Patent Issued for System and method for the management of liability risk selection (USPTO 11379927): AllDigital Specialty Insurance

2022 JUL 27 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- A patent by the inventors Alwis, Athula (Belle Meade, NJ, US), filed on March 17, 2021, was published online on July 5, 2022, according to news reporting originating from Alexandria, Virginia, by NewsRx correspondents.

Patent number 11379927 is assigned to AllDigital Specialty Insurance (Princeton, New Jersey, United States).

The following quote was obtained by the news editors from the background information supplied by the inventors: “The underwriting process for insurance can be a long and drawn out process, depending upon the type of insurance being sought and the proposed insured. In the consumer market, a policy for automobile insurance can generally be underwritten within a day, and many companies are now offering insurance through online portals in which coverage can be obtained rapidly upon answering a few questions. In the commercial market, the time frame for a commercial insurer to rate, quote and bind a typical management liability policy can range from a period of from 4-6 weeks, and sometimes even longer. This time period starts after the proposed insured has completed a multi-page (around 10 pages), multi-question application (around 40 questions), and where many of the questions include several sub-parts. Next, the underwriting process can be considered to have two elements, or phases, the first phase being the decision-making process, and the second phase being the delivery process, that is, the steps of getting the approved policy to the proposed insured.

“As will be seen in an embodiment of the present invention, both of these steps are facilitated through the use a brief, often a single page application, containing a standardized series of questions, which is completed, either by the proposed insured, or a broker for the proposed insured. It is anticipated that this questionnaire, which will generally comprise 10 questions, can be completed within a few minutes. The data from the questionnaire is entered into the system, and within a few minutes an insurance price quote can be generated, and delivered to the proposed insured.”

In addition to the background information obtained for this patent, NewsRx journalists also obtained the inventors’ summary information for this patent: “The underwriting process can be considered to have two elements, or phases, the first phase being the decision-making process, and the second phase being the delivery process, that is, the steps of getting the approved policy to the proposed insured.

“For the last 300 years, the insurance underwriting process (risk selection) has been a manual task performed by humans with a comprehensive knowledge of insurance coverages and an understanding of risk vs. reward in conducting insurance. The key breakthrough in our approach is the recognition that for small companies within the United States, the technical problem of the Management Liability risk selection process can be transformed into a pattern recognition problem. The solution to this problem is provided by embodiments of the system of the present invention, that employ an Artificial Intelligence system.

“Embodiments of the present invention include a system and method to manage liability risk selection in the Small-to-Medium Enterprises business sector that facilitates the insurance underwriting and delivery process using a combination of blockchain and Artificial Intelligence technologies. The approval process and the insurance rates are determined using a process that analyzes the answers of potential insureds in response to a set of questions designed by insurance experts to evaluate the performance and the value of the potential insured’s organizations. The system also obtains information from social media and news data relating to the businesses of the potential insureds. The system enables an insurance broker to provide real time product delivery and collaboration through a private permissioned blockchain platform. The system and method provide a comprehensive and unique insurance underwriting and policy delivery solution that provides stable rates and rapid turnaround, in several minutes compared to several weeks using conventional commercial insurance underwriting processes for management liability insurance.

“One embodiment of the present invention is a method for the management of risk selection. The method comprises the steps of:

“inputting an application for insurance for a proposed insured into a node of a computer system, the computer system comprising a plurality of databases of properties of known insureds, the application comprising a plurality of data relating to the proposed insured;”

The claims supplied by the inventors are:

