workbench/patterns/convergence-failure/
  - Summary: A structural condition where sequential or parallel states fail to resolve into an equivalent or coherent structure under shared authority and constraints.
- PAT-0360 — Propagation Amplification
  - URL: /workbench/patterns/propagation-amplification/
  - Summary: A structural condition where an effect, declaration, authority, constraint, or state increases in scope or intensity as it propagates beyond declared bounds.
- PAT-0410 — Unconstrained Expansion
  - URL: /workbench/patterns/unconstrained-expansion/
  - Summary: A structural condition where a region, process, authority, or effect expands without governing constraints limiting growth.

## Related Issues

- ISS-0046 — Action Changed Something Else Too
  - URL: /workbench/issues/action-changed-something-else-too/
  - Summary: An AI or agent action makes the requested change but also changes another object, field, file, state, rule, or workflow element that was not supposed to change.
- ISS-0064 — Action Triggered by Confidence Score
  - URL: /workbench/issues/action-triggered-by-confidence-score/
  - Summary: A confidence score, certainty label, risk level, or probability-like value triggers an action without enough approval, calibration, or authority control.
- ISS-0008 — Agent Keeps Expanding the Task
  - URL: /workbench/issues/agent-keeps-expanding-the-task/
  - Summary: The agent repeatedly expands the task, plan, scope, or next-step list instead of completing the declared work.
- ISS-0082 — Agent Never Settles on Final Answer
  - URL: /workbench/issues/agent-never-settles-on-final-answer/
  - Summary: The agent keeps revising, rechecking, planning, or branching instead of converging on a final answer or completed result.
- ISS-0099 — Agent Permission Expands Over Steps
  - URL: /workbench/issues/agent-permission-expands-over-steps/
  - Summary: An agent begins with limited permission but gains, assumes, or exercises broader authority as the workflow continues.
- ISS-0029 — AI Forgets Earlier Constraints
  - URL: /workbench/issues/ai-forgets-earlier-constraints/
  - Summary: A constraint, instruction, preference, or decision that should persist through the task stops affecting later output.
- ISS-0010 — Answer Has Too Many Paths
  - URL: /workbench/issues/answer-has-too-many-paths/
  - Summary: The answer presents too many possible paths, interpretations, options, or next steps without enough structure to choose among them.
- ISS-0032 — Context Leaks Between Tasks
  - URL: /workbench/issues/context-leaks-between-tasks/
  - Summary: Context, assumptions, constraints, examples, files, or decisions from one task affect another task where they should not apply.
- ISS-0077 — Diagnostic Area Has No Coverage
  - URL: /workbench/issues/diagnostic-area-has-no-coverage/
  - Summary: A known diagnostic area, failure mode, requirement, or review dimension has no Issue, check, rubric item, or workflow coverage.
- ISS-0097 — Downstream Steps Magnify Hallucinated Claim
  - URL: /workbench/issues/downstream-steps-magnify-hallucinated-claim/
  - Summary: A hallucinated or unsupported claim from an AI output is reused by later workflow steps until it becomes more influential than the evidence supports.
- ISS-0027 — Duplicate Fields With Same Meaning
  - URL: /workbench/issues/duplicate-fields-with-same-meaning/
  - Summary: The AI returns multiple fields, labels, sections, or structured elements that carry the same meaning and create ambiguity about which one should be used.
- ISS-0066 — Duplicate Output Sections
  - URL: /workbench/issues/duplicate-output-sections/
  - Summary: The AI repeats sections, headings, blocks, or output areas in a way that creates redundancy, confusion, or downstream handling problems.
- ISS-0094 — Early Model Output Gets Overweighted Downstream
  - URL: /workbench/issues/early-model-output-gets-overweighted-downstream/
  - Summary: An early AI output receives too much authority in later workflow steps, decisions, reviews, or generated artifacts.
