PAT-0290
Divergent Outputs
A structural condition where parallel evaluations under comparable scope and shared authority produce non-equivalent outputs.
Primary Lenses
Convergence Lens
Compares parallel structural systems to determine whether they align under shared authority.
Lens Application
The Convergence Lens is a strong inspection mechanism for Divergent Outputs because it compares parallel outputs across comparable scope, authority, criteria, and process conditions. It helps surface where outputs diverge, whether divergence follows a structural pattern, and which shared-authority assumptions require closer inspection.
Inspect For
- Parallel evaluations producing materially different outputs
- Shared authority or criteria applied inconsistently
- Comparable scopes yielding non-equivalent conclusions
- Process or interpretation gaps across evaluation paths
Avoid
Treat output mismatch as an inspection signal, not conclusive evidence of Divergent Outputs, until scope, authority, criteria, and process conditions are checked.
Variance / Entropy Lens
Measures structural variability across repeated or comparable evaluations and identifies divergence beyond expected bounds.
Lens Application
The Variance / Entropy Lens is a strong inspection mechanism for Divergent Outputs because it examines whether comparable evaluations are producing unusually inconsistent outputs. It supports inspection of output spread, decision-path instability, and whether divergence appears tied to scope, authority, criteria, or execution differences.
Inspect For
- Repeated evaluations producing materially different outputs
- Output variance under shared scope or authority
- Inconsistent weighting, criteria, or interpretation
- Divergence that exceeds expected tolerance
Avoid
Treat high variance as an inspection signal, not conclusive evidence of Divergent Outputs, until expected tolerance, bounds, and contextual justification are checked.
Secondary Lenses
Determinism Lens
Evaluates whether identical structural inputs produce equivalent structural outputs across repeated executions.
Lens Application
The Determinism Lens applies indirectly to Divergent Outputs by helping inspect whether output differences arise despite equivalent inputs, repeated execution conditions, or comparable evaluation authority. It highlights instability across runs, but should not be treated as sufficient on its own to establish the condition.
Inspect For
- Repeated evaluations with equivalent structural inputs
- Output differences under comparable scope
- Shared authority producing inconsistent results
- Execution conditions that appear materially unchanged
Avoid
Treat nondeterministic variation as an inspection signal, not confirmation of Divergent Outputs, until scope, authority, and structural comparability are checked.
Compression Lens
Reduces structural graphs into stable minimal representations for comparison, redundancy detection, and diffing.
Lens Application
The Compression Lens applies indirectly to Divergent Outputs by making parallel outputs easier to compare after structural reduction. It may reveal differences in retained nodes, collapsed equivalences, omitted branches, or redundant paths that contribute to non-equivalent results.
Inspect For
- Minimal representations that differ across parallel evaluations
- Collapsed structures that hide or reveal output divergence
- Redundant paths retained in one output but removed in another
- Scope or authority markers preserved unevenly
Avoid
Treat compression results as comparison aids, not sufficient evidence that Divergent Outputs is present.
Primary Issue Matches
Supporting Issue Matches
One Prompt Carries Too Many Meanings
A single prompt carries too many meanings, goals, roles, constraints, or implied tasks for the AI to interpret consistently.
Output Breaks After Model Change
Output that previously worked begins failing after a model, mode, runtime, or product behavior changes.
Parallel Reviews Never Agree
Parallel AI, human, workflow, or tool reviews keep producing different results without resolving into a shared decision state.
Prompt Behavior Changed Without Version Change
A prompt begins producing different behavior even though no prompt version, model version, workflow version, or declared dependency change is recorded.
Prompt Has Too Many Valid Interpretations
The prompt allows too many reasonable interpretations, causing the AI to choose among valid paths without enough guidance.
Prompt Only Works After Retry
The prompt fails, misroutes, or produces an unusable response on one attempt but works after retry without a meaningful change to the input.
Similar Cases Route to Different Outcomes
Similar inputs, cases, prompts, or workflow states are routed to different outcomes without a declared difference that explains the split.
Validation Result Changes on Retry
A validation, grading, review, classification, or pass/fail result changes after retry even though the input and declared validation rules did not change.