Other parts of this series:
There are more and more practical, immediate use cases for artificial intelligence in insurance. One example: AI can be used to combine aerial imaging with other underwriting factors to allow for automated home inspection and improved underwriting.
In this blog series, I have looked at a number of potential applications of artificial intelligence (AI) in various areas of insurance, ranging from underwriting to risk management to contact center automation to product innovation.
One area of immediate interest for property and casualty insurers is the use of AI to combine aerial imaging with currently used underwriting factors to allow for automated inspection of the interior and exterior of a home, reducing the costs and increasing the accuracy of the underwriting process.
The data elements needed include aerial imaging and data overlays; connected home data; replacement cost estimates; building characteristics; and weather and environmental factors. AI is used to combine input from the safety and security features of connected devices with weather sensor information that tracks temperature, wind speed, humidity and other factors to initiate potential re-inspections. To this is added aerial imagery to assess roof conditions, general upkeep and other indicators to determine the proper frequency of property reassessments for insurance purposes. The data will provide details to complement existing data used for analysis and the ability to validate inputs in real time.
The benefits to the insurer (and to the policyholder) are clear. The insurer gains insights to be used for more accurate pricing of policies, as well as information that can help reduce property damage and decrease claims payouts. The policyholder obtains added protection from extreme events as well as early warnings on potentially dangerous conditions such as fires, break-ins and water damage.
Artificial intelligence creates many new opportunities for insurers, but deciding where to start may not be easy. One good first step is to conduct an inventory and analysis to identify what could be gained from existing data; at the same time, carriers can look at possible AI use cases in functional areas. As is true with other innovative technologies, starting small, moving quickly and building upon the momentum gained from early wins is often the best approach.