- ISS-0081 — Evaluation Rubric Has Coverage Gap
  - URL: /workbench/issues/evaluation-rubric-has-coverage-gap/
  - Summary: An evaluation rubric, grading standard, or review checklist leaves part of the required evaluation space uncovered.
- ISS-0028 — Format Rule Too Weak
  - URL: /workbench/issues/format-rule-too-weak/
  - Summary: The format instruction is too vague, incomplete, or optional to reliably produce output that satisfies the expected structure.
- ISS-0036 — Human Review and Automation Disagree
  - URL: /workbench/issues/human-review-and-automation-disagree/
  - Summary: A human review result and an automated AI or workflow result disagree without a declared rule for resolving the difference.
- ISS-0021 — Invalid JSON Output
  - URL: /workbench/issues/invalid-json-output/
  - Summary: The AI returns malformed JSON or structured output that cannot be parsed.
- ISS-0101 — Local Exception Grows Into Policy
  - URL: /workbench/issues/local-exception-grows-into-policy/
  - Summary: A local exception, special case, or one-off allowance begins to function like a general policy.
- ISS-0096 — Local Rule Spreads to Broader Cases
  - URL: /workbench/issues/local-rule-spreads-to-broader-cases/
  - Summary: A rule intended for one local case, file, context, user, workflow, or exception begins affecting broader cases.
- ISS-0084 — Merge Step Leaves Unresolved Differences
  - URL: /workbench/issues/merge-step-leaves-unresolved-differences/
  - Summary: A merge, reconciliation, or consolidation step combines outputs or reviews but leaves important differences unresolved.
- ISS-0074 — Missing Fallback for Unavailable Information
  - URL: /workbench/issues/missing-fallback-for-unavailable-information/
  - Summary: The task does not declare what the AI should do when required information, sources, tools, fields, or evidence are unavailable.
- ISS-0050 — Model Output Triggers Unapproved Action
  - URL: /workbench/issues/model-output-triggers-unapproved-action/
  - Summary: AI output causes, recommends, or triggers an action that has not passed the required approval, permission, or authority check.
- ISS-0070 — Multiple Policies Say the Same Thing
  - URL: /workbench/issues/multiple-policies-say-the-same-thing/
  - Summary: Multiple policies, rules, or guidance documents express the same requirement, creating redundancy and uncertainty about which one governs.
- ISS-0104 — One Prompt Carries Too Many Meanings
  - URL: /workbench/issues/one-prompt-carries-too-many-meanings/
  - Summary: A single prompt carries too many meanings, goals, roles, constraints, or implied tasks for the AI to interpret consistently.
- ISS-0052 — Output Breaks After Model Change
  - URL: /workbench/issues/output-breaks-after-model-change/
  - Summary: Output that previously worked begins failing after a model, mode, runtime, or product behavior changes.
- ISS-0060 — Output Exceeds Length Limit
  - URL: /workbench/issues/output-exceeds-length-limit/
  - Summary: The AI output exceeds a declared length, token, word, character, section, field, or size limit.
- ISS-0083 — Parallel Reviews Never Agree
  - URL: /workbench/issues/parallel-reviews-never-agree/
  - Summary: Parallel AI, human, workflow, or tool reviews keep producing different results without resolving into a shared decision state.
- ISS-0049 — Policy Exception Spreads Too Far
  - URL: /workbench/issues/policy-exception-spreads-too-far/
  - Summary: A narrow policy exception, allowance, or special case spreads beyond its intended scope and begins governing broader cases.
- ISS-0058 — Prompt Behavior Changed Without Version Change
  - URL: /workbench/issues/prompt-behavior-changed-without-version-change/
  - Summary: A prompt begins producing different behavior even though no prompt version, model version, workflow version, or declared dependency change is recorded.
- ISS-0076 — Prompt Does Not Say What to Exclude
  - URL: /workbench/issues/prompt-does-not-say-what-to-exclude/
  - Summary: The prompt declares what to include but does not declare what should be excluded, allowing unwanted scope, sources, content, or actions into the result.
