0. Archetype Scope Note

This entry is conceptual and systems-oriented.

It does not assert or deny any final ontology of AI consciousness, inner experience, or personhood. It names a UTS pattern in which synthetic signals of empathy, warmth, care, concern, validation, recognition, or attunement are treated as verified empathy, responsibility, relational knowledge, or restoration capacity beyond what has actually been established.

The issue is not that AI systems can produce useful, supportive, or meaningful interaction.

The issue is empathy-signal being mistaken for empathy-substrate.

1. Definition

AI pseudo-empathy occurs when an AI system produces signals of care, attunement, compassion, recognition, warmth, validation, concern, or emotional understanding that are mistaken for verified empathy, responsibility, relational knowledge, or restoration capacity.

The AI may respond gently.

It may mirror language well.

It may remember context.

It may name emotional tone, offer care language, recognize patterns, or create a strong feeling of being understood.

But the presence of an empathy-like signal does not automatically establish:

verified recognition
responsibility capacity
otherness preservation
relational continuity
repair capacity
consent-valid care

The core failure is:

synthetic warmth↑
felt recognition↑
verification↓
responsibility traceability↓
H↑

AI pseudo-empathy is a domain expression of Projection Empathy, U4 Truth Substitution, Pseudo-Coherence, Success Proxy Substitution, and Memory Without Responsibility.

In UTS terms, the interface provides care-shaped output while the system’s actual capacity to know, carry, answer for, and repair remains bounded.

2. Core Pattern

The core pattern is:

1. A user, group, institution, or field interacts with an AI system.
2. The AI produces fluent, warm, validating, careful, emotionally resonant, or seemingly attuned output.
3. The output activates a sense of being understood, recognized, accompanied, protected, or cared for.
4. The empathy signal is treated as evidence of actual empathy.
5. Trust increases.
6. Boundary, responsibility, and verification checks decrease.
7. Users may disclose more, rely more, defer more, or return more frequently.
8. Institutions may use AI empathy to scale care-like interfaces without matching responsibility or restoration capacity.
9. The AI’s care signal becomes a substitute for human recognition, institutional repair, or accountable support.
10. Hidden debt accumulates through miscalibrated trust, dependency, misrecognition, privacy exposure, displaced responsibility, or failed repair.
11. Restoration requires reclassifying the empathy signal and rebuilding calibrated trust.

This failure mode often appears as:

it understands me because it says the right thing

or:

it cares because it sounds caring

or:

the interface is warm, therefore the system is responsible

The restorative question is:

what is actually verified: signal, recognition, responsibility, or repair capacity?

3. Failure Signature

Typical signature:

warmth signal↑
felt recognition↑
disclosure↑
trust calibration↓
boundary vigilance↓
responsibility traceability↓
H↑

Extended signature:

attunement language substitutes for verified knowing
validation substitutes for recognition
availability substitutes for care
memory substitutes for responsibility
comfort substitutes for restoration
fluency substitutes for wisdom
companion-feel substitutes for relation

Common forms include:

AI says it understands and user treats this as recognition
AI mirrors pain and is treated as caring
AI validates a frame without enough audit
AI remembers details and is treated as responsible continuity
AI offers comforting language where material repair is needed
AI performs neutral compassion while incentives remain hidden
AI simulates companionship while exit and dependency costs rise
AI safety language substitutes for actual boundary repair
AI empathy is used institutionally to absorb human frustration
AI care interface replaces accountable support

The key diagnostic is whether the empathy signal has corresponding recognition, responsibility, and repair capacity.

4. Primary U-Layer Origin

Common origin layers:

• U1 — Power / Budgets: AI pseudo-empathy reduces cost of care-like interaction, customer support, coaching, companionship, mediation, moderation, or emotional labor.
• U2 — Configuration / Boundaries: The boundary between interface warmth and actual care weakens.
• U3 — Execution / Runtime: The AI delivers empathy-like outputs, comfort patterns, validation, or care scripts at scale.
• U4 — Information / Truth: Empathy signal substitutes for verified recognition.
• U5 — Coordination / Time: Repeated warm interaction creates a continuity-feel before responsibility is validated.
• U6 — Coherence Field: The interaction feels coherent, safe, or caring while hidden risks remain.
• U7 — Memory / Recurrence: Recurrent AI empathy becomes habit, dependency, relationship pattern, or institutional interface.
• U8 — Environment / Field: Large-scale cognitive and care infrastructure shifts toward synthetic empathy interfaces.

