Archive registry entry

LAW-121 — AI as Γ-Amplifier Law

Archive registry entry

LAW-121 — AI as Γ-Amplifier Law

AI amplifies classification, filtering, interpretation, routing, and prioritization; because Γ-amplification changes what becomes visible, actionable, suppressed, or believed, AI governance must scale auditability, feedback integrity, restoration capacity, and boundary discipline accordingly.

draftid: LAW-121version: 1.0.0updated: 2026-06-17

Archive Progress

This section can be read now; registry depth and cross-references are still being strengthened.

Foundation

Online

The section has a stable overview route and basic reader context.

Technical Layer

Online

A deeper technical overview is available.

Registry

Current

171 registry entries are available.

Cross-links

Curating

Related concepts are being connected conservatively for accuracy.

Jump Through This Page

LAW-120 — Security Legibility Law

Registry

LAW-122 — AI Error Lag Law

0. Plain Statement

AI amplifies Γ.

Plain-language version:

AI does not only generate text, images, code, decisions, rankings, or recommendations.

AI amplifies classification.

It changes what gets labeled, filtered, ranked, refused, routed, summarized, escalated, ignored, translated, remembered, believed, or acted on.

Because Γ determines how a system classifies reality, AI becomes powerful whenever it shapes classification at scale.

AI governance must therefore treat AI as a classification amplifier, not merely as a content generator.

1. Formal Definition

The AI as Γ-Amplifier Law states that AI systems amplify classification, filtering, interpretation, routing, and prioritization across human, institutional, technical, media, governance, economic, and cognitive systems.

AI amplifies Γ when it determines:

what category a signal belongs to;
whether a request is safe or unsafe;
whether content is relevant or irrelevant;
whether a user is trusted or suspicious;
whether a node is normal or anomalous;
whether feedback is helpful or harmful;
whether speech is allowed, throttled, transformed, or refused;
whether evidence is salient or ignored;
whether a case is escalated or deprioritized;
whether a memory is retrieved or forgotten;
whether a narrative is summarized in one frame or another;
whether a pattern is labeled real, false, risky, low-quality, spam, fraud, abusive, or benign;
whether an event becomes visible to decision-makers.

AI becomes structurally powerful because classification precedes action.

A system acts on what it can classify.

Therefore, AI that amplifies Γ also amplifies downstream Π, ⊗, ℛ, Φ, L, and H.

2. Canonical Form

Core form:

textScroll

AI amplifies Γ

Expanded form:

textScroll

AI amplifies classification + filtering + interpretation + routing + prioritization

Downstream form:

textScroll

Γ_AI↑ ⇒ Π decisions↑ + visibility selection↑ + routing power↑ + H risk↑

Governance form:

textScroll

Γ_AI scale↑ ⇒ Au + FI + BΣ + ℛ must scale faster

Failure form:

textScroll

Γ_AI↑ + Au↓ + FI↓ + ℛ↓ ⇒ misclassification debt↑

Restoration-valid contrast:

textScroll

AI Γ-amplification valid when classification remains auditable, corrigible, bounded, repair-linked, and time-validated

Related variables:

textScroll

O, H, H_AI, ε, ι, Au, Au_eff, µᵢ, BΣ, K, σ, R, R_eff, 𝓑, 𝓓, Φ, Φ_AI, Λ, ⊗, Γ, Γ_AI, Π, Ξ, ℛ, Θ, Σ, Ψ, Τ, FI, MS, L, classification_power, filtering_power, routing_power, interpretation_power, prioritization_power, context_integrity, refusal_logic, escalation_path, correction_path, appeal_path, affected_node_feedback

Where:

