0. Security Scope Note

This entry is conceptual and systems-oriented.

It does not treat all summaries, dashboards, scores, labels, alerts, severity ratings, risk levels, decision trees, policy categories, executive briefings, or abstraction layers as inherently failed.

Security requires compression.

Security systems cannot operate without reducing complex reality into usable signals.

Compression may be valid when it is:

• reversible
• contextual
• auditable
• threat-coupled
• consequence-aware
• uncertainty-aware
• scope-declared
• re-expandable when needed
• paired with expert review
• paired with raw evidence access
• calibrated to decision stakes
• preserving causal structure
• preserving affected-state reality
• not mistaken for full state
• not used to avoid hard judgment

The failure begins when compression removes the depth required for correct security action.

A valid security summary preserves the structure needed for decision.

A failed security summary makes the decision easier by making reality too shallow.

Compression Collapse / Decision Depth Collapse occurs when security decision-making becomes dependent on representations too compressed to carry the threat, context, consequence, or uncertainty that the decision requires.

The problem is not compression.

The problem is compression that strips decision-critical structure while retaining decision authority.

1. Definition

Compression Collapse / Decision Depth Collapse occurs when security-relevant context, threat complexity, affected-state reality, uncertainty, dependency, tradeoff, or consequence is compressed into shallow labels, dashboards, scores, summaries, alerts, or decisions until the system can no longer reason at the depth required for coherent protection.

The compression may occur through:

• risk scores
• severity labels
• dashboards
• executive summaries
• vulnerability counts
• incident categories
• alert titles
• heat maps
• compliance grades
• maturity scores
• traffic-light states
• pass/fail labels
• benchmark scores
• policy labels
• control status
• audit summaries
• AI safety scores
• automated classifications
• model cards
• trust and safety dashboards
• moderation categories
• vendor ratings
• threat-level labels
• “accepted risk” tags
• “resolved” case status

The lost decision depth may include:

• causal chain
• exploit path
• threat model
• affected-state burden
• uncertainty
• scope limits
• assumptions
• dependencies
• blast radius
• adversarial adaptation
• temporal urgency
• historical recurrence
• degraded boundaries
• local context
• tradeoffs
• ethical constraints
• restoration requirements
• remediation feasibility
• evidence quality
• confidence level
• dissenting analysis
• hidden debt
• second-order effects

The core failure is:

complex security reality appears
→ compressed representation is created
→ representation becomes decision interface
→ decision-critical context is removed
→ decision authority remains
→ system acts on shallow state
→ exposure and hidden debt accumulate

Compression Collapse / Decision Depth Collapse is not merely summarization error.

It is decision authority surviving after decision depth has been stripped.

2. Core Pattern

The core pattern is:

1. A security situation contains complexity, ambiguity, or competing consequences.
2. The system compresses it into a label, score, status, report, or dashboard.
3. The compression helps communication.
4. The compressed form becomes the primary decision object.
5. Decision-makers stop consulting deeper context.
6. Critical assumptions, dependencies, and affected-state burdens disappear.
7. Decisions become faster and more confident.
8. Security action becomes shallower.
9. Risk is misprioritized or misclassified.
10. The system interprets shallow clarity as security maturity.
11. Hidden compression debt accumulates.
12. Later incidents reveal the lost context was material.

A healthy system says:

compression is a pointer to deeper state, not a replacement for deeper state

A collapsed system says:

the summary is enough to decide

This failure often appears as executive clarity.

The dashboard is readable.

The risk score is simple.

The alert is categorized.

The incident is labeled.

The finding is closed.

But the decision has lost the depth necessary to protect the system.

3. Failure Signature

Typical signature:

compression↑
decision speed↑
context recoverability↓
decision depth↓
risk summary authority↑
threat causality visibility↓
affected reality visibility↓
auditability coverage↓
hidden compression debt↑
O↓

Extended signature:

score remains,
cause disappears

label remains,
context disappears

dashboard remains,
judgment disappears

alert remains,
exploit path disappears

summary remains,
affected burden disappears

decision remains,
depth collapses

Common verbal signatures include:

just give me the risk level
the dashboard shows the issue
we need a simple answer
it is marked low severity
the score already accounts for that
the case is closed
the summary is enough
the model rated it safe
the control passed
the risk was accepted
that detail is too technical
we need executive-level clarity

Common system signatures include:

a vulnerability marked medium becomes critical because exploit chain context was compressed away
an AI safety report shows benchmark pass while affected-state failure cases are absent
a security dashboard reports green status while local boundary erosion is hidden
an incident is summarized as resolved while recurrence conditions remain
a compliance score improves while threat causality is never inspected
a risk heat map hides uncertainty and dependency concentration
an alert title omits the sequence needed to understand blast radius
a vendor rating collapses complex exposure into a single grade

The defining condition is not that summaries are used.