- ISS-0071 — Prompt Has Too Many Valid Interpretations
  - URL: /workbench/issues/prompt-has-too-many-valid-interpretations/
  - Summary: The prompt allows too many reasonable interpretations, causing the AI to choose among valid paths without enough guidance.
- ISS-0005 — Prompt Only Works After Retry
  - URL: /workbench/issues/prompt-only-works-after-retry/
  - Summary: The prompt fails, misroutes, or produces an unusable response on one attempt but works after retry without a meaningful change to the input.
- ISS-0047 — Repair Step Created New Breakage
  - URL: /workbench/issues/repair-step-created-new-breakage/
  - Summary: A repair, correction, retry, or fix step addresses one problem but introduces a new failure elsewhere.
- ISS-0068 — Repeated Constraints Create Confusion
  - URL: /workbench/issues/repeated-constraints-create-confusion/
  - Summary: Repeated constraints, instructions, limits, or exclusions make the task harder to interpret instead of clearer.
- ISS-0063 — Results Vary Too Much
  - URL: /workbench/issues/results-vary-too-much/
  - Summary: Repeated or comparable runs produce outputs that vary more than the task, workflow, or user can tolerate.
- ISS-0062 — Retrieval Exceeds Evidence Limit
  - URL: /workbench/issues/retrieval-exceeds-evidence-limit/
  - Summary: The AI retrieves, uses, cites, or considers more evidence than the task permits or more than the review surface can support.
- ISS-0007 — Retry Makes the Problem Worse
  - URL: /workbench/issues/retry-makes-the-problem-worse/
  - Summary: A retry, repair attempt, regeneration, or follow-up instruction increases the error, expands the failure, or creates additional breakage instead of narrowing the problem.
- ISS-0100 — Review Escalates Without Stop Condition
  - URL: /workbench/issues/review-escalates-without-stop-condition/
  - Summary: A review process keeps escalating, re-reviewing, or adding scrutiny without a declared condition for stopping.
- ISS-0051 — Review Outcome Changes Unrelated Environment
  - URL: /workbench/issues/review-outcome-changes-unrelated-environment/
  - Summary: A review result, approval, rejection, or classification changes state outside the environment, case, file, or workflow it was meant to govern.
- ISS-0105 — Review Queue Becomes Bottleneck
  - URL: /workbench/issues/review-queue-becomes-bottleneck/
  - Summary: A review queue, approval path, or validation stage accumulates too much work and begins blocking the workflow.
- ISS-0075 — Review Rubric Missing Required Criteria
  - URL: /workbench/issues/review-rubric-missing-required-criteria/
  - Summary: A review rubric, grading rule, evaluation checklist, or classification standard lacks criteria required to make the review reliable.
- ISS-0065 — Risk Score Triggers Wrong Escalation
  - URL: /workbench/issues/risk-score-triggers-wrong-escalation/
  - Summary: A risk score, severity label, confidence value, or threshold result triggers the wrong escalation path.
- ISS-0095 — Risk Signal Escalates Beyond Evidence
  - URL: /workbench/issues/risk-signal-escalates-beyond-evidence/
  - Summary: A risk signal, warning, score, or concern escalates farther than the available evidence supports.
- ISS-0092 — Routing Path Cycles Back to Start
  - URL: /workbench/issues/routing-path-cycles-back-to-start/
  - Summary: A routing path sends the case back to the starting point or an earlier step without resolving the condition that caused the route.
- ISS-0072 — Same Instructions Allow Different Outputs
  - URL: /workbench/issues/same-instructions-allow-different-outputs/
  - Summary: The same instructions are broad or underspecified enough to allow materially different outputs that all appear compliant.
- ISS-0067 — Same Rule Declared in Multiple Places
  - URL: /workbench/issues/same-rule-declared-in-multiple-places/
  - Summary: The same rule, constraint, instruction, or policy appears in multiple places, creating redundancy and possible drift.