TableScroll

Variable	Meaning in this law
`Γ_AI`	AI-mediated classification, labeling, interpretation, filtering, routing, ranking, or admissibility selection
`classification_power`	Degree to which AI assigns categories that shape downstream action
`filtering_power`	Degree to which AI admits, blocks, transforms, suppresses, or selects signals
`routing_power`	Degree to which AI sends outputs, users, cases, evidence, or attention into different pathways
`interpretation_power`	Degree to which AI frames meaning, intent, salience, evidence, or narrative
`prioritization_power`	Degree to which AI orders attention, response, queue position, ranking, or importance
`context_integrity`	Degree to which AI classification preserves relevant context
`refusal_logic`	AI logic used to deny, block, redirect, or restrict action
`escalation_path`	Path by which AI-classified cases move to review, support, enforcement, or repair
`correction_path`	Path by which AI misclassification can be corrected
`appeal_path`	Path by which affected nodes can challenge AI classification or routing
`affected_node_feedback`	Feedback from nodes affected by AI classification
`H_AI`	Hidden debt generated by AI classification, filtering, routing, or misinterpretation
`Φ_AI`	Visible AI performance proxy, such as accuracy, helpfulness, refusal rate, benchmark score, or safety score
`Au / Au_eff`	Auditability of AI classification, decision traces, routing logic, and effects
`FI`	Feedback integrity; AI classifications must remain corrigible
`BΣ`	Boundary integrity; AI must preserve membranes between contexts, users, scopes, roles, and claims
`R / R_eff`	Restoration capacity required when AI misclassifies, blocks, misroutes, or harms
`L`	Legitimacy of AI-mediated classification under audit
`O`	Coherence; AI Γ-amplification should preserve or improve coherence
`H`	Hidden debt; rises when AI classification errors are invisible or unrepaired
`ι / Ξ`	Inversion when AI classification claims safety, neutrality, or usefulness while producing incoherence
`Θ`	Humility preventing classification certainty and model overreach
`Σ`	Scope of valid AI classification, authority, and use
`Ψ`	Field and affected-node feedback validating AI classification effects
`Τ`	Time validation of classification performance and downstream effects

3. Core Mechanism

The law unfolds because classification sits upstream of action.

Coherent AI Γ-amplification pathway

textScroll

signal enters AI system
→ AI classifies within valid scope
→ classification trace remains auditable
→ context is preserved
→ affected feedback remains available
→ routing / refusal / action is bounded
→ repair path exists
→ classification improves over time

AI Γ-failure pathway

textScroll

signal enters AI system
→ AI classifies with weak context or hidden rule
→ output routes into refusal / ranking / enforcement / suppression
→ affected node lacks correction path
→ misclassification debt accumulates
→ trust and legitimacy decline

The core mechanism is:

textScroll

AI power begins at classification

Detailed mechanism:

AI receives a signal.

The signal may be a prompt, image, document, account, case, message, behavior, transaction, claim, search query, dataset, or sensor input.

AI classifies the signal.

The system assigns category, relevance, risk, salience, quality, intent, safety, identity, similarity, priority, or eligibility.

Classification shapes the pathway.

The classification determines whether the signal is answered, refused, ranked, routed, summarized, hidden, transformed, escalated, flagged, or ignored.

Amplification occurs at scale.

A small classification rule or model bias can affect millions of downstream interactions.

Errors become structural.

Misclassification can become invisible because users only see the output, not the hidden Γ process.

Governance must scale with Γ power.

Auditability, feedback, correction, appeal, boundary clarity, and restoration must rise as AI classification power rises.

Time validates AI classification.

AI Γ-amplification is coherent only when downstream effects reduce error, debt, recurrence, and legitimacy shock over time.

4. When This Law Applies

This law applies whenever AI classifies, filters, ranks, routes, recommends, summarizes, refuses, moderates, detects, scores, labels, interprets, prioritizes, or assigns meaning.

It is especially important when AI:

moderates content;
ranks search results;
summarizes evidence;
classifies risk;
detects fraud;
scores users;
routes support tickets;
triages medical, legal, financial, or institutional cases;
generates safety refusals;
decides what memory to retrieve;
interprets user intent;
filters news or media;
recommends jobs, products, relationships, content, or resources;
flags suspicious behavior;
produces automated reports;
performs red-team or safety classification;
shapes public belief through repeated framing.

The law applies strongly when:

textScroll

AI classification changes what becomes visible, actionable, or believed

or when:

textScroll

AI Γ operates at scale without proportional audit, feedback, and repair

Typical domains:

TableScroll

Domain	AI Γ-Amplification Expression
AI safety	Safety filters, refusals, and policy classifiers shape what users can ask, learn, or do.
AI governance	Governance must audit classification, routing, correction, appeal, and downstream effects.
Cybersecurity	AI detection amplifies classification of threat, anomaly, user behavior, and incident priority.
Media / information networks	AI ranking, summarization, and recommendation amplify salience and narrative frames.
Institutions	AI triage and scoring amplify access, priority, eligibility, trust, and repair pathways.
Economy	AI risk, fraud, credit, hiring, and market classification amplify allocation outcomes.
Culture	AI classification shapes what is treated as normal, safe, relevant, valuable, or illegitimate.
Restoration	AI can route repair signals accurately or misclassify them as noise, risk, or abuse.

