Reinsurer SCOR warns against abandoning GenAI too soon
Global reinsurer SCOR released a report on generative AI adoption in life and health insurance, cautioning that organizations may scrap initiatives before long-term benefits materialize. The report draws on SCOR's internal deployment experience and a February 2026 industry survey.
Eighty percent of life and health insurers have already deployed GenAI in at least one core function, according to the Gallagher survey cited in the report. The technology has moved beyond pilot phase into standard operations across the industry.
Four primary use cases gaining traction
SCOR identified the following applications:
- AI-powered customer engagement tools
- Automated summarization of unstructured data
- Augmented underwriting and claims support
- Advanced data analytics
Streamlining document-intensive underwriting and claims processes ranks as SCOR's highest priority. Medical record review typically consumes several hours to several days of staff time per case.
SCOR's proprietary AI Assistant tool, deployed to internal users starting in 2023, now supports more than 100 underwriting and claims professionals across major markets. The tool processes over one million document pages monthly.
Early gains mask deeper challenges
SCOR identified four obstacles that insurers should prepare for. In early pilots, the AI Assistant produced an unexpected result: case management times initially increased rather than decreased.
Staff spent additional hours verifying AI-generated outputs against source documents. The report warned that return on investment may take longer to materialize than expected, creating pressure to abandon initiatives before benefits emerge.
Legacy systems present another barrier. Data trapped in incompatible databases prevents the information flow needed for AI to reach full efficiency.
Regulatory compliance frameworks including the EU AI Act, GDPR, and HIPAA impose strict requirements around explainability, data protection, and non-discrimination. These obligations can delay scaling decisions.
The compliance concern has basis in practice. Separate research from MoneyGeek found that nearly one in three health insurers do not test their AI models for racial bias, according to the National Association of Insurance Commissioners.
Learn more about AI for Insurance and Generative AI and LLM applications in your organization.
Your membership also unlocks:
