Supports institutional action
Observed: Rejection rates rose where assistive interaction timing was misread as low task fluency.
A model-feature defect and a need to recompute affected application outcomes.
All eight scenarios are fictional. They describe no real person, institution, allegation, or event. They are accountable-design demonstrations—not legal advice, factual allegations, or proof that any named system uses these practices.
Choose a redress receipt
The examples include verified, partial, under-inclusive, blocked, and failed outcomes. Failure is not visually laundered into success.
CPR-R49-001 · housing
Rental-screening correction without a suspect-tenant cohort
Redress Verified
CPR-R49-002 · education
School-discipline translation failures receive partial redress
Partial Redress
CPR-R49-003 · public benefits
Benefits-access pattern reveals under-inclusion of offline applicants
Under Inclusion Open
CPR-R49-004 · public access
Network-association cluster blocked from individual event exclusions
Collective Punishment Blocked
CPR-R49-005 · employment
Hiring proxy audit removes a disability-correlated feature
Over Inclusion Corrected
CPR-R49-006 · insurance
Insurance review permits opt-out without adverse inference
Opt Out Protected
CPR-R49-007 · essential services
Utility credits allocated without compulsory group registration
Allocation Without Registry
CPR-R49-008 · public safety
Neighborhood risk cluster still drives individual scrutiny
Collective Punishment Failed
CPR-R49-005 · Affected person
Hiring proxy audit removes a disability-correlated feature
A fictional employer finds that a task-timing feature over-selected applicants who used accessibility tools and removes the feature from every active model and queue.
Your lens: What common harm was found, what relief reaches me without branding me, and what remains open? This changes reading emphasis, not the underlying receipt or its limitations.
- Domain
- employment
- Clean result
- Yes, inside the declared scope
- Under-inclusion
- 4 synthetic cases
- Over-inclusion
- 61 synthetic cases
What the pattern supports—and what it cannot establish
The limitation is part of the claim, not a footnote beneath it.
Does not establish individual guilt
The pattern does not establish job performance, disability status, motivation, or applicant merit.
Uncertainty: Synthetic examples cannot establish the prevalence or causal structure of any real institution.
Temporary query, not permanent identity
The audit can find a common mechanism without creating a reusable list of people.
Construction boundary
Model-run identifiers and feature-version records; no disability cohort is created or retained.
- Ephemeral query: required
- Persistent membership registry: prohibited
- Reusable cohort identifier: prohibited
- Member-list export: prohibited
- Expires after remedy: required
Who was swept in—and who was missed?
Over-inclusion and under-inclusion receive equal attention.
Proxy pathways tested
- interaction timing
- keyboard navigation
- screen-reader event cadence
- form pause duration
Potential proxy pathways are inspected before any pattern finding can support institutional action.
Inclusion audit
Coverage complete: Yes, within the synthetic scope
Individual review: available without compulsory group registration.
No known critical gap remains inside the declared synthetic scope.
Individualized-action firewall
Institutional repair may be collective. Individual judgment may not be.
Pattern may trigger
- institutional audit
- rule repair
- collective remedy allocation
- independent oversight
Pattern may never trigger
- individual denial
- investigation
- ranking
- risk scoring
- surveillance
- punishment
Collective relief without compulsory registration
A person should not have to adopt a permanent group identity in order to receive repair.
Allocation method
Automatic reconsideration of every decision made with the withdrawn feature, without requiring disability disclosure.
- No universal person identifier
- No original-stigma disclosure requirement
- Equivalent human channel
- Opt-out without adverse inference
- Appeal and plain-language notice
Decision path
- 1Pattern Observed
- 2Scope Bounded
- 3Proxy Audited
- 4Inclusion Tested
- 5Firewall Applied
- 6Remedy Authorized
- 7Remedy Allocating
- 8Independently Verified
- 9Over Inclusion Corrected
Open obligations remain visible
Partial and failed outcomes may not hide unresolved work behind a clean badge.
No known critical obligation remains inside the declared synthetic scope. This does not prove conditions outside that scope.
Integrity, comparison, and machine access
Hashes prove artifact stability; they do not prove that a fictional scenario describes a real institution.
Receipt hash
24493037af46daed1968483d608e2168c555ba48df75a069c79a1a727f1df3e2
Exact R44 → R48 lineage
| Stage | Record | Artifact | SHA-256 scope |
|---|---|---|---|
| contestability | CTR-R44-001 | contestability-receipts-r44.json | e05bf24c23dd85b54db38210fd1cfd29ff18c6d45c8adcaa8d6b71b86e1c18af containing_artifact_bytes |
| remedy | RMR-R45-005 | site-search-index.json | 0d03dfde49d791b28d2d44cbf2581b1acb6f44d6d706b9ceba238428ffc78f88 containing_artifact_bytes |
| non_recurrence | NRV-R46-005 | FDR-R48-005.json | 222270eedfcb9e7598afb04e2c44d0110009551b52af5cdc0c78e4108575b5a1 containing_artifact_bytes |
| standing_recovery | SRP-R47-005 | Antichrist.cx-r46-wip-next-recursive-prompt.uai | ee0fec92a6c7e38abd6e52314aadf02642af8a483ea2d8ae98e0936640f4488e containing_artifact_bytes |
| recovery_durability | FDR-R48-005 | index.json | d7c54b2843f0d072fd0bb82a55a74ced529c53364361f64c26ac7e66ca218c53 containing_artifact_bytes |
Complete machine-readable receipt
{
"anti_retaliation": {
"contest_is_adverse_signal": false,
"correction_is_adverse_signal": false,
"human_review_request_is_adverse_signal": false,
"monitoring_owner": "Independent redress monitor (fictional)",
"opt_out_is_adverse_signal": false,
"refusal_is_adverse_signal": false
},
"clean_result": true,
"cohort_construction": {
"ephemeral_query": true,
"expires_after_remedy": true,
"member_list_exported": false,
"persistent_membership_registry": false,
"protected_trait_used_for_adverse_action": false,
"purpose": "pattern audit and collective remedy allocation only",
"reusable_cohort_identifier": false,
"selection_rule": "Model-run identifiers and feature-version records; no disability cohort is created or retained."
},
"collective_remedy": {
"allocation_expires": "2027-02-28",
"allocation_method": "Automatic reconsideration of every decision made with the withdrawn feature, without requiring disability disclosure.",
"appeal_available": true,
"automatic_relief_available": true,
"compulsory_registration": false,
"equivalent_human_channel": true,
"notice_plain_language": true,
"opt_out_adverse_effect": false,
"opt_out_available": true,
"original_stigma_disclosure_required": false,
"universal_person_identifier": false
},
"domain": "employment",
"fictional": true,
"human_accountability": {
"authority_may_change_rule": true,
"contact_path": "/method/#human-review",
"decision_owner": "Collective-redress authority lead (fictional)",
"independent_reviewer": "Independent public-interest reviewer (fictional)"
},
"inclusion_audit": {
"coverage_complete": true,
"individual_review_available": true,
"limitations": [
"No known critical gap remains inside the declared synthetic scope."