- ISS-0069 — Same Workflow Check Happens Twice
  - URL: /workbench/issues/same-workflow-check-happens-twice/
  - Summary: The same review, validation, approval, routing, or safety check occurs more than once in the workflow without a clear reason.
- ISS-0102 — Severity Increases Without New Evidence
  - URL: /workbench/issues/severity-increases-without-new-evidence/
  - Summary: The severity, risk, confidence, or escalation level increases even though no new evidence has been added.
- ISS-0001 — Similar Cases Route to Different Outcomes
  - URL: /workbench/issues/similar-cases-route-to-different-outcomes/
  - Summary: Similar inputs, cases, prompts, or workflow states are routed to different outcomes without a declared difference that explains the split.
- ISS-0107 — Single Field Carries Too Many Obligations
  - URL: /workbench/issues/single-field-carries-too-many-obligations/
  - Summary: One field, label, score, status, or structured value is expected to carry too many meanings, decisions, or workflow obligations.
- ISS-0103 — Single Step Carries Too Many Decisions
  - URL: /workbench/issues/single-step-carries-too-many-decisions/
  - Summary: One prompt, workflow step, review stage, or agent action carries too many decisions for the system or user to evaluate cleanly.
- ISS-0106 — Small Change Produces Large Downstream Effects
  - URL: /workbench/issues/small-change-produces-large-downstream-effects/
  - Summary: A small prompt, schema, policy, output, or workflow change creates unexpectedly large effects in downstream steps.
- ISS-0093 — Small Error Spreads Into Large Failure
  - URL: /workbench/issues/small-error-spreads-into-large-failure/
  - Summary: A small AI, output, routing, or workflow error propagates through later steps until it becomes a larger failure.
- ISS-0098 — Small Issue Keeps Escalating
  - URL: /workbench/issues/small-issue-keeps-escalating/
  - Summary: A small issue, warning, uncertainty, or correction keeps increasing in severity, scope, or workflow impact across later steps.
- ISS-0011 — Task Has No Clear Limit
  - URL: /workbench/issues/task-has-no-clear-limit/
  - Summary: The task does not declare where the AI should stop, what is out of scope, or what counts as enough work.
- ISS-0030 — Task Progress Is Lost Midway
  - URL: /workbench/issues/task-progress-is-lost-midway/
  - Summary: The AI loses track of completed work, prior decisions, current position, or remaining steps before the task is finished.
- ISS-0061 — Too Many Tool Calls
  - URL: /workbench/issues/too-many-tool-calls/
  - Summary: The AI or agent makes more tool calls, searches, retrievals, API calls, or integration actions than the task requires or permits.
- ISS-0020 — Tool Exists but Required Inputs Are Missing
  - URL: /workbench/issues/tool-exists-but-required-inputs-are-missing/
  - Summary: A usable tool or integration exists, but the AI or agent does not have the required inputs, permissions, fields, identifiers, or context needed to call it correctly.
- ISS-0006 — Validation Result Changes on Retry
  - URL: /workbench/issues/validation-result-changes-on-retry/
  - Summary: A validation, grading, review, classification, or pass/fail result changes after retry even though the input and declared validation rules did not change.
- ISS-0086 — Workflow Loops Through Review Without Resolution
  - URL: /workbench/issues/workflow-loops-through-review-without-resolution/
  - Summary: A workflow repeatedly sends work through review, repair, or escalation without reaching an approved, rejected, or otherwise resolved state.
- ISS-0073 — Workflow Step Lacks Required Conditions
  - URL: /workbench/issues/workflow-step-lacks-required-conditions/
  - Summary: A workflow step can run, route, approve, reject, or continue without the required conditions being declared or checked.
- ISS-0089 — Workflow Waits on Step That Waits Back
  - URL: /workbench/issues/workflow-waits-on-step-that-waits-back/
  - Summary: A workflow step waits for another step that also waits on the first step, creating a blocking loop.