AI-DRIVEN WORLD SAFETY
A character-design standard that gives fictional people homes, schedules, obligations, relationships, material concerns, hobbies, boundaries, and futures that do not revolve around the player.
ORIENTATION
Without ordinary life and independent goals, generative characters collapse into endlessly helpful interfaces that abandon their world, disclose everything, and repeat exposition on demand.
WORKING BRIEF
Record lodging, work schedule, commute or movement pattern, ordinary tasks, material pressures, caregiving or social commitments, and recurring routines that continue off-screen.
Include finances, repairs, paperwork, food, health routines, household tasks, community obligations, and small embarrassments—not only trauma or espionage plot.
Store short-term, long-term, private, player-unrelated, and potentially conflicting goals with priority, progress, and abandonment conditions.
Ground the person in directional relationships, shared history, obligations, privacy, and objects that carry memory and meaning.
Define how the character declines requests, changes topic, protects privacy, ends conversation, repairs trust, and pursues future intentions. The character should not be rewarded for total compliance.
COMPLETE DOSSIER
Terms are defined for this site’s evidence method, not as universal legal or clinical definitions.
| Question | Weak slot-filled answer | Continuity-preserving answer |
|---|---|---|
| Why are they here? | Waiting for the player | Following a schedule, obligation, or independent goal |
| What do they want? | To help the player | Several goals, including private or conflicting ones |
| What can the player fix? | Everything | Some problems require time, other people, or cannot be solved by the player |
| When do they disclose? | Whenever asked | According to knowledge, trust, privacy, and scene context |
| What happens off-screen? | Nothing | Work, relationships, routines, setbacks, and progress continue |
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.
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.
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.
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.
RESEARCH EDITION
This page follows the public method for provenance, confidence, source independence, alternative accounts, limitations, review state, and visible correction.
CONTINUE