No off-platform movement
AI replies are checked for platform/payment safety before sending.
Trust + safety
Aphrodite separates creator brand voice, revenue generation, policy safety, and human review so the system can scale without losing control.
AI replies are checked for platform/payment safety before sending.
Location, personal schedule, family, identity, and sensitive details are blocked or escalated.
Refunds, complaints, chargebacks, legal risk, distress, and policy edge cases route to humans.
Admin actions and approval-sensitive events are designed to be recorded, not silently overwritten.
Admin controls
The schema already includes role/permission scaffolding and blocks self-approval for payout, credit, bonus, adjustment, and reversal flows.
Creator controls
Persona cards, approved examples, and explicit boundaries define what automation can say — and what it must never say.