Global economic losses from natural catastrophes hit $380 billion in 2023, with insured losses at $118 billion, according to a new report from Aon. More than two-thirds (67%) of global insured losses occurred in the United States, and “while no event reached the $10 billion mark, there were at least 37 billion-dollar disasters in total, marking a new historical record,” according to Aon.
Severe convective storms (SCS) in the U.S. were the most damaging peril for insurers. Severe convective storms include straight-line winds, tornados, hail, and thunderstorms. According to CoreLogic, due to the ongoing threat of climate change, the frequency and severity of these events will continue to increase, translating to more property-related damage.
How do insurers, businesses, and governments leverage artificial intelligence (AI) and data-driven models to assess and manage SCS and other natural disaster risks such as wildfires, hurricanes, floods, and earthquakes?
AI Utilization
AI technology is already transforming the fight against climate change. For example, Japan is using artificial intelligence to analyze satellite images of the earth to predict future natural disasters, while NASA and data analysis firm Development Seed tracked Hurricane Harvey’s path in the United States six times more accurately than traditional monitoring, allowing authorities to plan ahead of time.
With AI, insurers can also take a more proactive role in assisting communities in better preparing for climate change. AI can assist by leveraging data to move communities away from poor planning decisions like overbuilding in high-risk coastal flood zones and can encourage local governments to make better choices, such as building more robust infrastructure to combat rising sea levels.
In addition, AI has the potential to significantly minimize challenges for customers affected by climate change related disasters by tracking and informing them before an occurrence occurs. Hippo already provides this service. The company employs smart home sensors and internet of things (IoT) technology to detect problems and forecast damage before it occurs, reducing the effect of incidents. Following an event, Hippo’s AI technology provides faster, more accurate, and seamless claims processing, helping consumers start their recovery process sooner.
Owned by IBM, The Weather Company uses AI capabilities to develop leading-edge technologies that measure and analyze an array of geospatial, water, and climate data. The insights from this data help insurers understand how natural ecosystems are impacted and take preventive actions. AI can also tailor risk alerts to elicit action from insureds to materially reduce their risks.
Zurich uses AI and machine learning in risk modeling, creating neural networks to produce global hazard maps for the current and future climate. The AI networks generate these hazard maps by combining data from climate models with current hazard maps.
Several insurance companies are using AI to help predict storms. They are using an AI-enabled network of rooftop sensors developed by weather tech company Understory to track and monitor weather patterns for businesses. According to a report by IBM, one of these insurers estimates that the technology has already saved as much as 20% in claims by improving claim accuracy after recent storms.
Data-Driven Models, Advanced Analytics
The uncertainty of a changing environment, combined with the diversity and increasing frequency of perils, has rendered historical loss data used in catastrophe models less effective for predicting future losses. Several insurers and reinsurers are actively collaborating with the climate scientific community to stay current on the most recent data and loss control developments.
Advanced analytics assists insurers in evaluating historical weather records, insured property data, and predictions for future climate conditions to optimize risk selection and pricing. Augmenting climate change models with big data/social media information and predictive analytics has the potential to significantly widen risk assessment considerations.
Conclusion
AI, machine learning, climate and environmental modeling, and data assist insurers, governments, and businesses in better understanding risk and climate science to mitigate the physical effects of climate change. With advancements, AI will allow insurers to reduce some of the uncertainty associated with climate change, make more precise assessments, eliminate errors, and increase their risk-taking capacity. Furthermore, as AI advances, it will be able to assist insurers in providing more usage-based insurance solutions to address climate concerns.