0. Plain Statement

Feedback without integrity becomes capture.

Plain-language version:

Feedback only helps a system regulate when the feedback channel remains honest, usable, auditable, and connected to real effects. If feedback can be gamed, suppressed, redirected, or turned into performance theater, the system learns the wrong thing.

1. Formal Definition

The Feedback Integrity Law states that feedback only improves coherence when the feedback channel preserves fidelity between field effects, affected-node reality, classification, selection, and corrective action.

Feedback is not automatically corrective. A system can receive many signals and still learn incorrectly if those signals are distorted, suppressed, gamed, filtered, redirected, over-weighted, under-weighted, performative, or routed into the wrong classification pathway.

When feedback integrity fails, selection misfires. The system begins selecting for the performance of responsiveness rather than real correction, or it learns from manipulated signals rather than actual coherence conditions.

In UTS terms, feedback without integrity corrupts Γ, feeds inversion Ξ, and increases hidden debt H.

2. Canonical Form

FI failure → Γ_mis → Ξ → H↑

Expanded canonical form:

when feedback integrity fails, the system misclassifies field effects, generates inversion, and accumulates hidden debt

Failure expression:

feedback channel gamed / suppressed / redirected / performative ⇒ selection misfire

Related variables:

O, H, ε, ι, Au, R, BΣ, K, µᵢ, Φ, Γ, Π, Θ, Ψ, Τ, FI

Where:

3. Core Mechanism

The Feedback Integrity Law unfolds whenever a system uses feedback to classify, adapt, govern, repair, optimize, or steer.

Coherent feedback pathway

field effect occurs
→ affected-node signal can reach the system
→ channel preserves context
→ feedback is auditable
→ classification updates accurately
→ corrective action targets the source
→ hidden debt decreases
→ recurrence weakens

Feedback capture pathway

field effect occurs
→ feedback channel is gamed / suppressed / redirected / performative
→ system misclassifies the state
→ correction targets wrong object
→ visible responsiveness may improve
→ hidden debt accumulates
→ recurrence persists

The core mechanism is:

feedback regulates only when the channel itself resists capture

When the channel is compromised, feedback no longer connects system action to real effects.

4. When This Law Applies

This law applies whenever systems use feedback, complaints, metrics, surveys, appeals, reviews, incident reports, telemetry, comments, ratings, engagement, user signals, biological signals, security alerts, audits, monitoring, elections, public response, training data, performance evaluation, or model evaluation to steer action.

It is especially important when:

• feedback can be gamed;
• complaints disappear into process;
• appeals are performative;
• surveys are designed to produce desired results;
• feedback is punished;
• affected nodes lack safe channels;
• metrics replace field reality;
• alerts are suppressed;
• dashboards become performance theater;
• AI evaluations are benchmark-gamed;
• user feedback is reduced to engagement;
• governance treats participation as legitimacy;
• institutions classify criticism as noncompliance;
• security treats silence as safety;
• feedback is collected but cannot change policy.

The law applies strongly when:

feedback exists but does not reduce recurrence or hidden debt

or when:

the system can perform responsiveness without being corrected

Typical domains:

5. When This Law Does Not Apply

This law should not be used to demand that all feedback be treated as correct.

Feedback can be noisy, adversarial, incomplete, mistaken, emotionally charged, strategically manipulated, or based on partial information. Feedback integrity does not mean obeying every signal. It means preserving a channel capable of distinguishing useful, harmful, misleading, and valid feedback.

Feedback integrity allows filtering, triage, sandboxing, weighting, and review.

This law does not imply:

• every complaint is true;
• every alert is valid;
• every metric should steer action;
• every user preference should be obeyed;
• every signal deserves equal weight;
• every feedback source is high fidelity;
• feedback should bypass boundaries or safety.

False-positive cases:

Important distinction:

Feedback integrity does not mean feedback obedience. It means feedback remains traceable, corrigible, uncaptured, and connected to real effects.

6. Diagnostic Signature

Canonical diagnostic:

FI failure → Γ_mis → Ξ → H↑

Warning signature:

feedback volume↑
field correction↓
recurrence unchanged
channel gameability↑
suppression↑
performance theater↑
H↑
⇒ feedback capture

Common indicators:

Additional diagnostics:

7. Failure Pattern

If ignored, this law produces feedback theater, selection misfire, hidden debt, and legitimacy collapse.

General failure pathway:

feedback enters system
→ channel is gameable / suppressed / redirected / performative
→ system misclassifies feedback
→ selection favors appearances
→ corrective action targets proxy or theater
→ hidden debt accumulates
→ affected nodes lose trust
→ recurrence persists
→ legitimacy shock appears later

Common failure modes:

• Feedback Capture — feedback channel is captured by incentives, power, or performance.
• Selection Misfire — system learns the wrong lesson.
• Feedback Suppression — valid signals are blocked or punished.
• Feedback Redirection — signals are routed away from the corrective layer.
• Feedback Gaming — actors optimize to manipulate feedback outcomes.
• Performance Theater — system performs listening without being changed.
• Misclassification — feedback is classified incorrectly.
• Auditability Collapse — feedback path cannot be reconstructed.
• Pseudo-Coherence — feedback metrics improve while coherence declines.
• Hidden Debt Accumulation — real effects remain unrepaired.
• Control-Restoration Confusion — managing feedback replaces repairing the source.
• Legitimacy Shock — captured feedback is later exposed.

