California’s challenging homeowners insurance market is pushing specialty providers to find more precise ways of assessing wildfire exposure. Risk Theory is using property-level AI from ZestyAI to support underwriting through Jupiter Platinum Home, its new excess and surplus (E&S) homeowners programme for high-value properties.
The vertically integrated specialty insurance platform has deployed ZestyAI’s Z-FIRE model as part of the programme, which is aimed at homes that can be difficult to place with admitted carriers or existing E&S providers.
Jupiter Platinum Home covers properties with dwelling limits starting at $750,000 and total insured values of up to $25m. The programme is particularly focused on homes with elevated wildfire exposure, as insurers continue to reassess their appetite for properties in high-risk areas.
Rather than assessing wildfire exposure primarily through broader geographic zones, Risk Theory is using Z-FIRE to provide a more detailed view of individual properties. The model uses machine learning to analyse factors such as defensible space, nearby vegetation, terrain, construction materials and local fire behaviour.
These inputs are combined to assess both the potential wildfire hazard and the characteristics of the property itself, giving underwriters additional information when evaluating risk and determining pricing.
Risk Theory operates a portfolio of managing general agents across specialty insurance lines, with capabilities spanning distribution, underwriting, claims, analytics and compliance. The company focuses on building insurance businesses in areas where technology and underwriting expertise can support more specialised approaches.
ZestyAI provides risk decision technology for property and casualty insurers, reinsurers, brokers and regulators across the US. Its platform combines property-level data with predictive AI across several perils, including wildfire, severe convective storms and non-weather water.
Z-FIRE has received approval across every Western wildfire market and was the first AI-based wildfire model accepted as part of a carrier rate filing in California. ZestyAI said its broader portfolio of models has received more than 200 regulatory approvals across the US.
Risk Theory executive vice president of underwriting Jessica Furrow-Young said, “Jupiter Platinum Home was built for risks that are increasingly difficult to place in California, including homes with elevated wildfire exposure.”
“For us, that means understanding the differences between individual properties rather than treating wildfire exposure as a simple yes-or-no decision,” she added.
Furrow-Young said Z-FIRE provides underwriters with insight into the hazards around a property as well as the structural characteristics that could affect how an individual home performs.
For ZestyAI founder and CEO Attila Toth, more precise risk differentiation will be important as insurers look to expand capacity in California.
“Risk Theory built property-level risk selection into its underwriting approach from the outset,” he said. “That level of insight can help insurers approach even challenging wildfire exposures with greater confidence and discipline.”
The deployment highlights the growing use of property-level data and AI in catastrophe underwriting, as insurers and specialty providers seek more detailed ways to evaluate risks in markets facing pressure from wildfire exposure.
Copyright © 2026 InsurTech Analyst