“1. A method for the management of risk selection, the method comprising the steps of: inputting an application for insurance for a proposed insured into a node of a computer system, the computer system comprising a plurality of databases of properties of known insureds, the application comprising a plurality of data relating to the proposed insured; comparing the plurality of data relating to the proposed insured to a plurality of data comprising risk factors and risk levels; determining the risk level for the proposed insured, based upon the inputted application; and determining whether to offer an insurance policy to the proposed insured, wherein the computer system further comprises: a plurality of nodes that are input sources; a plurality of insurance company computer systems, the insurance company computer systems being in electronic communication with one or more of the nodes; a plurality of broker company computer systems, the broker company computer systems being in electronic communication with one or more of the nodes; and wherein the plurality of databases further comprises: a database comprising information specific to one or more insurance companies; a database comprising information specific to one or more businesses other than insurance companies; a database comprising a plurality of risk factors and risk classes, the risk classes being determined by a range of amounts previously paid to an insured, the risk classes being defined as low risk, medium risk, and high risk, the low risk class characterized by the lowest range of payments, the high risk class characterized by the highest range of payments made to an insured, and the medium risk class characterized by a range of payments that are between the payments of the low risk class and the high risk class; a database comprising a plurality of business characteristics; a database comprising a plurality of claims data; a database comprising a plurality of insureds data; a database comprising a plurality of pricing data; a database comprising a plurality of actuarial information; a database comprising a plurality of newspaper data; a database comprising a plurality of business publications data; and a database comprising a plurality of public source data, wherein the one or more of the databases are stored on a blockchain, and data obtained during the method for the management of risk selection added to the blockchain, enabling the method for the management of risk selection to be completed within a time period ranging from about 1 minute to about 2 hours; and the method further comprises the step of training the computer system to evaluate risk, including the step of updating one or more of the databases, at periodic intervals, with new risk factors data, new business characteristics data, new claims data, and new insureds data, whereby the computer system is taught to make a decision whether to offer insurance to the proposed insured based on the evaluated risk, without requiring the input of an insurance underwriter; and wherein the step of training the computer system to evaluate risk further comprises the steps of: providing a plurality of rules for evaluating risk, the rules for evaluating risk including rules for approving insurance, declining insurance, and pending insurance approval, the rules for evaluating risk prepared by an insurance underwriter; providing a plurality of insurance company data records; and providing a plurality of data-driven inferences; generating a synthetic dataset for training an Artificial Intelligence (AI) system from the rules for evaluating risk, the insurance company data records, and the data driven inferences; using the synthetic dataset being to train the AI system to evaluate risk, creating a trained risk assessment module; and further training the trained risk assessment module to underwrite insurance, utilizing a second dataset based on review of decisions and input from an insurance underwriter, the second dataset comprising a plurality of actual insurance company records of evaluated risk, the insurance underwriter determining an accuracy rate that the evaluated insurance risk is similar to the second dataset, thereby allowing the computer system to evaluate risk without further review by the insurance underwriter.

“2. The method as described in claim 1, wherein the insurance company computer system further comprises regional demographic data, industry segment and sub-segment data, occupational data, current market trends data, and city and zipcode data.

“3. The method as described in claim 2, wherein the plurality of computers are in real time electronic communication with each other, the electronic communication being either a wired or a wireless connection, and with one or more of the connections selected from the group consisting of an electronic network, a direct connection to the insurance company computer system, the world wide web, the Internet and a virtual private network (“VPN”).

“4. The method as described in claim 3, wherein the node is selected from the group consisting of a computer terminal, a portable computer, a laptop computer and a smartphone.

“5. The method as described in claim 4, further comprising the step of searching the internet to obtain additional information relating to the proposed insured to incorporate into one or more of the plurality of databases, followed by the step of comparing the obtained information with data in one or more of the plurality of databases, to determine a risk class for the proposed insured.

“6. The method as described in claim 4, further comprising the step of comparing the risk class for the proposed insured to the database of pricing data, and determining a price for the proposed insured.

“7. The method as described in claim 6, wherein if the proposed insured is determined to be in the low risk class, the proposed insured is offered an insurance policy.

“8. The method as described in claim 6, wherein if the proposed insured is determined to be in the high risk class, the proposed insured is not offered an insurance policy.

“9. The method as described in claim 6, wherein if the proposed insured is determined to be in the high risk class, and prior to the step of determining whether or not to offer an insurance policy, the method further comprises the step of seeking additional information relating to the proposed insured.

“10. The method as described in claim 6, wherein the management of risk selection process is completed within a time period ranging from 2 minutes to 1 hour.

“11. The method as described in claim 10, wherein the management of risk selection process is completed within a time period ranging from 2 minutes to 10 minutes.

“12. The method as described in claim 1, further comprising the step of verifying the correctness of the one or more updated databases.

“13. The method as described in claim 12, wherein the step of updating one or more of the databases, at periodic intervals, further includes previously unincluded historical data for risk factors, for business characteristics, claims, and other insureds, thereby providing the computer system an improved ability to make decisions whether to offer insurance to the proposed insured based on the evaluated risk.

“14. The method as described in claim 4, wherein the method is used for underwriting insurance, chosen from the group consisting of employment practices liability insurance (“EPLI”), wage and hours practices disputes, discrimination disputes and harassment disputes.

“15. The method as described in claim 1, further comprising the step of updating the computer system at periodic intervals, with a plurality of additional examples of situations being characterized as a good risk and a plurality of additional examples of situations being characterized as a high risk, based upon data from recent claims provided by the insurance company, and repeating the training step and the updating step at periodic intervals, and repeating the training step one or more additional times wherein the percentage of risks identified keeps increasing, such that the decisions made by the computer system maintain a percentage that is comparable to the percentage of decisions made by an insurance underwriter, the decisions having been made by the computer system in the absence of the insurance underwriter.”

URL and more information on this patent, see: Alwis, Athula. System and method for the management of liability risk selection. U.S. Patent Number 11379927, filed March 17, 2021, and published online on July 5, 2022. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=11379927.PN.&OS=PN/11379927RS=PN/11379927

(Our reports deliver fact-based news of research and discoveries from around the world.)