Munich Re engineers risk models with AI
Reinsurance firm Munich Re is re-engineering its risk operations by combining AI with climate science. The 140-year-old company is transitioning to a tech-driven model, using advanced analytics to ingest real-time climate data and underwrite emerging threats like wildfires and AI-related losses. This requires platforms capable of processing vast, heterogeneous data streams.
- To assess storm damage, Munich Re uses a solution called CatAI which analyzes high-resolution aerial imagery to detect and classify property-specific roof damage, often before the policyholder files a claim. The company worked with Google researchers to adapt visual inspection AI, originally developed for manufacturing assembly lines, to accelerate the deployment of these catastrophe models. - The company provides its machine learning models for damage analytics to the Geospatial Intelligence Center (GIC), an insurance industry consortium. In a pilot program during Hurricane Michael, this remote sensing technology delivered damage assessments for over 40% of property claims before the insurer received the first notice of loss. - Munich Re's NATHAN Risk Suite includes specific "climate change modules" that allow clients to model how the risks of natural catastrophes may shift under various climate scenarios through the year 2100. This platform integrates the firm's natural disaster database with clients' own data to visualize and evaluate physical risk exposure globally. - The company offers a specific insurance product called aiSure™ that provides coverage for the performance of other companies' AI models, covering financial losses that result from issues like prediction errors or data drift. One application is providing reinsurance for the AI-powered identity fraud detection service from the company Instnt. - Expanding beyond climate and property risk, Munich Re has partnered with Spectrum Labs, a company that uses contextual AI to identify toxic online behavior. Munich Re provides a performance guarantee for Spectrum's AI, signaling trust in the model's ability to make online platforms safer. - As part of its technology-driven focus, the company created the new board-level position of Chief Technology Officer, appointing Robin Johnson to the role effective August 1, 2025. This is part of a broader leadership transition that will see Chief Financial Officer Dr. Christoph Jurecka become Chair of the Board of Management on January 1, 2026. - The company's Tech Trend Radar 2025 report highlights "spatial intelligence"—fueled by satellite data, AI, and geospatial analytics—as a key technology for enabling more accurate risk assessments in catastrophe-prone areas. This builds on its established use of remote sensing data from planes, satellites, and drones to feed its AI engines. - Through its innovation arm, Munich Re has also partnered with Metabiota, a California startup that models the risk of epidemics. This allows the company to develop new insurance products for risks that were previously considered uninsurable, such as the spread of infectious diseases.