Algorithms, AI and Proxy Discrimination: Insurance Executives Continue to View Personal Auto Insurance; Homeowners and Life Insurance Up Next


The NAIC’s Committee on Innovation, Cybersecurity and Technology (H) – the first “committee of correspondence” established by the NAIC in almost 20 years – consists of several working groups, including drafts (Working Group). The Working Group met on July 14, 2022, to provide an initial review of industry responses to its Artificial Intelligence and Machine Learning (AI/ML) Survey. The first study, which focused on personal car insurance, was carried out with the help of nine countries and was only given to the main car insurance writers (more than $ 75 million each year), showed that almost 90% of the respondents are using AI. /ML to one or more business functions. Insurers reported that AI/ML is frequently used for strategic purposes, followed by fraud detection, marketing, rating, underwriting and loss prevention, among others. The results of the full survey will be released at the upcoming NAIC Summer National Meeting in Portland, Oregon, when the Working Group meets on August 10, 2022. The administration plans to use the first AI/ML survey and will conduct a survey of homeowners and insurers. about the afterlife. a year. In addition to its investigative work, the Working Group is reviewing third-party data services and model vendors and is developing a formal plan to monitor and control the industry’s use of said vendors.


Washington, DC, Hearing

On June 29, 2022, the Department of Insurance, Securities & Banking (DISB) of the District of Columbia to hear and stakeholders to discuss developing a mechanism by which the DISB can collect and analyze information related to bias in private auto insurance. The data will be used to inform the DISB’s diversity, equity and inclusion of insurers in the use of non-driving factors in underwriting and rating. The DISB Commissioner, Karima Woods, explained that 27 insurance groups are writing personal car insurance, with $387 million in premiums, in the District of Columbia, the top five groups are writing 85% of premiums.

California Bulletin 2022-5

On June 30, 2022, Insurance Commissioner Ricardo Lara issued Pictures of 2022-5 (Bulletin) “to remind all insurance companies [including nonadmitted/surplus lines insurers] it is the licensee’s responsibility to advertise and provide insurance, to collect premiums, to investigate suspected fraudsters, and to pay insurance premiums in a manner that is fair to all similarly situated individuals.” The Bulletin describes the need to “avoid cognitive bias or ignorance that can occur and often result from the use of artificial intelligence, as well as other forms of ‘Big Data.'” who claim to be neutral as a proxy for prohibited conditions that result in discrimination, unfair discrimination, or discrimination.”

The bulletin explains that “Before using any data collection method, deceptive method, statistical/recording or marketing tool, insurers and licensees must exercise due diligence to ensure compliance with all applicable laws” and “inform the California public by informing consumers of the real reasons for any wrong decisions.” The Bulletin also advises “all persons engaged in insurance business in California,” in general, “to review all applicable laws and to train their employees in the proper application of all insurance laws.” Finally, the Bulletin states that the California Department of Insurance “has the right to investigate and evaluate all insurance business practices including insurance marketing, accounting, claims, and documentation, software, algorithms, and models.”

Connecticut Requires Insurers to Certify Annual Compliance

As previously mentioned, the Connecticut Department of Insurance (CID) has also published a reminder (Note) that the state expects insurers and other licensees to fully comply with state and federal laws regarding discrimination. Although not obvious, CID requires an annual license and home insurance policies; the first such certification is due by September 1, 2022. The notice reminds insurers that, as the California Bulletin above explains, the CID will include discrimination coverage in the insurer’s periodic examination. The notice also reminds insurers that, with respect to AI and ML, CID will assess the technology whether it is developed in-house or purchased from third parties. As for the scope of CID’s comments on the use of Big Data, in fact, the entire Big Data universe—from social media to the Internet of Things—will be within it.


The above provides examples of the increased scrutiny of insurance regulators on the use of new technologies, including AI, ML and Big Data, and the determination of insurance regulators to ensure compliance with state-by-state anti-discrimination policies. These companies will continue to be asked (if not required) to provide survey responses as regulators assess the use of AI, ML and Big Data in various business sectors and for various other purposes. Managers are beginning to realize that financial and market tests will periodically incorporate these tools, as will benchmarks based on algorithms developed by such tools. In terms of full compliance, both insurers and licensees should consider employee training recommendations, especially in states such as California and Connecticut that will require periodic certification.

McDermott’s fully integrated Insurance Transactions and Regulation team represents a wide and diverse range of insurance clients. Please do not hesitate to contact the authors of this article or your regular McDermott contact with any questions.