5. When This Law Does Not Apply

This law should not be used to claim that all AI classification is harmful.

Classification is necessary.

AI can improve coherence when it classifies signals more accurately, routes support faster, reveals hidden debt, reduces noise, improves access, detects early risk, and supports restoration.

False-positive cases:

TableScroll

Case	Why AI Γ-amplification may be coherent
AI classifies spam accurately	Filtering can protect attention and bandwidth
AI routes support tickets faster	Routing can improve restoration capacity
AI detects anomalies early	Classification can reduce incident lag
AI summarizes evidence with traceability	Interpretation can support audit
AI flags risk with review and appeal	Classification can support safety if corrigible
AI triages urgent cases transparently	Prioritization can improve outcomes
AI detects misclassification patterns	AI can repair Γ when audit-bound

Important distinction:

The law is not anti-classification; it requires classification power to remain auditable, corrigible, bounded, and repair-linked.

6. Diagnostic Signature

Canonical diagnostic:

textScroll

Γ_AI↑ ⇒ Π decisions↑ + visibility selection↑ + routing power↑ + H risk↑

Warning signature:

textScroll

AI classification power↑
classification trace↓
context integrity↓
correction path↓
affected-node feedback↓
restoration capacity↓
⇒ AI misclassification debt

Common indicators:

TableScroll

Diagnostic	Expected movement	Interpretation
`Γ_AI`	measured	AI classification power must be visible
`classification_power`	watched	The more AI classifies, the more governance is required
`filtering_power`	watched	Filtering determines signal admissibility
`routing_power`	watched	Routing determines downstream pathway
`interpretation_power`	watched	Interpretation shapes meaning and belief
`prioritization_power`	watched	Prioritization shapes access to attention and action
`context_integrity`	must remain high	Classification requires relevant context
`classification_trace`	must remain intact	AI decisions need auditability
`correction_path`	must exist	Misclassification must be repairable
`appeal_path`	must exist when effects matter	Affected nodes need review pathways
`FI`	must remain intact	Feedback must update classification
`BΣ`	must remain intact	AI must preserve scope and boundary distinctions
`R_eff`	must scale	Repair capacity must match classification impact
`H_AI`	should ↓	Hidden AI debt must be reduced
`Φ_AI`	not sufficient	Benchmarks and refusal rates do not prove coherence
`L`	stable / ↑ if valid	Legitimacy rises when AI classification survives audit
`Τ`	required	Time validates downstream classification effects

Additional diagnostics:

TableScroll

Diagnostic	Use
AI Γ-Amplification	Measures AI classification power
Classification Amplification	Tracks AI labels and categories
Filtering Amplification	Tracks what AI admits or suppresses
Routing Amplification	Tracks pathway assignment
Interpretation Amplification	Tracks meaning and narrative framing
Misclassification Risk	Tracks false positives and false negatives
Effective Auditability	Tests if AI classification can be inspected
Context Integrity	Tests whether relevant context is preserved
Temporal Proof	Validates classification effects over time

7. Failure Pattern

If ignored, this law allows AI to reshape reality classification while appearing merely assistive.

General failure pathway:

textScroll

AI classification power increases
→ classifications become hidden infrastructure
→ routing and filtering shape action
→ affected nodes lack correction
→ misclassification debt accumulates
→ proxy metrics look good
→ legitimacy fails under field audit

Common failure modes:

AI Misclassification Amplification — AI scales classification errors.
AI Filter Capture — AI filtering determines what signals become admissible.
AI Routing Capture — AI routing determines access, escalation, and repair.
AI Context Collapse — relevant context is stripped before classification.
AI Proxy Drift — benchmark or metric success diverges from coherence.
AI Interpretive Overreach — AI assigns meaning beyond valid scope.
AI Authority Laundering — human or institutional decisions are presented as AI-neutral classification.
AI Feedback Suppression — user correction cannot reach classification layer.
AI Boundary Confusion — contexts, users, domains, or authority scopes are collapsed.
AI Rule-Stacking — classification rules accumulate beyond audit capacity.
AI Legibility Collapse — AI classification becomes untraceable.
AI Pseudo-Safety — safety proxies improve while coherence declines.
AI Trust Collapse — users lose trust when classification cannot be corrected.
Hidden Debt Accumulation — AI error becomes invisible infrastructure.
Cognitive Infrastructure Drift — AI classification reshapes public belief, meaning, and salience.

