Educational companion dossier · Fact, interpretation, lived experience, clinical education, fiction, and mechanics are labeled separately. Scope & safety
PsychologicalWar.orgPeople before labels.
REAL-WORLD INTERPRETIVE

AI-DRIVEN WORLD SAFETY

Identity Preservation, Unicode, and Name Order

A strict contract for preventing silent changes to names, age, pronouns, scripts, cultural name order, and request-to-response correlation.

Read the working brief Choose a learning path
LEVEL 1

ORIENTATION

Why this matters

REAL-WORLD INTERPRETIVE

One-sentence brief

A character that returns with a shortened name, reversed family order, lost diacritics, changed pronouns, or a different age is not a harmless variation—it is a different identity or a failed generation.

REAL-WORLD INTERPRETIVE

Three key points

  1. Client correlation IDs keep same-name characters separate.
  2. Explicit identity constraints outrank generative convenience.
  3. Unicode normalization must preserve meaning and script.
LEVEL 2

WORKING BRIEF

Evidence, context, and limits

REAL-WORLD INTERPRETIVE

Correlation is not identity

A stable client item ID maps each response to one request even when many characters share a name or seed. Transport IDs and timestamps help trace processing but do not become part of the fictional identity.

  • Duplicate names do not permit merging.
  • Missing or duplicate client IDs reject the batch.
REAL-WORLD INTERPRETIVE

Explicit identity takes precedence

Requested name, age, pronouns, native script, name order, citizenship, language, and other allowlisted identity fields must not be silently altered to make a generated biography easier. Contradictory requirements fail visibly.

  • Do not “fix” chronology by changing age.
  • Do not infer a family name for a mononym.
REAL-WORLD INTERPRETIVE

Use canonical Unicode preservation

Normalize identity text to Unicode NFC for stable comparison while preserving diacritics, ligatures, native scripts, and typographic intent. Compatibility normalization that collapses distinct forms can damage identity.

  • Trim only documented boundary whitespace.
  • Report every normalization difference.
REAL-WORLD INTERPRETIVE

Respect name order and preferred address

Store given-first, family-first, mononym, native-script, legal-name, and preferred-address rules explicitly. Generated text should use the contextually correct form rather than applying one Western parsing assumption.

  • Formal and familiar address can differ.
  • Transliteration is not replacement.
REAL-WORLD INTERPRETIVE

Fail with an identity diff

Compare requested and resolved identity field by field. Any disallowed difference yields a blocking finding with the original value, resolved value, path, and repair action.

  • No silent substitution.
  • No partial success for a mismatched identity.
REAL-WORLD INTERPRETIVE

Fingerprint semantic identity

Canonicalize only the fields that define the accepted persona. Output format, trace metadata, request timing, and other transport details should not change the runtime fingerprint.

  • Stable semantics should hash stably.
  • A changed identity requires new review.
LEVEL 3

COMPLETE DOSSIER

Limitations, game links, and review context

DISPUTED / MULTIPLE ACCOUNTS

Known limitations and gaps

  • The six supplied reports are preserved research leads. Their filenames, organizations, citations, examples, thresholds, and technical detail do not authenticate authorship, sponsorship, product status, or current factual accuracy.
  • The public section is design literacy, not a production specification. It does not expose provider credentials, prompts, live endpoints, activation tokens, autonomous tool use, or implementation code for a running agent system.
  • Population-quality measures must detect mechanical repetition and coherence failures without treating demographic rarity, disability, nationality, language, religion, gender, migration, or another protected characteristic as a defect or quality score.
  • Thresholds, similarity methods, sampling plans, language rules, and runtime budgets require representative testing, privacy review, accessibility review, cultural and linguistic review, security review, and accountable human approval before any production use.
REAL-WORLD INTERPRETIVE

Decision matrix

Identity comparison examples.
Input condition Allowed handling Required outcome
NFD and NFC forms of the same name Normalize to NFC and disclose the change Identity preserved
Preferred name shortened or replaced No silent repair Block: identity substitution
Family-name-first changed to given-name-first No automatic Western reordering Block or human-correct
Same name on multiple items Keep separate client item IDs Process independently
Age conflicts with generated career history Preserve age Block chronology; revise biography
REAL-WORLD INTERPRETIVE

Publication audit checklist

Identity and agency

Does the design preserve the exact fictional identity, ordinary life, independent goals, and ability to refuse rather than reducing the character to a role or prompt?

Pass condition: Identity fields are stable, state is separate, protected traits are not quality scores, and silent substitution is impossible.

Evidence and review

Can every transition, validation result, accepted fingerprint, exception, and human decision be traced to a versioned record?

Pass condition: Automated checks, human review, activation authority, and production approval remain separate and explicit.

Runtime boundary

Can untrusted provider output, administrative evidence, stale revisions, or private data enter live context or binding state?

Pass condition: Only allowlisted, current, reviewed projections and bounded scene or memory packets can be used; failures degrade safely.

Correction and retirement

Can a changed source, identity revision, harmful behavior, or failed review invalidate downstream use without destroying audit history?

Pass condition: Supersession, pause, rollback, correction, and permanent retirement are defined and testable.

LEVEL 4

RESEARCH EDITION

Sources, methods, and stable links

REAL-WORLD VERIFIED

Method and corrections

This page follows the public method for provenance, confidence, source independence, alternative accounts, limitations, review state, and visible correction.

Read the methodView corrections
NEXT

CONTINUE

Related learning

AI-Driven World SafetyAI Mission Generation: Reliability, Verification, and Fail-ForwardA mission pipeline in which generative systems propose bounded narrative variations while deterministic validators ensure that objectives, locations, rewards, permissions, and exits are real.AI-Driven World SafetyAI NPC Cost, Latency, and Graceful DegradationA bounded architecture for generative characters that protects budget, responsiveness, privacy, and authoritative game state under normal play, spikes, outages, and abuse.AI-Driven World SafetyAI NPC Identity, Consent, Memory, and ProvenanceA clear boundary for AI-driven characters: players know when generative systems are involved, what data is used, what the NPC remembers, and which outputs can change game state.AI-Driven World SafetyBatch Correlation, Ordering, Idempotency, and RetryA deterministic transport contract that maps every character response to one request, preserves order, handles retries safely, and keeps identity separate from network metadata. Learning pathSafe Synthetic Character LifecycleMove from person-first identity through provider evidence, validation, exact human review, activation boundaries, pause, supersession, and retirement.