How AI is Powering Our Next Gen Products

May 20, 2022

Sompo’s new Retail Property product is founded on the innovative use of artificial intelligence and data. Richard Brown explores how this benefits both clients and the underwriters themselves.


There’s no doubt that advances in data engineering and artificial intelligence (AI) are opening up all sorts of opportunities for insurers to contemplate. In fact, the range of possibilities is almost overwhelming. So, the key, then for insurers, is how to translate all this potential into real, tangible benefits for clients and our broker partners.


Within Sompo’s Retail Property team, we’ve been examining the best way in which to make use of new technology. Specifically, we’ve been studying how to use the powerful combination of data and AI to create an entirely new line of property business. As Retail Property is a relatively new class for us in the UK and EMEA, it seems the ideal place to begin flexing our muscles. Our objective is to build a framework for underwriters to take advantage of the rich data sets we have access to and real-time data that is now available, developing underwriting tools that will help us better assess evolving risks.


The thrust of this initiative is not to reduce the role of underwriters, but to free them from the more process-driven aspects of the job in order to make better use of their skills. The availability of new data and AI tools will enable underwriters to focus time on pinpointing areas for risk management or controls to benefit a client’s risk profile.


Consider a large multinational risk with a presentation that includes 50 risk engineering reports, each running to upwards of 50 pages. Each of these has to be reviewed by the underwriter in order to fully assess the proposal and reach an underwriting conclusion. But an AI tool can extract the relevant data from 2,500 pages in a fraction of the time and with greater accuracy. The AI can then extract relevant information based on key triggers and compile it into a two-page report for the underwriter to review. This same technology can be applied to policy wordings to sweep for pre-defined words which are material to the underwriting of the risk. This offers the underwriter greater oversight of the risk in a shorter space of time and increased flexibility in terms of decision-making.


It’s about smart use of an underwriter’s time combined with better underwriting outcomes. The more information, the better our understanding of the risk, the greater our ability to price it appropriately and offer valuable advice on risk management.


Another great benefit of this AI-data-driven approach is that it will allow us to address unmodelled risk. During 2021, for example, there were a considerable number of unmodelled cat losses in the market, which clearly created challenges for insurers. By better utilising data and AI, we can increase the depth of our analysis, allowing our systems to better evaluate risks. By harnessing the kind of data available in elevation maps, real-time aerial shots, 3-D imagery, we can better understand the risks, allowing us to see issues such as third-party exposures or things like solar panels on building roofs, which were previously not taken into account or fully understood.


At the moment, the market’s use of AI tools is quite binary. For example, when assessing nat cat exposures, insurers tend to use a single tool. In my experience, this is simply providing one view of the truth. What we’re looking to do is provide an underwriting platform that will utilise a number of data points to build a sophisticated real-time 3D picture of the exposures.


For a client, this means the underwriter is better informed about the risk and can focus their time on the key issues that are material to the management of the client’s risk. This can better drive the underwriting process – for example, enhanced river flood exposures – allowing the underwriter to be transparent with the client and explain how decisions have been reached. From an underwriting perspective, this allows the underwriter to focus on technical excellence and building a stronger relationship with the client and broker.


Ultimately, this is about the role of the underwriter changing and enabling them to have time to focus on relationship management and production. Smart technology and rich data sets help the shift from being a desk-bound decision-maker to underwriters taking a more holistic approach. This is important because, increasingly, clients want to meet the decision-makers.


The skills, experience and flexibility of underwriters are one of the insurance industry’s most powerful assets. By allowing underwriters to focus on what they do best, this new Retail Property product puts those attributes right at the heart of our offering.


Visit Sompo Holdings, Inc.