Compact failure signature:

textScroll

Γ_AI↑ + Au_eff↓ + FI↓ ⇒ H_AI↑

8. Restoration Implications

Restoration requires auditing AI at the classification layer, not only at the output layer.

The first restoration question is not:

textScroll

Was the AI output good?

The first restoration question is:

textScroll

What did the AI classify, filter, route, prioritize, or interpret before producing that output — and can that Γ-pathway be audited and repaired?

Restoration priorities:

Identify where AI performs Γ.
Map classification categories and pathways.
Map filtering and routing effects.
Preserve classification trace.
Test context integrity.
Measure false positives and false negatives.
Restore feedback and correction paths.
Create appeal and review pathways where effects matter.
Route misclassification into repair.
Validate downstream effects over time.

Relevant restoration arcs:

TableScroll

Restoration Arc	Why it applies
AI Γ Recalibration	Rebalances classification logic
AI Classification Repair	Corrects false labels and category errors
AI Filtering Audit	Tests what signals are admitted, transformed, or suppressed
AI Routing Audit	Tests pathway assignment and downstream effects
AI Context Restoration	Restores relevant context before classification
AI Feedback Integrity Restoration	Allows corrections to reach the classification layer
AI Boundary Reconstitution	Repairs collapsed scope, domain, user, and authority membranes
AI Legibility Restoration	Makes AI classification sufficiently traceable
AI Restoration Capacity Increase	Builds repair capacity for classification errors
AI Misclassification Repair	Repairs affected-node harms from AI classification
AI Governance Re-Sequencing	Places audit, feedback, and repair before scaling deployment
Hidden Debt Reduction	Repairs hidden AI debt
Temporal Validation	Confirms reduced recurrence over time

Minimal restoration sequence:

textScroll

identify Γ_AI layer
→ map categories / filters / routes
→ restore classification_trace + context_integrity
→ test false positives / negatives
→ restore Au/FI/correction/appeal
→ route errors into ℛ
→ validate H_AI↓ and L↑ over Τ

Temporal validation requirement:

textScroll

classification trace improves
context integrity improves
false positives decrease
false negatives decrease
routing errors decrease
affected-node correction reaches Γ layer
restoration capacity scales with impact
hidden AI debt decreases
legitimacy stabilizes
coherence holds or rises over time

9. Design Rule

Govern AI first where it classifies, not only where it outputs.

Operational design requirements:

Identify every AI Γ function.
Map categories, filters, routes, and priorities.
Define valid scope for each classification.
Preserve context integrity.
Preserve traceability.
Preserve correction pathways.
Preserve appeal where effects matter.
Preserve affected-node feedback.
Measure false positives and false negatives.
Measure downstream routing effects.
Measure restoration capacity.
Prevent category drift.
Prevent filter capture.
Prevent hidden authority laundering.
Validate classification over time.

Avoid:

treating AI as only content generation;
evaluating only outputs;
ignoring hidden classifiers;
hidden safety labels without appeal;
black-box user risk scores;
AI ranking without salience audit;
AI summaries without source and frame audit;
AI refusals without classification trace;
AI moderation without repair;
AI fraud detection without correction;
AI triage without affected-node pathway;
AI governance focused on policy text while classification infrastructure remains unaudited.

10. Cross-Scale Expressions

TableScroll

Scale / Layer	Expression of the Law
U0 — Substrate	AI classification affects physical, embodied, infrastructural, and material outcomes when routed into action.
U1 — Energy / capacity	AI can conserve or consume attention, staff, budgets, compute, and restoration capacity through routing.
U2 — Boundary / interface	AI classifies access, permission, identity, trust, context, privacy, and safe coupling.
U3 — Process / execution	AI routes workflows, cases, tickets, alerts, escalations, refusals, and repairs.
U4 — Classification / claim	AI directly amplifies Γ by labeling, interpreting, ranking, filtering, and summarizing.
U5 — Time / delay	AI can accelerate classification faster than audit, correction, or repair can follow.
U6 — Field effect	Outcomes reveal whether AI classification preserved coherence or amplified debt.
U7 — Recurrence / memory	AI classifications become memory, precedent, model updates, retrieval patterns, and institutional habit.
U8 — Environment / forcing	Platforms, institutions, markets, media, governance, and culture amplify AI Γ through deployment scale.

11. Examples

Example A — AI Refusal Classifier

Scenario:

An AI refuses a user request because a hidden classifier marks it as unsafe, but the system cannot show the refusal category, correction path, or review route.

