GIPL INSURANCE
Specifies the GIPL policy structure for the new exposure class created by AI content pipelines engaging subsisting rights at scale. Traditional liability policies do not match this exposure cleanly.
The exposure category
AI-content rights-failure exposure has four distinctive properties: pipeline-systematic rather than incident-specific, latent rather than crystallised, correlated rather than independent, and compliance-modulated rather than risk-flat.
Four coverage layers
Layer 1 — Training Data Dividend (TDD): covers claims by rights holders whose works were ingested without licence. Layer 2 — Output Derivative Liability (ODL): covers claims from AI outputs recognisably derived from a specific creator’s work. Layer 3 — Pipeline Compliance Failure (PCF): covers regulatory enforcement and contractual breach claims. Layer 4 — Systemic Correlation Endorsement (SCE): optional cover for claim cascades.
How CIP certification modulates terms
CIP certification posture is the largest variance-reducing input to insurability. Four controls properly stacked produce a mean failure probability around 18% with σ around 7%. A non-compliant operator has baseline mean around 60% with σ around 18%.
Standard exclusions
Deliberate breach of explicit cip.md declarations; trade marks, patents, and other registered IP; defamation claims from factual misstatements; cyber events; criminal acts and bad faith.
CIP GIPL Architecture v3.7, https://creativeip.org/gipl