Life insurance underwriting has long relied on manual review for a significant share of applications — particularly those involving complex medical histories, financial disclosures, or multi-source documentation. In the US market, even sophisticated automated triage systems refer a substantial portion of cases to human underwriters, who must reconcile data from health records, lab results, physician statements, motor vehicle reports, and prescription histories spread across disparate systems. For reinsurers like Swiss Re, this inefficiency compounds across thousands of ceded policies, creating bottlenecks that slow policy issuance, increase operational costs for cedants, and limit underwriter capacity for genuinely complex risk assessment. The status quo imposes a measurable productivity ceiling on life insurance operations at scale.
Swiss Re developed Underwriting Ease, a platform that combines OCR, NLP, and large language models to address the data aggregation problem at the core of manual underwriting. When a case is referred from an automated triage system, the platform extracts and standardizes information from application disclosures, motor vehicle reports, electronic health records, prescription histories, and lab results, then surfaces everything in a unified dashboard. NLP models parse unstructured clinical and financial documents, reducing the time underwriters spend locating and interpreting source data. The system integrates with Swiss Re's proprietary Life Guide underwriting manual and can be configured to work with other insurer-specific guidelines, enabling flexible deployment across different cedant environments. The platform launched in North America and is being evaluated via proof-of-concept programs in the UK, France, Germany, and select Asian markets.
Early production results from SBLI, an early adopter of the platform, demonstrate meaningful efficiency gains across the manual underwriting workflow:
Beyond throughput, the platform delivers a structural benefit for underwriting teams: by presenting data in a consistent, standardized interface, it accelerates skill development for new underwriters, shortening the time required to reach proficiency. Usage data generated through the platform also creates feedback loops that can improve the accuracy of upstream automated decision models over time.
Have a similar implementation?
Share your customer's AI results and link it to your vendor profile.
Submit a case study →