Compact failure signature:

FI↓ + Γ_mis + recurrence↑ ⇒ feedback-driven hidden debt

8. Restoration Implications

Restoration requires repairing the feedback channel before trusting the system’s learning.

The first restoration question is not:

What does the feedback say?

The first restoration question is:

Can this feedback channel carry truth, correction, and affected-node reality without capture?

Restoration priorities:

1. Map the feedback channel.
2. Identify who can speak and who cannot.
3. Identify whether feedback can change system behavior.
4. Audit suppression, redirection, and gaming pathways.
5. Restore traceability of feedback inputs and outcomes.
6. Separate feedback performance from actual correction.
7. Reclassify corrupted feedback as provisional.
8. Reconnect feedback to origin-layer repair.
9. Protect affected-node feedback from retaliation or erasure.
10. Time-validate recurrence reduction.

Relevant restoration arcs:

Minimal restoration sequence:

map feedback channel
→ identify capture / suppression / gaming
→ restore Au and FI
→ protect affected-node signal
→ reclassify feedback proportionally
→ route into origin-layer repair
→ validate recurrence↓ and H↓

Temporal validation requirement:

FI↑
Au↑
Γ_mis↓
feedback changes system behavior
affected-node signal preserved
H↓
recurrence↓
R targets origin
Ξ / ι↓
O stable or rising

9. Design Rule

Never optimize from feedback channels whose integrity cannot be audited.

Operational design requirements:

• Treat feedback integrity as a first-class requirement.
• Preserve traceability from feedback to action.
• Protect feedback from suppression and retaliation.
• Detect and reduce channel gameability.
• Separate feedback collection from feedback correction.
• Ensure feedback can change the system.
• Preserve affected-node feedback.
• Validate feedback through recurrence and field effects.
• Audit whether feedback improves coherence.
• Repair feedback channels before scaling them.

Avoid:

• treating feedback volume as feedback integrity;
• treating surveys as legitimacy;
• treating engagement as meaning;
• treating appeals as repair when outcomes never change;
• treating complaint closure as correction;
• treating benchmark improvement as safety;
• treating silence as satisfaction;
• treating alert suppression as security;
• treating compliance as feedback success;
• treating performance theater as restoration.

10. Cross-Scale Expressions

11. Examples

Example A — AI Evaluation Gaming

Scenario:

An AI model improves benchmark scores, but field failures persist because the evaluation channel is gamed or too narrow.

Law expression:

FI_eval failure → Γ_mis → Ξ → H_AI↑

Interpretation:

Evaluation feedback has been captured by benchmark performance rather than real coherence.

Example B — Platform Appeals

Scenario:

A platform offers appeal buttons, but appeals rarely change outcomes, explanations are generic, and no policy correction occurs.

Law expression:

appeal channel exists but FI↓ ⇒ performance theater

Interpretation:

The feedback channel performs responsiveness without correction.

Example C — Security Alert Suppression

Scenario:

Alert noise is reduced by suppressing alerts without preserving root-cause trace. Incident recurrence later rises.

Law expression:

alert suppression → Au↓ → Γ_mis → H_security↑

Interpretation:

The system improved alert appearance while reducing feedback integrity.

Example D — Institutional Complaint Closure

Scenario:

Complaints are logged and closed, but the underlying process never changes and the same complaint pattern recurs.

Law expression:

complaint closure Φ↑ while recurrence↑ ⇒ feedback capture

Interpretation:

Complaint management replaced complaint correction.

Example E — Governance Public Comment

Scenario:

Public feedback is collected, but policy direction is predetermined and feedback cannot alter outcomes.

Law expression:

feedback channel without behavioral effect ⇒ legitimacy theater

Interpretation:

Participation signal is not governance feedback integrity.

Example F — Biological Symptom Feedback

Scenario:

A symptom is suppressed, but recurrence continues because the underlying signal was not interpreted or repaired.

Law expression:

symptom feedback suppressed ⇒ Γ_mis + H_bio↑

Interpretation:

Biological feedback must guide repair, not only symptom management.

12. Relationship to Nearby Laws

Aliases folded into this law:

• Feedback Integrity Law
• Feedback Without Integrity Becomes Capture
• Feedback Capture Law
• Selection Misfire Law
• Feedback Theater Law

Deduplication note:

This law should remain the root cybernetic feedback-integrity law. LAW-049 should handle feedback absorbability under low slack; LAW-050 should distinguish control from restoration after feedback is received.