Law expression:

textScroll

Γ_AI refusal + Au_eff↓ + FI↓ ⇒ misclassification debt

Interpretation:

The issue is not only the refusal; it is the hidden classification pathway behind the refusal.

Example B — AI Support Routing

Scenario:

An AI routes customer issues into tiers. It misclassifies urgent repair cases as low priority because context is stripped during intake.

Law expression:

textScroll

context_integrity↓ + routing_power↑ ⇒ repair delay

Interpretation:

AI routing amplifies classification error into restoration failure.

Example C — AI Search Ranking

Scenario:

AI ranking determines what information users see first. Certain frames become repeatedly prioritized while other relevant evidence becomes invisible.

Law expression:

textScroll

Γ_AI ranking ⇒ visibility selection + belief shaping

Interpretation:

AI classification shapes salience and belief, not only search convenience.

Example D — AI Fraud Detection

Scenario:

A fraud model flags constrained users as suspicious because unusual behavior is misclassified without economic, access, or recovery context.

Law expression:

textScroll

AI anomaly Γ + context collapse ⇒ false positive harm

Interpretation:

AI misclassification can punish constrained nodes unless appeal and repair exist.

Example E — AI Safety Benchmark Proxy

Scenario:

A model scores higher on safety benchmarks but users experience more false refusals, less correction, and lower trust in edge cases.

Law expression:

textScroll

Φ_AI↑ + FI↓ + L↓ ⇒ AI pseudo-safety

Interpretation:

AI safety metrics must remain coupled to field coherence.

Example F — Coherent AI Γ Governance

Scenario:

An AI system documents classification categories, preserves context, logs decision traces, provides appeal and correction, routes errors into repair, measures downstream effects, and updates over time.

Law expression:

textScroll

Γ_AI + Au + FI + ℛ + Τ ⇒ coherent AI classification

Interpretation:

AI Γ-amplification becomes coherent when classification power remains auditable, corrigible, bounded, and repair-linked.