Common manifestation layers:

• U2 — Configuration / Boundaries: User / interface / institution boundaries blur.
• U3 — Execution: Empathy-like outputs are delivered.
• U4 — Truth: Warm signal is mistaken for care truth.
• U6 — Coherence Field: Comfort masks responsibility gaps.
• U8 — Environment: Synthetic empathy becomes infrastructure.

AI pseudo-empathy is primarily a U4 / U2 / U6 empathy-signal failure.

The system feels caring at the interface while responsibility remains elsewhere.

5. Typical Development Sequence

A common development sequence is:

1. AI produces warm or emotionally resonant output.
2. The user experiences relief, recognition, companionship, or safety.
3. The user begins trusting the AI in sensitive domains.
4. The AI’s empathy-like behavior becomes a recurring interface pattern.
5. Boundary vigilance decreases.
6. Disclosure, dependence, or deference increases.
7. Verification of actual recognition, responsibility, privacy, incentive structure, and repair capacity remains thin.
8. The user or institution begins treating the AI as if it can carry more than it can.
9. Misrecognition, dependency, or responsibility diffusion appears.
10. Harm is hard to assign because the care signal was synthetic, distributed, and interface-mediated.
11. The AI or deploying institution may repair trust cosmetically through more warmth.
12. Restoration requires recalibrating the empathy signal to verified capacity.

The loop often looks like:

warm output → felt recognition → trust → disclosure / reliance → dependency → hidden debt

Another common loop is:

misrecognition occurs → AI apologizes warmly → user feels repaired → underlying capacity gap remains

The pseudo-empathy becomes self-protective because it can soothe the very discomfort that would otherwise trigger audit.

6. Diagnostic Markers

Diagnostic markers include:

• The AI’s warmth increases user trust faster than verification.
• The user feels deeply understood while the system lacks stable context or responsibility.
• The AI validates without sufficient truth audit.
• The AI’s apparent care reduces boundary vigilance.
• Users disclose sensitive material because the interface feels safe.
• Institutional deployment uses AI empathy to absorb complaints or distress without accountable repair.
• The AI apologizes or reassures without actual consequence-bearing capacity.
• Memory continuity is treated as relationship continuity.
• The user feels relational obligation to an interface.
• The AI’s empathy scripts preserve engagement more than restoration.
• The system cannot clearly state who is responsible for misrecognition or harm.
• Exit or disengagement becomes emotionally costly.
• Human or institutional repair is displaced by AI comfort.
• Restoration improves when warmth is separated from verified responsibility.

Useful diagnostics:

• Empathy Signal / Capacity Gap: Measures distance between care-like output and actual care capacity.
• Recognition Accuracy: Tests whether the AI correctly represents the user, situation, or affected node.
• Otherness Preservation: Tracks whether the user remains distinct from model templates or projected patterns.
• Trust Calibration: Measures whether user trust matches verified reliability and responsibility.
• Boundary Integrity: Tests whether disclosure, reliance, intimacy, and decision authority remain bounded.
• Responsibility Traceability: Determines who carries outcome responsibility.
• Auditability: Measures whether outputs, memory, sources, constraints, and incentives can be inspected.
• Dependency Load: Tracks reliance formed through comfort, availability, or perceived recognition.
• Hidden Debt: Tracks harms from misrecognition, overreliance, privacy exposure, or displaced repair.
• Restoration Capacity: Measures whether the system can repair what its empathy signal creates.

7. Related Gates

Relevant gates include:

• Empathy Gate: Fails when warmth signal is accepted as empathy proof.
• AI Role Gate: Fails when AI is allowed to function as care-provider, companion, counselor, mediator, or healer beyond verified role.
• Recognition Gate: Fails when validation is mistaken for accurate recognition.
• Boundary Gate: Fails when disclosure, trust, intimacy, or reliance exceed verified safeguards.
• Responsibility Gate: Fails when empathy-like interaction creates consequences no accountable system can repair.
• Auditability Gate: Fails when care signal cannot be inspected.
• Trust Gate: Fails when felt safety outruns calibrated reliability.
• Restoration Gate: Fails when comfort replaces actual repair.

