Around the globe, the past two to three years have been incredibly difficult for insurers and their policyholders. In 2023, Europe was hit with more than
We must also consider that
Claims and underwriting
We also know that increased claims volume and workloads create new opportunities for bad actors to commit fraud. Finding fraud in normal claims volumes is hard enough. It becomes much more difficult during a catastrophic event. Fraudsters know this and are not afraid to take full advantage of the situation. In addition, the catastrophe itself opens the door for fraudsters to file bogus claims for non-existent damage "caused" by the event.
Even during the best of times it is not always easy to determine who may be responsible for part or all of a claim. So, it should come as no surprise that subrogation and recovery efforts are also impacted during catastrophic events. This is partially due to claims volume making it more difficult to spot claims with subrogation opportunities. Insurers may also find that damage caused by catastrophic events may obfuscate liability for a claim.
And even though we may not immediately associate catastrophes with impacting underwriting, these events can elevate the potential for fraud and risk during this crucial process.
Faced with impacts throughout their businesses caused by catastrophic events, are there realistic options for insurers to mitigate these impacts? As we have seen from other "big picture" issues facing the insurance industry — such as the combined ratio crisis and the looming talent gap — the answer is a resounding yes. Artificial Intelligence (AI) and Generative Artificial Intelligence (GenAI) have proven adept at providing insurance professionals the analysis and insights required to tackle some of the industry's biggest challenges. Catastrophic events are no different.
Catastrophes and customer experience
Let us first look at how catastrophes affect the overall customer experience. These events are proven to increase claims volumes significantly. This is also true for severity and complexity. Yet, the number of insurance professionals — claims handlers, adjusters, subrogation specialists, etc. —working these, as well as the normal course of business claims, is finite. Because of this, the time between FNOL and payment for all claims during a catastrophe gets extended, causing friction with policyholders. In this environment, AI/GenAI can deliver exceptional value.
From simply reducing the amount of time spent on manual tasks to supporting full claims automation/straight through processing, AI/GenAI is well suited to shortening the claims lifecycle for standard course of business claims. As important, these technologies can also reduce the time required to process complex claims by eliminating or reducing the time claims professionals spend on the manual tasks — such as reading and analyzing documents — associated with claims processing. Their time can now be focused on providing an empathetic and compassionate solution. Policyholders feel supported, claims are settled more quickly and efficiently, and insurers mitigate the risk of customer churn.
Unfortunately, we also know that catastrophic events create an environment in which it becomes more difficult to see both suspicious claim activity as well as subrogation and recovery opportunities. When thinking about the impact on claims fraud, bad actors feel empowered to take advantage of an overburdened claims team. It becomes easier to inflate damages, report claim damages unrelated to the catastrophe, or file wholly false claims believing they will be obfuscated by the volume and severity created by the event. In these situations AI/GenAI becomes an invaluable tool to insurers. Solutions based on these technologies are able to spot details in a claim that may indicate fraud, more easily determine the veracity of documents supporting the claim, and through AI-aided entity resolution help insurers know exactly who is filing the claim and their claim history with the company. All of which helps insurers more effectively spot disingenuous claims, speeds investigations, and helps them focus on legitimate claims.
Subrogating the loss
Subrogation and recovery can also be negatively impacted by catastrophic events. We know that for even what may be considered standard claims it is not always easy to identify with certainty who is ultimately responsible. The chaos and confusion created by increased claim volume and severity — and a desire to settle claims as quickly as possible — makes it even more difficult. The real cause of the damage reported may not be the catastrophic event, but rather something entirely unrelated and for which the insurer is not responsible.
In such cases, AI/GenAI would be able to quickly analyze claims data, including policyholder statements, adjuster notes, and even relevant third-party data, to develop a holistic view of the case. Through that deeper understanding of the claim, it may be determined that, for example, a failed sump pump under manufacturer recall was responsible for water damage reported during a massive winter storm event. This gives insurers new subrogation and recovery opportunities and the ability to add hard dollars back to the bottom line.
Finally, we look at how AI/GenAI can impact the underwriting process as it relates to catastrophic events. At the individual policyholder level AI/GenAI can be used to identify applications/renewals where the property may have pre-existing damage from previous catastrophic events where the insured has taken a claims payout but not fixed the damage. If we take a wider view of the situation, AI/GenAI can be used to help understand how to better aggregate risk in certain geographies or analyze historical weather patterns to predict future events to best manage exposure to future catastrophic events.
Catastrophic events are, unfortunately, a reality of the insurance business. They cause incredible stress to policyholders and employees. They cost billions of dollars. And they open the door to fraud, missed subrogation and recovery opportunities, and future underwriting losses. However, the same technologies that are proving effective during "normal course of business" claims can deliver huge benefits during catastrophic claims. AI/GenAI can break through the noise caused by increased claims severity and frequency to spot suspicious claims or recovery opportunities. AI/GenAI can augment a workforce that is spread too thin, helping to make them not only more efficient but also giving them the time to be more empathetic. AI/GenAI is not going to stop catastrophic events from happening, but it will empower insurers to respond as effectively as possible.