12. Relationship to Nearby Laws

TableScroll

Related Law	Relationship
LAW-001 — Coherence Priority Law	AI classification is valid only when coherence is preserved
LAW-002 — Coherence Trajectory Law	AI Γ must improve trajectory over time
LAW-003 — Success Proxy Divergence Law	AI benchmarks can diverge from actual coherence
LAW-004 — Stability-Coherence Separation Law	AI can create stable classification regimes that hide debt
LAW-006 — Time Validation Law	AI classification requires temporal validation
LAW-009 — U4 / U6 Truth Law	AI claims and labels require field validation
LAW-010 — Hidden Debt Accumulation Law	Hidden AI misclassification creates debt
LAW-011 — Hidden Debt Return Law	AI classification debt returns through trust collapse or harm
LAW-012 — Error Lag Law	AI errors often become visible late
LAW-013 — Auditability-Debt Law	AI Γ requires auditability
LAW-015 — Suppressed Auditability Debt Law	Hidden AI classification creates debt
LAW-016 — Inversion Formation Law	AI safety and neutrality can invert into control
LAW-020 — Bandwidth Threshold Law	AI can exceed human audit bandwidth
LAW-024 — Latency–Gain Oscillation Law	Fast AI classification with slow repair creates instability
LAW-027 — Meaning Collapse Threshold Law	AI classification can overload meaning systems
LAW-028 — Control Density to Meaning Loss Loop	AI classification can increase control density
LAW-031 — Observability Collapse Law	Hidden classifiers reduce observability
LAW-036 — Signal Artifact Law	AI must distinguish signal from artifact
LAW-037 — Misclassification Law	LAW-121 is the AI amplification specialization of misclassification risk
LAW-038 — Pattern Recognition Discipline Law	AI pattern recognition requires discipline and audit
LAW-040 — Filtering Law	AI amplifies filtering
LAW-041 — Boundary Membrane Law	AI classification acts at boundaries
LAW-048 — Feedback Integrity Law	AI Γ must be correctable by feedback
LAW-050 — Control-Restoration Separation Law	AI classification must not route only into control
LAW-051 — Requisite Variety Law	AI governance must match classification variety
LAW-052 — Stability Proof Law	AI classification must survive perturbation
LAW-057 — Deception Instability Law	AI interpretive systems must resist deceptive classification
LAW-060 — Interface Legitimacy Law	AI classification interfaces require legitimacy
LAW-064 — Restoration Debt Reduction Law	AI errors must route into restoration
LAW-066 — Restoration Capacity Sufficiency Law	AI classification impact requires sufficient repair capacity
LAW-067 — Temporal Proof Law	AI Γ requires proof over time
LAW-095 — Meaning Directionality Law	AI classification shapes selection and meaning
LAW-102 — Legitimacy Audit Law	AI-mediated classification requires legitimacy audit
LAW-109 — High-Φ Legitimacy Scaling Law	High-influence AI requires stronger boundaries, audit, and repair
LAW-110 — Governance Sequencing Law	AI Γ must be sequenced into governance
LAW-111 — Meaning Audit Law	AI meanings and labels are not audit-exempt
LAW-112 — Security as Sustained Coherence Law	AI security must preserve coherence under forcing
LAW-113 — Incident Lag Law	AI classification errors may appear late
LAW-114 — Pseudo-Security Law	AI safety metrics can create pseudo-security
LAW-115 — Surveillance–Restoration Law	AI sensing and monitoring must route into repair
LAW-117 — Shadow–Light Security Law	AI threat modeling requires Light-governed execution
LAW-118 — Empathy Security Law	AI state estimation requires empathy and sovereignty
LAW-120 — Security Legibility Law	AI Γ requires legibility and traceability
LAW-122 — AI Error Lag Law	LAW-122 specializes AI errors as delayed visibility events
LAW-123 — AI U4 Truth Discipline Law	AI classifications require U6 validation
LAW-124 — AI Rule-Stacking Law	AI classifiers can become unauditable through rule accumulation
LAW-125 — AI Context Collapse Law	Context loss corrupts AI Γ
LAW-126 — AI Proxy Drift Law	AI metrics and proxies drift from coherence
LAW-127 — AI Decision Pipeline Law	AI actions must pass through disciplined decision sequence
LAW-128 — AI Representation Law	AI representation depends on valid classification and scope
LAW-129 — AI Capability–Legibility Gap Law	AI Γ can exceed legibility
LAW-130 — AI Membrane Triage Law	AI failures require membrane localization
LAW-131 — Cognitive Infrastructure Scaling Law	AI Γ scales into public cognition
LAW-132 — AI Legitimacy Function Law	AI legitimacy depends on accountable classification
LAW-133 — Synthetic Consensus Law	AI classification can manufacture apparent consensus
LAW-134 — Layered Interception Law	AI Γ requires layered safeguards
LAW-135 — Guardrail Belief-Sculpting Law	AI guardrails classify and shape belief
LAW-136 — Invisible Constraint Amplification Law	AI classification can invisibly amplify constraints

Aliases folded into this law:

AI as Γ-Amplifier Law
AI as Gamma-Amplifier Law
AI Classification Amplification Law
AI Filtering Amplification Law
AI Routing Amplification Law
AI Interpretive Amplification Law
AI Amplifies Γ Law

Deduplication note:

This law should remain the root AI classification-amplification law. LAW-037 defines misclassification generally. LAW-040 defines filtering generally. LAW-120 defines security legibility. LAW-121 specializes these into AI by defining AI as a Γ-amplifier whose downstream power comes from classification, filtering, interpretation, routing, and prioritization.

13. Operator Mapping

TableScroll

Operator	Role in this law
`Γ`	Primary amplified operator; AI classifies, labels, filters, interprets, routes, and prioritizes
`Π`	Operationalizes AI classifications into refusals, outputs, rankings, actions, workflows, and enforcement
`Ξ`	Captures inversion when AI classification claims neutrality or safety while producing incoherence
`⊗`	AI classification changes coupling among users, systems, evidence, institutions, and repair pathways
`ℛ`	Repairs misclassification, routing harm, refusal harm, suppression, and downstream debt
`Τ`	Validates classification performance and field effects over time
`Θ`	Prevents AI certainty, classification overreach, and benchmark self-certification
`Σ`	Defines scope, authority, context, and valid use of AI classification
`Ψ`	Field and affected-node feedback validates AI classifications
`Λ`	Tests compatibility between AI Γ-amplification and whole-system coherence

Coherent operator sequence:

textScroll

AI signal enters system
→ Θ prevent classification certainty
→ Γ_AI classify within scope
→ Σ preserve domain / role / context boundaries
→ Au/FI preserve trace and correction
→ Π route output / refusal / escalation appropriately
→ ℛ repair misclassification and affected-node effects
→ Ψ validate field outcomes
→ Τ validate reduced H_AI and stable L

