Why 25% of P&C Insurers Are Using AI to Tackle Extreme Weather Risks

Hurricanes Helene and Milton in 2024 are stark reminders of how extreme weather can cause chaos, pushing providers to rethink their insurance risk management strategies. Helene, which struck North Carolina hard, led to significant damage across 39 counties, impacting nearly half of the state’s small businesses and disrupting the lives of over a million people. The aftermath required a massive response effort, with National Guard troops delivering over 13,500 tons of humanitarian aid and supporting the recovery of essential infrastructure like water and electricity.

Meanwhile, Hurricane Milton carved a destructive path through Florida, hitting areas like Tampa and St. Petersburg. Milton’s impact again spotlighted the vulnerability of homeowners and smaller enterprises, many of which faced prolonged shutdowns and financial strain due to severe flooding and power outages.

As the frequency of such catastrophic events rises, it’s no surprise that 25% of Property and Casualty insurers in America are now leveraging AI to better predict and respond to extreme weather risks. This data was from a survey of 200 top insurance executives by ZestyAI, a climate and property risk analytics provider. 

Riding Out the Storm: How Insurers Improved Their Response During Major Climate Events

The obvious application of AI is helping customers in their hour of need but also in better catastrophe risk management. In the wake of Hurricanes like Helene and Milton, which wreaked havoc across the Southeast, insurers had to think fast. Thanks to AI, they managed to navigate through a storm of over 400,000 claims according to information provided by Florida’s Office of Insurance Regulation. From predicting storm impacts to speeding up claims payouts, many insurers went all-in on tech to better serve their policyholders.

Here's a quick tour of how some big names in insurance used AI to their advantage:

• Military family insurer, USAA, took a high-tech route by rolling out AI-powered claims processing. This wasn't just about shaving off a few minutes—it drastically cut down the time it took to get money into the hands of its members. They even used drones to zip through damage assessments in areas that were hard to reach, getting quicker inspections and accelerating repair processes. It's all about getting people back on their feet faster.

• Zurich North America didn’t just wait for the storm to hit—they were ahead of the curve. By using AI-driven risk modeling, they could predict the hurricanes' impact even before they made landfall. This allowed Zurich to ramp up staffing, deploy emergency response teams, and be ready for the surge in claims. It was a proactive approach that paid off, optimizing their operations right when it mattered most.

AI has already proven its value in transforming catastrophe claims handling as shown by. Zurich's in-house AI tool, CATIA. It streamlines claims tagging in minutes by analyzing loss causes and descriptions, improving reinsurance recoveries. 

According to Christian Westermann, Zurich's Group Head of AI, CATIA combines traditional AI techniques with generative AI, showcasing the powerful impact of evolving technologies.

Generative AI, enhanced CATIA’s ability to analyze complex claim descriptions and loss causes, even when the language was unstructured or ambiguous. This allowed the system to interpret nuances in claims data more effectively and streamline the tagging process.

• Allstate went all-in on virtual adjusters powered by AI. These virtual adjusters assessed the severity of catastrophe claims damages in real-time, helping prioritize urgent claims and ensuring that policyholders got swift responses. This not only kept their customers happy but also trimmed down operational costs—we could talk about a win-win here except that it was humans against the fury of nature!

McKinsey estimates that by 2030, over half of claims activities could be automated. 

AI quickly reads and summarizes claims reports and correspondence, then decides on or even takes the next steps. What used to take a claims manager hours—or even days—can now happen in seconds. And the future looks even more streamlined. McKinsey estimates that by 2030, over half of claims activities could be automated. Meanwhile, research from Oliver Wyman points out that automating these tasks could save up to 20% of the time spent on them.

• Farmers Insurance wasn’t far behind in the way they handled catastrophe risk. They used machine learning to analyze satellite images and drone footage. This tech-savvy approach allowed them to quickly assess property damage and settle catastrophe claims faster. By spotting damage patterns, they also gained insights for future underwriting, making their coverage more resilient.

• Citizens Property Insurance Corporation, Florida’s last-resort insurer, took AI to the next level. By combining historical data with real-time weather updates, they fine-tuned their weather risk assessments and adjusted pricing models to reflect the new reality of climate change. This smart strategy helps them better prepare for the next big one.

These are just a few examples of how AI is turning the insurance industry on its head, transforming it from reactive to proactive in dealing with extreme weather risks.

Research shows 45% of industry leaders prioritize peer adoption when choosing AI risk models, outpacing price (37%) and regulatory approval (31%) in decision-making. - ZestyAI

As insurers see the effects of AI in catastrophe risk management, many more are expected to join these pathbreakers.

Raising Rates or Limiting New Policies in High-Risk Areas? Is There Another Route?

Financial Health Network reveals, a staggering 40% of Americans—about 103.4 million people—live in 11 states where natural disasters hit harder than average, causing above-normal annual losses.

As natural disasters grow more frequent and severe, property and casualty insurers face tough choices—raising rates or limiting new policies in high-risk areas. In 2022, companies like Allstate and State Farm restricted or stopped issuing new policies in California altogether. This raises real concerns about affordability and accessibility for vulnerable communities.

Instead of abandoning entire regions deemed "too risky," Insurers now have the tools to anticipate challenges and create solutions tailored to specific communities.

Insurance relies on probability and statistics, which is exactly where generative AI shines. It enhances climate impact assessments and models potential scenarios with greater accuracy. By understanding regional weather patterns more deeply, insurers can design localized policies that manage risk more effectively. Instead of abandoning entire regions deemed "too risky," they can anticipate challenges and create solutions tailored to specific communities.

Here's how this works:

1. Localized Climate Modeling: Generative AI in insurance can process large datasets, such as historical weather patterns, infrastructure vulnerabilities, and local socio-economic factors. This allows insurers to predict how specific regions might be impacted by climate events and tailor their policies accordingly.

2. Custom Risk Assessment: Instead of relying on broad risk categories, generative AI enables insurers to develop nuanced risk profiles for individual areas. For example, it can identify neighborhoods within a high-risk state that are less likely to face catastrophic damage due to infrastructure or topography.

3. Dynamic Pricing Models: Using AI-generated insights, insurers can create pricing models that reflect localized risk rather than blanket increases. This ensures that premiums are fair and aligned with the actual risk level in specific regions.

4. Policy Innovation: AI can help insurers design creative solutions, like parametric insurance policies that provide quick payouts when specific thresholds (e.g., rainfall levels or wind speeds) are met, reducing the administrative burden and providing faster relief.

5. Collaboration with Local Governments: Generative AI in insurance can simulate the long-term effects of local policies, such as improved building codes or flood defenses. This enables insurers to work with policymakers to mitigate risks proactively, reducing future claims while keeping coverage accessible.

Only a third of property and casualty firms (32%) focus on non-renewals as a strategy, signaling a shift in priorities. Among companies leveraging AI risk models, 81% feel more equipped to tackle climate change challenges, a notable edge over the 66% relying on traditional approaches.

With the rising threat of extreme weather events, the insurance industry is moving faster than ever to adopt AI-driven models to ensure sustainability.