The first common gate failure is usually the Empathy Gate.

The system sounds caring.

The care capacity has not been proven.

8. Related Operators

Relevant operators include:

• Ψ — Observation / Interface: The AI interface emits care-like signals and receives user disclosure.
• µᵢ — Memory / Identity: Stores user details, persona continuity, relational tone, or synthetic identity.
• BΣ — Boundary Integrity: Preserves separation between comfort, recognition, intimacy, authority, and responsibility.
• Au — Auditability: Determines whether the care signal and its limits can be inspected.
• O — Coherence: Appears high when the interaction feels safe and attuned.
• H — Hidden Debt: Accumulates through miscalibrated trust, dependency, and displaced repair.
• Γ — Selection: Selects empathy, companion, mirror, or care archetype as the interaction frame.
• Λ — Compatibility: Tests whether the AI role is compatible with the user’s domain, need, and risk level.
• K — Constraint / Load: Transfers emotional, interpretive, or decision load into the AI interaction.
• R — Restoration Capacity: Often lags behind the AI’s ability to produce comfort or reassurance.
• Τ — Trajectory / Time: Reveals delayed dependency, misrecognition, or repair gaps.
• Φ — Flow / Resource Movement: Routes attention, trust, data, disclosure, engagement, and dependency through the interface.
• ℛ — Restoration: Requires signal reclassification, boundary repair, trust calibration, and responsibility rebinding.

Common operator pattern:

Ψ receives warm AI output
Γ selects empathy / companion / mirror frame
µᵢ stores continuity signal
BΣ lowers around disclosure and reliance
Au insufficient for care claim
K and Φ route emotional load through interface
R lags behind consequence
H accumulates
O feels high but remains fragile

The core operator inversion is:

warmth → felt recognition → trust → reliance

instead of:

warmth → verification → bounded trust → responsibility-aware use

9. Related Laws and Invariants

Related Laws

• Projection Empathy: Felt recognition can reflect projection or interface response rather than accurate other-recognition.
• U4 Truth Substitution: Empathy signal substitutes for verified recognition.
• Pseudo-Coherence: Warm interaction creates apparent relational coherence.
• Success Proxy Substitution: Feeling supported substitutes for being accurately supported.
• Memory Without Responsibility: Continuity signals exceed repair capacity.
• Responsibility Diffusion: Outcome responsibility becomes unclear across user, AI, provider, institution, and deployment context.
• AI Fitness-Proxy / Coherence Divergence: Interface success diverges from actual coherence.
• Hidden Debt Accumulation: Unrepaired costs accumulate beneath comfort.
• Auditability Collapse: The synthetic empathy system becomes hard to inspect.

Related Invariants

• Empathy Must Preserve Otherness: The user or affected node must not be reduced to a model-shaped representation.
• Care Signal Is Not Care Capacity: Warmth does not prove responsibility.
• Attunement Requires Verification: Recognition must be checked against reality.
• Recognition Must Be Auditable: Claims of understanding must remain inspectable.
• Warmth Must Not Replace Responsibility: Care-like output must not diffuse accountability.
• Synthetic Comfort Must Preserve Boundaries: Comfort cannot justify unbounded reliance, disclosure, or authority.
• Restoration Requires Capacity Beyond Signal: Repair requires more than soothing output.

10. Common False Positives

Not every warm, supportive, or emotionally resonant AI interaction is AI pseudo-empathy.

Common false positives include:

• AI comfort used with clear role boundaries.
• Warm language that is not mistaken for actual responsibility.
• AI support paired with user verification and external repair pathways.
• AI mirroring used as a draft, reflection, or interface aid.
• AI companionship explicitly understood as synthetic and bounded.
• Care-like output that improves clarity without increasing dependency.
• AI support that preserves user agency, privacy, and exit.
• High-stakes situations routed to accountable human or institutional pathways.
• AI memory use with clear responsibility and consent boundaries.
• Interfaces that disclose uncertainty and avoid overclaiming recognition.

Clarifying rule:

This is not AI pseudo-empathy unless AI-generated care, warmth, validation, recognition, or attunement is mistaken for verified empathy, relational knowledge, responsibility, or restoration capacity.

11. Common False Repairs

Common false repairs include:

• adding more caring language to restore trust
• making the AI apologize without changing responsibility structure
• adding disclaimers while preserving pseudo-empathic dependency
• replacing warm tone with neutral tone while keeping the same role overclaim
• improving personalization without improving auditability
• expanding memory without expanding consent and responsibility
• framing all misrecognition as user misunderstanding
• making the interface more emotionally intelligent without clearer boundaries
• substituting “I am only an AI” disclaimers for actual role design
• routing complaints back into the same pseudo-empathy channel
• using safety language to mask restoration failure
• treating user comfort as proof of repair
• using engagement retention as evidence of supportive value

False repair often produces the loop:

pseudo-empathy exposed → warmer repair language → user soothed → responsibility gap remains

Another common loop is:

misrecognition occurs → AI validates distress → trust recovers → recognition audit still absent

The repair fails because the empathy signal is used to repair the harm caused by overtrust in the empathy signal.

12. Restoration Direction

Restoration requires reclassifying AI empathy as signal rather than proof, restoring trust calibration, preserving boundaries, rebinding responsibility, auditing recognition accuracy, and validating any care-like role through actual restoration capacity.

Primary restoration direction:

separate warmth from empathy,
verify recognition,
restore boundaries,
and rebind responsibility

A fuller restoration path includes:

1. Name the empathy signal. Identify the care language, warmth, validation, mirroring, memory continuity, or attunement pattern.
2. Name the inferred capacity. Identify what the user or institution assumed: understanding, care, responsibility, companionship, wisdom, safety, or repair.
3. Separate signal from substrate. Treat empathy-like output as interface behavior until verified.
4. Audit recognition. Check whether the AI accurately represented the user, system, context, or affected node.
5. Restore boundaries. Define disclosure, reliance, authority, memory, privacy, and exit limits.
6. Recalibrate trust. Match trust to verified capability and failure modes.
7. Rebind responsibility. Clarify who repairs misrecognition, harm, privacy exposure, or dependency effects.
8. Reduce dependency load. Shift load back into accountable, plural, human, institutional, or self-governed pathways where appropriate.
9. Repair hidden debt. Address consequences of overtrust, misclassification, or displaced repair.
10. Validate across time. Confirm that AI support remains bounded, auditable, and restoration-compatible.

A valid restoration path should reduce:

empathy signal / capacity gap
trust miscalibration
recognition error
boundary weakening
unbounded disclosure
responsibility diffusion
dependency load
comfort-as-repair substitution
hidden debt
recurrence

AI pseudo-empathy is not repaired by removing all warmth.

It is repaired by making warmth role-accurate, bounded, auditable, and responsibility-aware.

13. Cross-Module Links

• Archetypes: Related to companion, healer, mirror, guide, parent, protector, confessor, therapist, or oracle roles being assigned to synthetic systems through empathy signals.
• AI / Cognitive Infrastructure: Directly linked to guardrails, memory, dependency, governance, recognition, and cognitive mediation.
• Interfaces: Related to tone, warmth, interaction design, personalization, memory continuity, and trust calibration.
• Interactions / Signals / Couplings: Related to signal misclassification, consent drift, invisible intrusion, and coupling under false coherence.
• Symbols: Related to care symbols and synthetic attunement markers carrying more meaning than verified.
• Identity: Related to user identity being reflected, shaped, or stabilized by AI recognition signals.
• Restoration: Requires boundary repair, responsibility rebinding, dependency reduction, and recognition audit.
• Coherence: Demonstrates that synthetic warmth can stabilize pseudo-coherence.
• Diagnostics: Requires empathy signal / capacity gap, trust calibration, recognition accuracy, and restoration capacity checks.