13. Operator Mapping

Coherent operator sequence:

Ψ(field / affected-node signal) → Σ(channel scope) → Au(trace) → FI check → Γ(classify feedback) → Π(adjust) → ℛ(repair) → Τ(validate recurrence↓)

Inverted operator sequence:

feedback signal → FI failure → Γ_mis → Π wrong adjustment → Ξ / ι↑ → H↑ → recurrence persists

14. Machine-Readable Summary

id: "LAW-048"
name: "Feedback Integrity Law"
type: "law"
status: "draft"
family:
  - "Cybernetic and Meta-Theory Laws"
summary: "Feedback without integrity becomes capture."
canonical_statement: "Feedback without integrity becomes capture."
canonical_form: "FI failure → Γ_mis → Ξ → H↑"
failure_form: "feedback channel gamed / suppressed / redirected / performative ⇒ selection misfire"
variables:
  primary:
    - "FI"
    - "Γ_mis"
    - "Ξ"
    - "H"
    - "Ψ"
    - "Au"
    - "R"
  secondary:
    - "O"
    - "ε"
    - "ι"
    - "BΣ"
    - "K"
    - "µᵢ"
    - "Φ"
    - "Π"
    - "Θ"
    - "Τ"
diagnostics:
  - "Feedback Integrity"
  - "Feedback Capture Risk"
  - "Channel Gameability"
  - "Suppression Risk"
  - "Redirection Risk"
  - "Performance Theater"
  - "Classification Fidelity"
  - "Effective Auditability"
  - "Selection Accuracy"
  - "Hidden Debt"
  - "Inversion Index"
  - "Recurrence"
failure_modes:
  - "Feedback Capture"
  - "Selection Misfire"
  - "Feedback Suppression"
  - "Feedback Redirection"
  - "Feedback Gaming"
  - "Performance Theater"
  - "Misclassification"
  - "Auditability Collapse"
  - "Pseudo-Coherence"
  - "Hidden Debt Accumulation"
  - "Control-Restoration Confusion"
  - "Legitimacy Shock"
restoration_arcs:
  - "Auditability Restoration"
  - "Feedback Integrity Repair"
  - "Boundary Reconstitution"
  - "Controlled Decoupling"
  - "Restoration Capacity Rebuild"
  - "Origin-Layer Repair"
  - "Temporal Validation"
  - "Recurrence Reduction"
  - "Basin Supersession"
related_laws:
  - "LAW-006"
  - "LAW-009"
  - "LAW-013"
  - "LAW-015"
  - "LAW-016"
  - "LAW-017"
  - "LAW-031"
  - "LAW-036"
  - "LAW-037"
  - "LAW-038"
  - "LAW-040"
  - "LAW-041"
  - "LAW-043"
  - "LAW-044"
  - "LAW-045"
  - "LAW-047"
  - "LAW-049"
  - "LAW-050"
  - "LAW-051"
  - "LAW-054"
  - "LAW-102"
  - "LAW-111"
  - "LAW-120"
  - "LAW-121"
  - "LAW-124"
related_invariants:
  - "INV-001"
  - "INV-004"
operator_sequence:
  coherent:
    - "Ψ field / affected-node signal"
    - "Σ channel scope"
    - "Au trace"
    - "FI check"
    - "Γ classify feedback"
    - "Π adjust"
    - "ℛ repair"
    - "Τ validate recurrence↓"
  inverted:
    - "feedback signal"
    - "FI failure"
    - "Γ_mis"
    - "Π wrong adjustment"
    - "Ξ / ι↑"
    - "H↑"
    - "recurrence persists"
aliases:
  - "Feedback Integrity Law"
  - "Feedback Without Integrity Becomes Capture"
  - "Feedback Capture Law"
  - "Selection Misfire Law"
  - "Feedback Theater Law"
deduplication_note: "Root cybernetic feedback-integrity law. LAW-049 handles feedback absorbability under low slack; LAW-050 distinguishes control from restoration after feedback is received."
source: "content/archive/laws/technical.md"

15. Compact Card Version

LAW-048 — Feedback Integrity Law

Feedback without integrity becomes capture.

Canonical stack:

FI failure → Γ_mis → Ξ → H↑

Plain meaning:

Feedback only helps a system regulate when the channel remains honest, usable, auditable, and connected to real effects. If feedback can be gamed, suppressed, redirected, or converted into performance theater, selection misfires.

Failure form:

feedback channel gamed / suppressed / redirected / performative ⇒ selection misfire

Primary variables:

FI, Γ_mis, Ξ, H, Ψ, Au, R, O, ι, BΣ, K, µᵢ, Φ, Π, Θ, Τ

Diagnostic signature:

Feedback exists, but the channel can be gamed, suppressed, redirected, converted into performance theater, or prevented from changing system behavior. Recurrence and hidden debt do not decrease.

Failure risk:

Feedback capture, selection misfire, feedback suppression, feedback redirection, feedback gaming, performance theater, misclassification, auditability collapse, pseudo-coherence, hidden debt accumulation.

Restoration priority:

Repair the feedback channel before trusting system learning: restore traceability, protect affected-node feedback, reduce gaming and suppression, reconnect feedback to origin-layer repair, and time-validate recurrence reduction.