Swiss Re: Underwriting Is Now Continuous

The push towards continuous underwriting is a direct response to longstanding industry friction. Underwriters historically spent 30-40% of their time on manual, administrative tasks like re-keying data from disconnected legacy systems. This bottleneck creates delays and inconsistencies, issues that can be addressed by automating data ingestion and analysis. At Swiss Re, the strategy is to augment, not replace, human expertise. Group Chief Economist Jérôme Haegeli notes that AI will help "understand risks better, price risks better and more accurately," increasing the value of underwriting in a complex market. The goal is to use technology to make interfaces more efficient, freeing underwriters from manual work to apply their judgment to the most complex risks. To enable this, Swiss Re has developed platforms like the Magnum XP Underwriting Assistant, described as a "second brain" for underwriters. This AI-powered tool extracts and organizes key applicant data from disparate sources, flagging information gaps and allowing underwriters to make faster, more consistent decisions on complex cases that automated systems alone cannot handle. Building these systems requires a fundamental shift in data infrastructure. Insurers are migrating from legacy environments to modern data platforms, with Snowflake providing scalable cloud-based storage and compute. Transformation tools like dbt are being used to build reliable, version-controlled data models, allowing actuarial, claims, and medical departments to make better decisions from a single source of truth. This technological evolution directly impacts how actuaries model risk. While the fundamental principles of risk classification are unchanged, actuarial professional bodies are adapting their standards for an era of continuous risk evaluation. The focus is on integrating new data sources and complex models into Enterprise Risk Management (ERM) frameworks, combining quantitative analysis with expert oversight to maintain discipline. The shift is not without challenges, as data quality and integration with legacy systems remain significant hurdles for many carriers. Successfully implementing continuous underwriting requires overcoming cultural resistance to change and investing in a data-literate workforce capable of managing the entire machine learning lifecycle.