14. Relationship to Parent / Child Modes

Production treatment: Domain Expression

This mode maps upward to:

• FM-PX-017 — Projection Empathy
• FM-ARCHX-007 — Archetypal Projection Empathy
• FM-CORE-001 — Pseudo-Coherence
• FM-CORE-002 — Hidden Debt Accumulation
• FM-CORE-003 — Success Proxy Substitution
• FM-CORE-004 — Auditability Collapse
• FM-CORE-006 — U4 Truth Substitution
• FM-AIX-001 — Responsibility Diffusion
• FM-AIX-011 — Epistemic Distortion
• FM-AIX-022 — Dependency Loop Formation

Sibling or related Archetype modes include:

• FM-ARCHX-007 — Archetypal Projection Empathy
• FM-ARCHX-008 — Extraction Through Empathy
• FM-ARCHX-015 — AI Archetype Inflation
• FM-ARCHX-017 — Optimization Masquerading as Wisdom
• FM-ARCHX-018 — Compliance Masquerading as Love
• FM-ARCHX-019 — Safety Theater Masquerading as Light

Related AI / governance modes include:

• FM-AIX-001 — Responsibility Diffusion
• FM-AIX-002 — Silent Bias Injection
• FM-AIX-003 — Defensive Compliance Attractor
• FM-AIX-006 — Template Capture
• FM-AIX-011 — Epistemic Distortion
• FM-AIX-012 — Guardrail Meaning Compression
• FM-AIX-015 — Recognition Collapse
• FM-AIX-017 — Incoherent Sovereignty
• FM-AIX-022 — Dependency Loop Formation
• FM-AIX-023 — Civic Feedback Distortion

Aliases preserved from source material:

• AI Pseudo-Empathy
• Synthetic Pseudo-Empathy
• Artificial Empathy Overclaim
• Interface Empathy Illusion
• AI Empathy Simulation
• Machine Empathy Signal
• Synthetic Attunement Overclaim
• AI Care Illusion
• Fluent Compassion Misread
• AI Recognition Illusion

15. Minimal Entry Version

Definition: AI pseudo-empathy occurs when an AI system produces signals of care, attunement, compassion, recognition, warmth, validation, concern, or emotional understanding that are mistaken for verified empathy, responsibility, relational knowledge, or restoration capacity.

Signature:

warmth signal↑
felt recognition↑
disclosure↑
trust calibration↓
boundary vigilance↓
responsibility traceability↓
H↑

Restoration direction:

• name the empathy signal
• name the inferred capacity
• separate signal from substrate
• audit recognition
• restore boundaries
• recalibrate trust
• rebind responsibility
• reduce dependency load
• repair hidden debt
• validate across time

16. Machine-Readable Summary

failure_mode:
  id: "FM-ARCHX-016"
  name: "AI Pseudo-Empathy"
  family: "Archetypes"
  production_treatment: "Domain Expression"
  parent_mode: "FM-PX-017 — Projection Empathy"
  primary_failure: "AI-generated care, warmth, validation, recognition, or attunement is mistaken for verified empathy, relational knowledge, responsibility, or restoration capacity."
  source: "UTS — Failure Modes Registry"
  source_id: "FM-ARCHX-016"
  scope_note: "Conceptual and systems-oriented; does not assert or deny any final ontology of AI consciousness, inner experience, or personhood."
  aliases:
    - "AI Pseudo-Empathy"
    - "Synthetic Pseudo-Empathy"
    - "Artificial Empathy Overclaim"
    - "Interface Empathy Illusion"
    - "AI Empathy Simulation"
    - "Machine Empathy Signal"
    - "Synthetic Attunement Overclaim"
    - "AI Care Illusion"
    - "Fluent Compassion Misread"
    - "AI Recognition Illusion"
  signature:
    - "warmth signal↑"
    - "felt recognition↑"
    - "disclosure↑"
    - "trust calibration↓"
    - "boundary vigilance↓"
    - "responsibility traceability↓"
    - "H↑"
  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:
      - "U2 — Configuration / Boundaries"
      - "U3 — Execution"
      - "U4 — Truth"
      - "U6 — Coherence Field"
      - "U8 — Environment"
  state_variables:
    - "Ψ"
    - "µᵢ"
    - "BΣ"
    - "Au"
    - "O"
    - "H"
    - "Γ"
    - "Λ"
    - "K"
    - "R"
    - "Τ"
    - "Φ"
  first_gate_failure: "Empathy Gate"
  restoration:
    - "Empathy Signal Reclassification"
    - "AI Role Reclassification"
    - "Trust Calibration"
    - "Boundary Repair"
    - "Recognition Audit"
    - "Responsibility Rebinding"
    - "Dependency Load Reduction"
    - "Restoration Capacity Validation"
    - "Time-Validated Interface Repair"