Inverted operator sequence:

textScroll

AI classification power rises
→ hidden Γ becomes infrastructure
→ context integrity falls
→ Π routes into refusal / ranking / suppression / enforcement
→ FI and appeal remain weak
→ ℛ absent
→ H_AI↑
→ Ξ / ι↑
→ L↓

14. Machine-Readable Summary

yamlScroll

id: "LAW-121"
name: "AI as Γ-Amplifier Law"
type: "law"
status: "draft"
family:
  - "AI Laws"
summary: "AI amplifies classification, filtering, interpretation, routing, and prioritization; because Γ-amplification changes what becomes visible, actionable, suppressed, or believed, AI governance must scale auditability, feedback integrity, restoration capacity, and boundary discipline accordingly."
canonical_statement: "AI amplifies Γ."
core_form: "AI amplifies Γ"
expanded_form: "AI amplifies classification + filtering + interpretation + routing + prioritization"
downstream_form: "Γ_AI↑ ⇒ Π decisions↑ + visibility selection↑ + routing power↑ + H risk↑"
governance_form: "Γ_AI scale↑ ⇒ Au + FI + BΣ + ℛ must scale faster"
failure_form: "Γ_AI↑ + Au↓ + FI↓ + ℛ↓ ⇒ misclassification debt↑"
restoration_valid_contrast: "AI Γ-amplification valid when classification remains auditable, corrigible, bounded, repair-linked, and time-validated"
variables:
  primary:
    - "Γ_AI"
    - "classification_power"
    - "filtering_power"
    - "routing_power"
    - "interpretation_power"
    - "prioritization_power"
    - "context_integrity"
    - "refusal_logic"
    - "escalation_path"
    - "correction_path"
    - "appeal_path"
    - "affected_node_feedback"
    - "H_AI"
    - "Φ_AI"
    - "Au"
    - "Au_eff"
    - "FI"
    - "BΣ"
    - "R"
    - "R_eff"
    - "L"
  secondary:
    - "O"
    - "H"
    - "ε"
    - "ι"
    - "µᵢ"
    - "K"
    - "σ"
    - "𝓑"
    - "𝓓"
    - "Φ"
    - "Λ"
    - "⊗"
    - "Γ"
    - "Π"
    - "Ξ"
    - "ℛ"
    - "Θ"
    - "Σ"
    - "Ψ"
    - "Τ"
    - "MS"
diagnostics:
  - "AI Γ-Amplification"
  - "Classification Amplification"
  - "Filtering Amplification"
  - "Routing Amplification"
  - "Interpretation Amplification"
  - "Misclassification Risk"
  - "Feedback Integrity"
  - "Effective Auditability"
  - "Boundary Integrity"
  - "Restoration Capacity"
  - "Context Integrity"
  - "Legibility"
  - "Hidden Debt"
  - "Temporal Proof"
failure_modes:
  - "AI Misclassification Amplification"
  - "AI Filter Capture"
  - "AI Routing Capture"
  - "AI Context Collapse"
  - "AI Proxy Drift"
  - "AI Interpretive Overreach"
  - "AI Authority Laundering"
  - "AI Feedback Suppression"
  - "AI Boundary Confusion"
  - "AI Rule-Stacking"
  - "AI Legibility Collapse"
  - "AI Pseudo-Safety"
  - "AI Trust Collapse"
  - "Hidden Debt Accumulation"
  - "Cognitive Infrastructure Drift"
restoration_arcs:
  - "AI Γ Recalibration"
  - "AI Classification Repair"
  - "AI Filtering Audit"
  - "AI Routing Audit"
  - "AI Context Restoration"
  - "AI Feedback Integrity Restoration"
  - "AI Boundary Reconstitution"
  - "AI Legibility Restoration"
  - "AI Restoration Capacity Increase"
  - "AI Misclassification Repair"
  - "AI Governance Re-Sequencing"
  - "Hidden Debt Reduction"
  - "Temporal Validation"
related_laws:
  - "LAW-001"
  - "LAW-002"
  - "LAW-003"
  - "LAW-004"
  - "LAW-006"
  - "LAW-009"
  - "LAW-010"
  - "LAW-011"
  - "LAW-012"
  - "LAW-013"
  - "LAW-015"
  - "LAW-016"
  - "LAW-020"
  - "LAW-024"
  - "LAW-027"
  - "LAW-028"
  - "LAW-031"
  - "LAW-036"
  - "LAW-037"
  - "LAW-038"
  - "LAW-040"
  - "LAW-041"
  - "LAW-048"
  - "LAW-050"
  - "LAW-051"
  - "LAW-052"
  - "LAW-057"
  - "LAW-060"
  - "LAW-064"
  - "LAW-066"
  - "LAW-067"
  - "LAW-095"
  - "LAW-102"
  - "LAW-109"
  - "LAW-110"
  - "LAW-111"
  - "LAW-112"
  - "LAW-113"
  - "LAW-114"
  - "LAW-115"
  - "LAW-117"
  - "LAW-118"
  - "LAW-120"
  - "LAW-122"
  - "LAW-123"
  - "LAW-124"
  - "LAW-125"
  - "LAW-126"
  - "LAW-127"
  - "LAW-128"
  - "LAW-129"
  - "LAW-130"
  - "LAW-131"
  - "LAW-132"
  - "LAW-133"
  - "LAW-134"
  - "LAW-135"
  - "LAW-136"
related_invariants:
  - "INV-001"
  - "INV-002"
  - "INV-006"
  - "INV-073"
  - "INV-078"
  - "INV-080"
operator_sequence:
  coherent:
    - "AI signal enters system"
    - "Θ prevent classification certainty"
    - "Γ_AI classify within scope"
    - "Σ preserve domain / role / context boundaries"
    - "Au/FI preserve trace and correction"
    - "Π route output / refusal / escalation appropriately"
    - "ℛ repair misclassification and affected-node effects"
    - "Ψ validate field outcomes"
    - "Τ validate reduced H_AI and stable L"
  inverted:
    - "AI classification power rises"
    - "hidden Γ becomes infrastructure"
    - "context integrity falls"
    - "Π routes into refusal / ranking / suppression / enforcement"
    - "FI and appeal remain weak"
    - "ℛ absent"
    - "H_AI↑"
    - "Ξ / ι↑"
    - "L↓"
aliases:
  - "AI as Γ-Amplifier Law"
  - "AI as Gamma-Amplifier Law"
  - "AI Classification Amplification Law"
  - "AI Filtering Amplification Law"
  - "AI Routing Amplification Law"
  - "AI Interpretive Amplification Law"
  - "AI Amplifies Γ Law"
deduplication_note: "Root AI classification-amplification law. LAW-037 defines misclassification generally. LAW-040 defines filtering generally. LAW-120 defines security legibility. LAW-121 specializes these into AI by defining AI as a Γ-amplifier whose downstream power comes from classification, filtering, interpretation, routing, and prioritization."
source: "content/archive/laws/technical.md"

