Machine jurisdiction
What the system may describe is not what the institution may use
The Jurisdiction Matrix separates technical access from legitimate institutional authority across thirteen roles.
Research framework / scenario analysisFact / scenario boundaries enforcedv0.14.0
Answer first
What is the governing principle?
TECHNICAL ACCESS ≠ LEGITIMATE JURISDICTION. Observe, infer, retain, share and act are separate permissions. Data that is useful inside one relationship does not automatically become evidence everywhere else.
Jurisdiction matrix
Five separate permissions
| Role | May observe | May infer | May retain | May share | May act |
|---|---|---|---|---|---|
| AI confidant | Conversation necessary to answer the user; acute emergency signal only under defined policy | Intent/character only for immediate response context; no general civic character inference | User continuity data only as chosen; emergency processing should minimize retained risk metadata | No generalized export; external access only under defined lawful process/emergency exception | Serve user; voluntary support; narrowly defined emergency response |
| Employer | Job-relevant facts | Only role-relevant inferences with validated purpose | Minimal employment record with expiration | No unrelated private-life export/import | Employment decision subject to notice, reason and contest where consequential |
| School | Educational/safety facts | No lifelong character inference from adolescent behavior | Short-lived, developmental records | No automatic export into employment/criminal risk | Education and specific campus safety action |
| Insurer | Lawfully obtained contract-relevant data | Only risk relevant to covered product | As required for defined insurance purpose | No intimate or political data absent direct lawful relevance | Price/coverage decisions with reason and contest rules |
| Bank / lender | Financial facts and fraud signals | Credit/fraud inference only within financial context | Purpose-limited financial records | No general morality or political profile | Transaction security and credit decisions with specific adverse reasons |
| Doctor / health system | Health data needed for care | Clinical inference within care relationship | Protected medical record | No unrelated employer/credit export absent lawful basis | Treatment and emergency care, not civic scoring |
| Vehicle / mobility system | Speed, control inputs, bounded impairment signal | Immediate driving safety only | Prefer ephemeral safety telemetry | No generalized character score export | Temporary vehicle-specific safety friction |
| Police / public safety | Observed conduct and authorized evidence | Risk hypotheses cannot become guilt | Case-bounded records under law | No dragnet civic-risk vector | Investigate specific conduct; emergency action on concrete imminent harm |
| Court / tribunal | Evidence properly before tribunal | Legal inferences under transparent adjudicative rules | Official record with correction/appeal provenance | Only as authorized by law | Adjudication, not generalized life optimization |
| Border agency | Border-relevant facts | No political/sexual/private-life character inference absent lawful individualized nexus | Purpose-limited travel/security record | No unrelated downstream social score | Screening proportional to lawful security mandate |
| Platform | Content and service-integrity data | Moderation/safety inference within platform context | Minimum service record | No civic-risk export to employers/insurers/government absent lawful process | Moderation and service safety, not general moral judgment |
| Benefits agency | Program-specific eligibility data | Eligibility/fraud inference limited to program | Program record with correction history | No unrelated employment/policing export | Eligibility and bounded fraud review with due process |
| Personal wearable | Sensor data needed for user-selected feature | Local inference for user benefit | Local/ephemeral by default | No secondary risk brokerage without meaningful consent/lawful basis | Inform the user; emergency feature only if explicitly scoped |
Judgment-free total cognitive freedom
NO JUDGMENT WHATSOEVER. Concresca coordinates without assigning moral worth, character, guilt, danger, trustworthiness, loyalty, purity, normality, or social standing. Questions, thoughts, identities, messages, content, and conduct are not objects of Concresca judgment.