Context & Business Challenges
Underwriting processes in property and marine insurance require consolidating multiple sources of information: expert reports, claims history, client data, local regulations, and historical cases, often distributed across disconnected systems.
The challenge was to accelerate underwriting, improve risk assessment accuracy, and strengthen underwriting expertise, while ensuring consistent decision-making across teams and geographies.
Unified data foundation
All relevant data, reports, claims history, internal and external documents, is centralized into a single, structured view. Each case becomes a coherent and actionable underwriting object.
AI that understands cases
NLP models analyze documents, extract key signals, and connect them with similar historical cases. The system moves from information retrieval to risk understanding.
Underwriters at the center
AI provides signals, comparisons, and alerts but decision-making remains human. Each decision feeds back into the system, continuously improving screening, evaluation, and internal rules.
Operational outcomes
- Faster underwriting decisions + 40%
- More consistent and fair pricing
- Stronger underwriting expertise across teams