AI Model Z-WATER™ Approved for Non-Weather Water Risk in Homeowners Insurance

ZestyAI's AI-driven model, Z-WATER™, has been approved for underwriting and rating in Illinois, Indiana, Iowa, Louisiana, and Wisconsin, enabling insurers to better address non-weather water risks in homeowners insurance. Non-weather water, including losses from burst pipes and leaks, has become the fourth-costliest peril, with claim severity increasing 80% over the past decade, surpassing losses caused by hurricanes. Traditional rating tools have struggled to accurately model these risks due to reliance on broad territory or age-based data. Z-WATER™ utilizes verified insurer loss data combined with computer vision from aerial imagery and incorporates detailed property-level data, permitting history, local climatology, and infrastructure context. This approach allows insurers to more accurately predict the frequency and severity of interior water losses, improving pricing accuracy and portfolio management. The model's predictive accuracy is up to 18 times greater than traditional methods. This AI-driven solution enables insurers to tailor property-specific rates, align coverage with vulnerabilities, and target inspections and mitigation measures such as smart water sensors. By reducing cross-subsidization and improving clarity and fairness in pricing, Z-WATER™ meets regulatory requirements and enhances underwriting precision. The approval of Z-WATER™ adds to ZestyAI's portfolio of AI models covering major perils like wildfire, hail, wind, and storm, reflecting broad regulatory acceptance. Their property and roof analytics solution, Z-PROPERTY™, has also received state-level approvals, delivering parcel-level insights with regulatory-grade transparency. ZestyAI's platform integrates machine learning, computer vision, and AI automation to enhance risk assessment and decision-making across underwriting, rating, reinsurance, and regulatory workflows. These advances aim to improve insurer resilience and operational efficiency by leveraging data-driven insights into complex property risks.