At its heart, underwriting is about evaluating how much risk is involved in insuring a given policy. While this critical component of the insurance life cycle has historically been made up of heavily manual processes, it is ripe for evolution. Leveraging generative artificial intelligence (GenAI) in insurance underwriting is helping carriers make their underwriting processes more accurate, efficient, and scalable.
In the past year, the industry has seen a host of use cases for GenAI emerge; there is now a better understanding of how this technology might be leveraged across the value chain most effectively, with some use cases being implemented and beginning to yield return on investment.
How Can Carriers Leverage Generative AI in Insurance Underwriting?
Insurers are striving to optimize their underwriting processes with tools like intelligent document processing and text ingestion, improved data sources and analytics, digital core systems, and increased fraud protection. Where does GenAI fit in?
We spoke with three industry experts, insurance consultant Jeff Goldberg; ValueMomentum’s VP and Principal Consultant, Product and Underwriting, Rajesh Narayan; and ValueMomentum’s AVP, Emerging Technologies, David Kuhn on how insurers are thinking about GenAI and its involvement in underwriting and risk assessment. Here are three areas where they see GenAI making an impact.
1. Augmenting underwriters’ decision-making processes
While Jeff Goldberg doesn’t see GenAI taking over for human underwriters, it is offering them guidance and helping make the process of risk decision-making more efficient. “Even if GenAI could outperform underwriters in decision-making, insurers need to move away from unpredictable risk decision-making that can’t be justified based on clear metrics and data factors,” Goldberg noted. “Underwriting risk decisions are better suited for machine learning for data analytics than GenAI.”
“But this isn’t to say that GenAI can’t aid in the process. It can draw powerful connections across its training data, recognizing similarities that humans likely would not,” Goldberg noted. “It can identify policies that share characteristics with a new submission and show how those turned out from a risk perspective. For complex submissions with many pages of information, GenAI can summarize key points in a way that humans can understand. All of this can help a human underwriter come to a decision, much the same way that a predictive score generated from a standardized model can help.”
Goldberg summarized the way GenAI can help with underwriting decision-making as, “While insurers might be comfortable turning over risk decisions to a repeatable and transparent predictive model, especially for simpler risks, given its creativity and unrepeatability, GenAI should not be driving underwriting decisions without a human counterpart.”
2. Accelerating onboarding and productivity for new underwriters
Rajesh Narayan spoke about how GenAI can be a powerful tool for insurers as they work to replace a wave of retiring underwriters and the institutional knowledge that is being lost. “A lot of underwriters who have worked with manual documents and built strong broker relationships will need to transition their knowledge to the new generation of workers. This challenge is also a new opportunity for the industry; we might be at an inflection point where technology can help bring greater value to new employees,” Narayan said.
“You can apply GenAI to all of those manual documents to summarize them for new underwriters. This not only helps onboard these employees faster, but it also helps make them more productive from the start,” Narayan explained. “If the insurance industry does this right, we can shift it to be ready for the new workforce and a new way of underwriting altogether.”
Even once newer underwriters are onboarded, GenAI can help them keep their rates of productivity up. “GenAI and analytics can be used to extract information from forms up front, then help prioritize an underwriter’s work for a day based on the elements of an application that a broker sent in,” Narayan noted. “If you’re dealing with 20 submissions and there are five submissions with incomplete information, you can use technology to email the broker requesting more information. These tools can figure out which accounts do not fit the market appetite or are above a certain threshold and route them to the right process flows before an underwriter even receives the submission.”
3. Improving organizations’ understanding of risk
The insurance industry thrives on the concept of risk transfer; with increased access to data and digital technologies across all industries, carriers have a much clearer understanding of risk than was possible in previous generations. GenAI is helping carriers go even further with their understanding of risk, which can ensure underwriting is more accurate.
David Kuhn explained, “GenAI is not yet at the point where it can truly assess risk, but it’s starting to come up with good analysis of what the risk might be. With more training, that will become an even bigger use case.”
However, Kuhn noted, GenAI comes with its own risks for insurers aside from their actual portfolio. He reminded insurers, “Your GenAI strategy is heavily tied to your data strategy, which is tied to your integration strategy. Integration is just data in motion—you have to have a cohesive strategy to understand how you’re going to use your data.”
What Does the Future Hold?
As GenAI continues to mature, use cases for the technology are likely to expand. For now, however, the insurance industry experts we spoke with see GenAI as more of an enhancement of employees’ capabilities than a replacement for the human touch in underwriting.
Tools like GenAI and advanced analytics can help underwriters make more data-driven decisions that will lead to more profitable outcomes, increase the speed of onboarding new underwriters, and help insurers gain a more holistic view of their risk.
To learn more about how insurers can leverage GenAI to get the most out of their underwriting processes, check out our infographic Generative AI in Insurance: Underwriting Use Cases.