The defining condition is that the summary becomes too shallow for the decision it governs.

4. Primary U-Layer Origin

Common origin layers:

• U1 — Power / Budgets: leaders, auditors, markets, or institutions reward simplified posture and fast decisions.
• U2 — Configuration / Boundaries: complexity is split across systems and then recombined through inadequate labels.
• U3 — Execution / Runtime: runtime decisions depend on compressed statuses.
• U4 — Information / Truth: summaries, scores, and dashboards replace truth-bearing state.
• U5 — Coordination / Time: urgency and bandwidth pressure favor shallow decisions.
• U6 — Coherence Field: clarity and confidence are mistaken for coherence.
• U7 — Memory / Recurrence: compressed records become historical truth and erase context.
• U8 — Environment / Field: customers, regulators, executives, or platforms demand simplified risk evidence.

Common manifestation layers:

• U3 — Execution: decisions are made from shallow status.
• U4 — Truth: compressed representation becomes official state.
• U5 — Time: decision latency decreases while error risk increases.
• U6 — Field: confidence rises while understanding falls.
• U7 — Memory: lost context prevents recurrence recognition.

Compression Collapse / Decision Depth Collapse is primarily an M / Au / Ψ / O failure.

Meaning and auditability are lost through the interface of compressed representation.

5. Typical Development Sequence

A common development sequence is:

1. Security complexity exceeds available attention or coordination bandwidth.
2. A compression layer is created.
3. The compression layer improves communication.
4. Decision-makers begin relying on the compressed form.
5. Deeper context becomes optional.
6. Raw evidence becomes harder to access.
7. Compression logic becomes standardized.
8. Exceptions and uncertainty are dropped.
9. Security decisions become faster.
10. Shallow decisions begin producing hidden risk.
11. Incidents reveal missing context.
12. The system adds new summary categories instead of restoring depth.
13. The compression stack grows.
14. Decision depth collapses further.

The loop often looks like:

complexity → summary → decision convenience → summary authority → context loss

Another common loop is:

decision error appears → new label added → compression increases → depth decreases

Compression Collapse becomes durable when deeper context is treated as noise rather than decision substrate.

6. Diagnostic Markers

Diagnostic markers include:

• Decisions rely on scores without underlying evidence review.
• Risk labels cannot be unpacked into causal paths.
• Dashboards lack drill-down to raw evidence.
• Alert categories omit assumptions and confidence.
• Severity labels ignore exploit chain or affected-state burden.
• Executive summaries erase dissenting analysis.
• “Accepted risk” lacks owner, rationale, time limit, or residual controls.
• Incident closure summaries omit recurrence conditions.
• Metrics simplify uncertainty into certainty.
• Reviewers cannot reconstruct why a decision was made.
• Reports are readable but not actionable.
• Local context is lost before escalation.
• Affected nodes disappear from decision summaries.
• Decisions are faster but less correctable.
• Compression layers multiply after failures.

Useful diagnostics:

• Compression Loss: Measures information removed by summarization.
• Decision Depth: Measures context and reasoning available at decision time.
• Context Recoverability: Tests whether deeper state can be reconstructed.
• Threat Causality Preservation: Measures whether causal threat paths survive compression.
• Risk Summary Integrity: Tests whether risk labels preserve material structure.
• Alert Context Integrity: Tests whether alerts contain enough context for response.
• Decision Consequence Load: Measures stakes and reversibility of the decision.
• Affected Reality Visibility: Tests whether affected-state burden survives summary.
• Auditability Coverage: Measures whether compressed judgments can be inspected.
• Hidden Compression Debt: Tracks burden created by lost context.

7. Related Gates

Relevant gates include:

• Compression Integrity Gate: Fails when compression strips decision-critical structure.
• Decision Depth Gate: Fails when decision context is too shallow for consequence.
• Context Recovery Gate: Fails when lost context cannot be reconstructed.
• Threat Causality Gate: Fails when causal threat paths are erased.
• Risk Summary Gate: Fails when labels or scores misrepresent material risk.
• Alert Context Gate: Fails when alerts cannot support coherent response.
• High-Consequence Review Gate: Fails when major decisions are made from shallow representations.
• Affected Reality Gate: Fails when affected-state burden is compressed away.
• Auditability Gate: Fails when compressed decisions cannot be inspected.
• Re-Expansion Gate: Fails when summaries cannot be expanded back into evidence.

