Generative AI (GenAI) has the potential to transform the insurance industry by providing valuable insights to insurers in the following areas: Risk management, 2) Building and location details, 3) Insured operations.. This technology helps insurers identify more value in the application process and make better and more profitable underwriting decisions. Improved rating accuracy through CAT modeling means better, more accurate pricing and reduced premium leakage. In this post, we explore the opportunity areas, the capabilities of GenAI, and the potential impact of using GenAI in the insurance industry.
1) Risk management insights View detailed material data
Enabled by generative AI rRisk management analysis insights are highlighted to show the loss prevention measures in place and the effectiveness of those controls to reduce the potential for loss. These are critical to making informed underwriting decisions and can address areas and pain points that underwriters consistently miss in data collection. Currently, when it comes to reviewing applications, due to the large volume and disparate sources of information, underwriters are unable to review every application. Generative AI allows you to: Analyze completeness and quality across all submissions at scale. This means moving from a limited ability to compare information to similar risks to a scenario where you gain comparative insight into risk by evaluating submissions against UW guidelines and your current workbook. To do.
What you can do with generative AI:
- Generate a comprehensive description of the overall risk and its alignment with airline requirements and reservations.
- Need to flag, source, and identify missing material data
- Managing lineage of updated data
- Supplementary Sources Enrichment from TPA/external data (e.g. listed products/services for insureds)tion)
- Validate submitted data against those additional sources (e.g. geospatial data to validate vegetation management/proximity to building and roof construction materials)
Synthesizing submission packages and third-party data in this way allows submission packages to be presented in a meaningful and easy-to-use manner that ultimately supports decision-making.all of these are possibleEnables faster and improved pricing and risk mitigation recommendations. Enriching the information received from brokers with third-party data also eliminates the long lag times caused by today’s underwriter-broker interactions. This runs instantly on all submissions simultaneously and prioritizes them across your entire portfolio within seconds. What an insurance company does in his week can be done instantly and consistently, making informed and structured recommendations. underwriter Based on the details submitted, we instantly understand any control gaps where there may be significant deficiencies/gaps that may affect the potential for loss or technical pricing. of course,These must therefore be considered according to each insured’s individual risk tolerance. These improvements will ultimately allow you to write more risk without paying excessive premiums. Saying “yes” when you might have otherwise said “no.”
2) Insights into building and location details help improve risk exposure accuracy
Let’s take Example of a multi-property restaurant chain underwritten by an insurance company to illustrate detailed building insights.this One restaurant chain is located in a cat-prone area, such as Tampa, Florida. How can these insights be used to supplement the submission so that the insurer has a complete picture to accurately predict the risk exposure associated with this location? According to FEMA’s National Risk Index, Tampa’s high-risk hazards are hurricanes, lightning, and tornadoes. In this case, the insurance company had We have assigned the restaurant a medium risk level for the following reasons:
- Past safety inspection failures
- Lack of hurricane defense units
- Potential link between past maintenance failures and loss events
All of this increased the risk.
Meanwhile, in preparation for these dangers, the restaurant had implemented several mitigation measures.
- Mandatory hurricane training for all employees
- All windows have metal shutters
- Safely store furniture, signs, and other outdoor items that can be blown away during strong winds
These have all been added to the submission to demonstrate that the necessary measures have been taken to reduce the risk.
Building details insights reveal what is actually insured, while location details insights show the conditions in which a building is operated. RRevealing risk management analysis from building assessments and safety inspection reports Insights into which locations are causing the most losses, whether past losses are due to covered hazards or management deficiencies, and whether the appropriate management systems are in place. For example, in the case of a restaurant chain, Although it did not have its own hurricane protection unit, detailed geolocation data shows the building is located approximately three miles from the nearest fire station. What this means in practice is that from a context gathering perspective, insurers can go from being unable to triangulate through the large amount of information and documentation submitted to drilling into additional context around insights within seconds. This means that it has become like this. This enables insurers to gather insights and context to identify and track leakage factors and recommend risk mitigation actions more effectively.
3) Operational insights Assist in providing additional risk management recommendations
The insured business details integrate submission information from the broker, financial statements, and information regarding aspects not included in the broker’s Acord form/application. The danger class Information is also provided for each location associated with the insured’s operations, as well as the primary and secondary SIC codes. from now, You’ll be able to instantly see your loss history and your most costly driving locations compared to your total damages.
Taking the restaurant chain example again, it might be considered a “high” risk value instead of the “medium” risk value mentioned above for the following reasons: it is The location has potential risks such as from catering delivery operations. By analyzing operational risks, we identify high risks in catering as follows:
It has a maximum capacity of 1,000 people and is located within a shopping complex. The number of claims and average claim amount over the past 10 years may also indicate a higher risk of accidents, property damage, and liability issues. Some risk controls may be implemented, such as: OSHA compliant training, security guard, hurricane and fire drill response training conducted every 6 months.fFurther controls are required, such as specific risk management in catering operations and fire safety measures for outdoor pizza ovens.
This supplementary information is invaluable in calculating the actual risk exposure and identifying the correct risk level for the customer’s situation.
Benefits of Generated AI Beyond More Profitable Underwriting Decisions
These insights not only support more profitable underwriting decisions, but also provide additional value. like them Teach new insurers to understand data/guidelines and risk insights (in significantly reduced time). Incorporating all complete and accurate submission data by risk into his CAT model improves analysis/rating accuracy and reduces significant confusion between actuaries/pricing. Underwriting of risk information.
Please refer to the following. Summary of the potential impact of Gen AI on underwriting:
our recent AI for everyone From this perspective, we talk about how generative AI is transforming work and reinventing business.These are just three ways Underwriters can gain insights from generative AI. Watch this space to see how generative AI will transform the entire insurance industry over the next decade.
If you would like to discuss further please contact me here.