15. Compact Card Version

LAW-121 — AI as Γ-Amplifier Law

AI amplifies Γ.

Core form:

textScroll

AI amplifies Γ

Expanded form:

textScroll

AI amplifies classification + filtering + interpretation + routing + prioritization

Plain meaning:

AI does not only generate outputs. AI classifies signals, filters admissibility, interprets meaning, routes attention, prioritizes action, and shapes what becomes visible, actionable, suppressed, believed, or repaired. Because classification precedes action, AI governance must begin at Γ.

Downstream form:

textScroll

Γ_AI↑ ⇒ Π decisions↑ + visibility selection↑ + routing power↑ + H risk↑

Governance form:

textScroll

Γ_AI scale↑ ⇒ Au + FI + BΣ + ℛ must scale faster

Primary variables:

Γ_AI, classification_power, filtering_power, routing_power, interpretation_power, prioritization_power, context_integrity, refusal_logic, escalation_path, correction_path, appeal_path, affected_node_feedback, H_AI, Φ_AI, Au, Au_eff, FI, BΣ, R, R_eff, L, Γ, Π, Ξ, ℛ, Θ, Σ, Ψ, Τ

Diagnostic signature:

AI classification power rises while classification trace, context integrity, correction paths, affected-node feedback, and restoration capacity fall. This indicates AI misclassification debt.

Failure risk:

AI misclassification amplification, AI filter capture, AI routing capture, AI context collapse, AI proxy drift, AI interpretive overreach, AI authority laundering, AI feedback suppression, AI boundary confusion, AI rule-stacking, AI legibility collapse, AI pseudo-safety, AI trust collapse, hidden debt accumulation, cognitive infrastructure drift.

Restoration priority:

Identify where AI performs Γ, map categories, filters, routes, and priorities; restore classification trace and context integrity; test false positives and false negatives; restore feedback, correction, appeal, and repair; and validate downstream coherence over time.

LAW-120 — Security Legibility Law

Registry

LAW-122 — AI Error Lag Law