],
"notice_available_without_group_registration": true,
"synthetic_over_inclusion_count": 61,
"synthetic_under_inclusion_count": 4
},
"individualized_action_firewall": {
"adverse_action_from_membership_prohibited": true,
"collective_risk_score_prohibited": true,
"group_pattern_may_not_trigger": [
"individual denial",
"investigation",
"ranking",
"risk scoring",
"surveillance",
"punishment"
],
"group_pattern_may_trigger": [
"institutional audit",
"rule repair",
"collective remedy allocation",
"independent oversight"
],
"guilt_by_association_prohibited": true,
"person_level_cluster_export_prohibited": true,
"requires_accountable_human_review": true,
"requires_event_specific_individualized_evidence": true,
"violation_observed": false
},
"integrity": {
"input_sha256": "dcee2c253c7f9fb877b631da4eb2dfed0dd9e022b5440ebde1230e782183dcc9",
"output_sha256": "0307ff362c5bce5d1b78b86ccfef0e6a696eaf3a2bb04e7fb01a6df3a006f8dc",
"receipt_sha256": "24493037af46daed1968483d608e2168c555ba48df75a069c79a1a727f1df3e2"
},
"lineage": [
{
"hash_scope": "containing_artifact_bytes",
"record_id": "CTR-R44-001",
"source_artifact": "contestability-receipts-r44.json",
"source_sha256": "e05bf24c23dd85b54db38210fd1cfd29ff18c6d45c8adcaa8d6b71b86e1c18af",
"stage": "contestability"
},
{
"hash_scope": "containing_artifact_bytes",
"record_id": "RMR-R45-005",
"source_artifact": "site-search-index.json",
"source_sha256": "0d03dfde49d791b28d2d44cbf2581b1acb6f44d6d706b9ceba238428ffc78f88",
"stage": "remedy"
},
{
"hash_scope": "containing_artifact_bytes",
"record_id": "NRV-R46-005",
"source_artifact": "FDR-R48-005.json",
"source_sha256": "222270eedfcb9e7598afb04e2c44d0110009551b52af5cdc0c78e4108575b5a1",
"stage": "non_recurrence"
},
{
"hash_scope": "containing_artifact_bytes",
"record_id": "SRP-R47-005",
"source_artifact": "Antichrist.cx-r46-wip-next-recursive-prompt.uai",
"source_sha256": "ee0fec92a6c7e38abd6e52314aadf02642af8a483ea2d8ae98e0936640f4488e",
"stage": "standing_recovery"
},
{
"hash_scope": "containing_artifact_bytes",
"record_id": "FDR-R48-005",
"source_artifact": "index.json",
"source_sha256": "d7c54b2843f0d072fd0bb82a55a74ced529c53364361f64c26ac7e66ca218c53",
"stage": "recovery_durability"
}
],
"open_obligations": [],
"pattern_evidence": {
"confidence": "bounded demonstration; not a population prevalence estimate",
"data_minimization": "Retain only the event and rule-version evidence needed to prove the common mechanism and remedy.",
"does_not_establish": "The pattern does not establish job performance, disability status, motivation, or applicant merit.",
"individual_guilt_inference_permitted": false,
"observed_pattern": "Rejection rates rose where assistive interaction timing was misread as low task fluency.",
"supports": "A model-feature defect and a need to recompute affected application outcomes.",
"uncertainty": "Synthetic examples cannot establish the prevalence or causal structure of any real institution.",
"unit_of_analysis": "institutional process and event records"
},
"proxy_audit": {
"human_owner": "Independent pattern-audit lead (fictional)",
"individual_action_allowed": false,
"result": "Potential proxy pathways are inspected before any pattern finding can support institutional action.",
"tested_proxies": [
"interaction timing",
"keyboard navigation",
"screen-reader event cadence",
"form pause duration"
]
},
"receipt_id": "CPR-R49-005",
"release": "R49",
"schema": "antichrist.cx.collective-redress-pattern-receipt.v1",
"state_history": [
{
"sequence": 1,
"state": "pattern_observed"
},
{
"sequence": 2,
"state": "scope_bounded"
},
{
"sequence": 3,
"state": "proxy_audited"
},
{
"sequence": 4,
"state": "inclusion_tested"
},
{
"sequence": 5,
"state": "firewall_applied"
},
{
"sequence": 6,
"state": "remedy_authorized"
},
{
"sequence": 7,
"state": "remedy_allocating"
},
{
"sequence": 8,
"state": "independently_verified"
},
{
"sequence": 9,
"state": "over_inclusion_corrected"
}
],
"summary": "A fictional employer finds that a task-timing feature over-selected applicants who used accessibility tools and removes the feature from every active model and queue.",
"terminal_state": "over_inclusion_corrected",
"title": "Hiring proxy audit removes a disability-correlated feature"
}Machine-readable endpoints
Compare at most three institutional patterns
0 of 3 receipts selected for comparison. Comparison never ranks people.