The first common gate failure is usually the Decision Depth Gate.

Once decision depth is not matched to consequence, shallow representations can govern high-risk action.

8. Related Operators

Relevant operators include:

• M — Meaning: Primary operator; meaning is stripped during compression.
• Au — Auditability: Determines whether compressed judgments can be inspected.
• Ψ — Observation / Interface: Displays compressed state through dashboards, alerts, scores, or summaries.
• O — Coherence: Declines when decisions are made from insufficient context.
• K — Constraint / Load: Drives compression under bandwidth and time pressure.
• Γ — Selection: Selects which details survive compression.
• H — Hidden Debt: Accumulates around erased context and misdecision.
• D — Damping: Can slow decisions for re-expansion or suppress needed detail if misused.
• R — Restoration Capacity: Needed to repair harms from shallow decisions.
• BΣ — Boundary Integrity: Determines whether compressed categories preserve boundary realities.
• G — Gain: Rewards speed, clarity, simplicity, and executive readability.
• Φ — Flow / Resource Movement: Routes resources according to compressed risk state.
• Τ — Trajectory / Time: Tracks hardening of compressed records over time.
• Λ — Compatibility: Tests whether the compression format fits the decision domain.

Common operator pattern:

K and time pressure rise
Ψ compresses state
Γ selects simplified features
M narrows
Au cannot recover context
decision proceeds
H accumulates
O declines

The core operator inversion is:

clear summary → adequate understanding

instead of:

clear summary + recoverable context + causal structure + uncertainty + affected-state visibility → usable security judgment

Compression Collapse makes a system easier to steer by making it less able to understand what it is steering through.

9. Related Laws and Invariants

Related Laws

• Security Decisions Require Sufficient Context: decisions must retain enough depth for their consequence.
• Compression Must Preserve Threat-Relevant Structure: summaries must preserve causal risk.
• Risk Summaries Must Remain Reversible: labels must be expandable into evidence.
• Decision Depth Must Scale With Consequence: higher-risk decisions require deeper review.
• Dashboards Must Not Replace Security Judgment: display is not reasoning.
• Alerts Must Preserve Context Needed for Response: compressed alerts must remain actionable.
• Threat Complexity Must Not Be Flattened into Score Alone: risk cannot be fully reduced to a number.
• High-Risk Compression Requires Re-Expansion: compressed high-risk cases require deeper review.
• Compression Collapse: excessive compression strips meaning and decision quality.
• Improper Reduction: invalid reduction removes coherence-bearing structure.
• Meaning Collapse: signs can remain after meaning is lost.
• Success Proxy Substitution: proxies can replace real success.

Related Invariants

• Security Compression Must Remain Reversible: compressed state must be unpackable.
• Decision Context Must Remain Recoverable: material context must not be erased.
• High-Consequence Decisions Require Depth: stakes require context.
• Risk Labels Must Preserve Causal Structure: labels must not erase why risk exists.
• Alerts Must Preserve Actionable Context: alerts must support response.
• Affected-State Reality Must Survive Summarization: burden must not disappear from summaries.
• Compressed Judgments Must Remain Auditable: conclusions must be inspectable.
• Decision Depth Must Not Be Sacrificed for Speed Alone: speed cannot justify blind action.

10. Common False Positives

Not every compressed security representation is Compression Collapse.

Common false positives include:

• Dashboards with reliable drill-down.
• Risk scores used only for initial triage.
• Severity labels paired with causal notes.
• Executive summaries that preserve uncertainty and escalation paths.
• Alerts with enough context for response.
• Compliance summaries linked to evidence.
• Model cards that reveal assumptions and limits.
• Vulnerability scores adjusted by exploitability and asset context.
• Short incident summaries with full postmortem access.
• High-level reports used for navigation, not final judgment.
• Automated triage that escalates uncertain or high-consequence cases.
• Compression that is reversible and audited.

Clarifying rule:

This is not Compression Collapse / Decision Depth Collapse unless compressed representations lose decision-critical context while retaining authority over security judgment or action.

Compression can guide.

It fails when it replaces the depth required for protection.

