AI will change how property/casualty insurers model climate risk

• The severity of the climate crisis is at its highest level in seven years
• The recently reduced frequency of danger will not continue.
• AI models are improving the assessment of visual real estate risk information

The growing severity and frequency of the climate crisis is forcing insurers to apply AI and other new technologies to risk modeling and mitigation, according to data analytics providers.
The severity of all hazards has increased significantly and is the highest in seven years, an increase of almost 26% from 2024 and an increase of more than 93% compared to 2019. LexisNexis Latest Housing Trends Report.

Processing details

Analysts at LexisNexis note that the cost of loss for all perils decreased by more than 4% in 2025, and the frequency of occurrences decreased by nearly 24% year over year. The decline in billing frequency should not be expected to continue.as a number, and Severity of billion-dollar weather events and storms rise.

George Hosfield, vice president and general manager of home insurance at LexisNexis Risk Solutions, said that despite the overall upward trend, the key to containing risk costs is how carriers apply technology to the problem. Mr Hosfield said that as risks fluctuate seasonally and annually, it is essential for airlines to understand both per-risk and macro-level trends through more granular real estate intelligence and technology.

AI model According to Hosfield and Kelly Rush, director of home insurance at LexisNexis Risk Solutions, the use of visual language is improving digital self-examination tools for assessing home preparedness and risk enhancement measures. These models interpret image-based information and apply it to risk assessment.

“Technology is increasingly being used to identify property-level problems before losses occur,” Hosfield and Rush said in an emailed response to questions. “Digital self-inspections give insurers a more cost-effective way to gather detailed information about a property’s condition and surrounding risk factors.”

AI is More suitable for evaluating visual information about property risksAccording to Chris Lucas, principal machine learning engineer at Fathom, a climate risk data provider specializing in flood data.

“Machine learning AI is very good at finding patterns that we can’t find,” he said. “If you give it lots of examples of terrain, it will be able to tell the differences in ways that we couldn’t. This will allow us to zoom in on areas and create much higher-resolution simulations.”