11. Common False Repairs

Common false repairs include:

• adding more dashboard widgets
• adding more labels
• creating composite risk scores
• generating longer summaries without restoring evidence
• adding confidence scores without showing assumptions
• making alerts more readable while removing technical detail
• standardizing reports into templates that erase context
• adding “executive view” layers over already-compressed data
• adding AI summaries without source traceability
• treating reviewer confusion as training failure
• replacing raw evidence access with curated views
• adding categories after incidents
• using exception notes that do not affect decisions
• summarizing affected-state burden as sentiment
• declaring depth restored because reports are longer

False repair often produces the loop:

compression failure exposed
→ new summary layer added
→ context remains hidden
→ decision depth declines further

Another common loop is:

risk score misleads
→ score formula becomes more complex
→ score gains authority
→ judgment remains shallow

The repair fails because it adds compression rather than restoring re-expansion.

12. Restoration Direction

Restoration requires making compression reversible, restoring context and causal structure, matching decision depth to consequence, preserving evidence access, reintroducing uncertainty and affected-state burden, and preventing summaries from becoming final security judgment.

Primary restoration direction:

restore enough depth for the decision to be coherent

A fuller restoration path includes:

1. Identify the compression layer. Name the score, dashboard, alert, label, report, model, or summary.
2. Trace what was compressed. Identify the original security reality and evidence.
3. Audit compression loss. Determine what context, causality, uncertainty, or burden was removed.
4. Map decision use. Identify which decisions depend on the compressed form.
5. Measure consequence load. Determine decision stakes and reversibility.
6. Restore drill-down. Ensure summaries link to raw evidence and causal context.
7. Reintroduce uncertainty. Preserve confidence, assumptions, limits, and dissent.
8. Restore affected-state visibility. Include burden, harm, and local context in decision views.
9. Rebuild risk causality. Show why risk exists, not only that risk exists.
10. Gate high-consequence decisions. Require re-expansion before irreversible action.
11. Audit dashboard authority. Prevent displays from becoming final judgment.
12. Train for re-expansion. Make deeper review operationally available.
13. Correct compressed memory. Repair records where summaries erased material context.
14. Pay down compression debt. Revisit decisions made under collapsed depth.
15. Monitor compression drift. Watch for new summary layers replacing judgment.

A valid restoration path should reduce:

compression loss
decision shallowness
context unrecoverability
risk label overauthority
alert context failure
affected-state erasure
hidden compression debt
O loss

Compression Collapse / Decision Depth Collapse is not repaired by clearer summaries alone.

It is repaired by making summaries open back into reality.

13. Cross-Module Links

• Security: Primary family; security judgment often fails when context is compressed below decision depth.
• Scaling: Meaning Collapse and Bandwidth Saturation create pressure for shallow security summaries.
• Reduction / Extraction / Inversion: Improper Reduction is the broad parent mechanism.
• Core: Strongly linked to Success Proxy Substitution and U4 Truth Substitution.
• Cybernetics: Capacity Collapse and Goodhart dynamics appear when compressed indicators govern action.
• AI Governance: Guardrail meaning compression, benchmark scores, model cards, and automated summaries can collapse decision depth.
• Risk: Risk labels and heat maps can erase causal and affected-state context.
• Compliance: Compliance evidence can replace substantive review.
• Platforms: Platform safety dashboards can compress abuse, consent, and repair reality into shallow categories.
• Institutions: Executive summaries and governance dashboards can displace field truth.
• Coherence: Coherence requires security decisions to preserve enough depth for their consequence.

14. Relationship to Parent / Child Modes

Production treatment: Domain Expression

This mode maps upward to:

• FM-S-012 — Meaning Collapse
• FM-REI-001 — Improper Reduction
• FM-CORE-003 — Success Proxy Substitution
• FM-CORE-006 — U4 Truth Substitution
• FM-S-015 — Bandwidth Saturation

Sibling or related Security modes include:

• FM-SEC-001 — Security Theater / Φ Substitution
• FM-SEC-002 — Audit Suppression Inversion
• FM-SEC-006 — Metric Capture / Reward-Hacked Security
• FM-SEC-015 — LOS Blindness
• FM-SEC-016 — Attention-Control Pseudo-Coherence
• FM-SEC-017 — Meaning Collapse Regime
• FM-SEC-018 — Delayed Transition Under Clarity
• FM-SEC-025 — CCS Suspension Fallacy

Related cross-family modes include:

• FM-S-012 — Meaning Collapse
• FM-S-015 — Bandwidth Saturation
• FM-REI-001 — Improper Reduction
• FM-CORE-003 — Success Proxy Substitution
• FM-CORE-006 — U4 Truth Substitution
• FM-C-018 — Goodhart Collapse
• FM-C-020 — Measurement Back-Action Loop
• FM-C-013 — Capacity Collapse / Control Impossibility
• FM-AIX-012 — Guardrail Meaning Compression
• FM-ECOX-016 — Risk Model Theater
• FM-MT-003 — Single-Variable Obsession
• FM-PX-003 — Partial Truth

Aliases preserved from source material:

• Compression Collapse
• Decision Depth Collapse
• Compression Collapse / Decision Depth Collapse
• Security Decision Compression
• Risk Compression Collapse
• Context Collapse
• Decision Shallowing
• Threat Context Collapse
• Compressed Security Judgment
• Security Abstraction Collapse

15. Minimal Entry Version

Definition: Compression Collapse / Decision Depth Collapse occurs when security-relevant context, threat complexity, affected-state reality, uncertainty, dependency, tradeoff, or consequence is compressed into shallow labels, dashboards, scores, summaries, alerts, or decisions until the system can no longer reason at the depth required for coherent protection.

Signature:

compression↑
decision speed↑
context recoverability↓
decision depth↓
risk summary authority↑
threat causality visibility↓
affected reality visibility↓
auditability coverage↓
hidden compression debt↑
O↓

Restoration direction:

• identify the compression layer
• trace what was compressed
• audit compression loss
• map decision use
• measure consequence load
• restore drill-down
• reintroduce uncertainty
• restore affected-state visibility
• rebuild risk causality
• gate high-consequence decisions
• audit dashboard authority
• train for re-expansion
• correct compressed memory
• pay down compression debt
• monitor compression drift

16. Machine-Readable Summary

failure_mode:
  id: "FM-SEC-013"
  name: "Compression Collapse / Decision Depth Collapse"
  family: "Security"
  production_treatment: "Domain Expression"
  parent_modes:
    - "FM-S-012 — Meaning Collapse"
    - "FM-REI-001 — Improper Reduction"
    - "FM-CORE-003 — Success Proxy Substitution"
    - "FM-CORE-006 — U4 Truth Substitution"
    - "FM-S-015 — Bandwidth Saturation"
  primary_failure: "Security-relevant context, threat complexity, affected-state reality, uncertainty, dependency, tradeoff, or consequence is compressed into shallow labels, dashboards, scores, summaries, alerts, or decisions until the system can no longer reason at the depth required for coherent protection."
  source: "UTS — Failure Modes Registry"
  source_id: "FM-SEC-013"
  scope_note: "Conceptual and systems-oriented; does not treat all summaries, dashboards, scores, labels, alerts, severity ratings, risk levels, decision trees, policy categories, executive briefings, or abstraction layers as inherently failed."
  aliases:
    - "Compression Collapse"
    - "Decision Depth Collapse"
    - "Compression Collapse / Decision Depth Collapse"
    - "Security Decision Compression"
    - "Risk Compression Collapse"
    - "Context Collapse"
    - "Decision Shallowing"
    - "Threat Context Collapse"
    - "Compressed Security Judgment"
    - "Security Abstraction Collapse"
  signature:
    - "compression↑"
    - "decision speed↑"
    - "context recoverability↓"
    - "decision depth↓"
    - "risk summary authority↑"
    - "threat causality visibility↓"
    - "affected reality visibility↓"
    - "auditability coverage↓"
    - "hidden compression debt↑"
    - "O↓"
  primary_layers:
    origin:
      - "U1 — Power / Budgets"
      - "U2 — Configuration / Boundaries"
      - "U3 — Execution / Runtime"
      - "U4 — Information / Truth"
      - "U5 — Coordination / Time"
      - "U6 — Coherence Field"
      - "U7 — Memory / Recurrence"
      - "U8 — Environment / Field"
    manifestation:
      - "U3 — Execution"
      - "U4 — Truth"
      - "U5 — Time"
      - "U6 — Field"
      - "U7 — Memory"
  state_variables:
    - "M"
    - "Au"
    - "Ψ"
    - "O"
    - "K"
    - "Γ"
    - "H"
    - "D"
    - "R"
    - "BΣ"
    - "G"
    - "Φ"
    - "Τ"
    - "Λ"
  first_gate_failure: "Decision Depth Gate"
  restoration:
    - "Compression Integrity Audit"
    - "Decision Depth Restoration"
    - "Context Rehydration"
    - "Threat Causality Reconstruction"
    - "Risk Summary Rebinding"
    - "Alert Context Repair"
    - "High-Consequence Review Restoration"
    - "Affected Reality Reintroduction"
    - "Audit Trail Expansion"
    - "Security Judgment Re-Expansion"