NAMIC Report Clarifies AI Myths in Insurance Pricing and Regulation

The National Association of Mutual Insurance Companies (NAMIC) released a report addressing several myths around the use of artificial intelligence (AI) and big data in the insurance industry, warning that these misconceptions could shape regulations that inadvertently harm policyholders. The report emphasizes that AI enhances the industry's ability to analyze risk, offering improved risk-based pricing which fosters competition and benefits consumers.

It explains that insurance pricing relies on underwriting and ratemaking processes to assess risk, which differs fundamentally from pricing in other industries, necessitating industry-specific regulatory approaches. One myth debunked is treating insurance regulation like other consumer goods, given insurance prices are prospective and risk-based rather than fixed. Another misconception is that higher precision in risk rating through AI would create a "risk pool of one" that disadvantages high-risk consumers. The report argues that better risk classification expands market access, benefiting even high-risk insureds due to competitive dynamics. The third myth relates to fairness, cautioning that equal pricing regardless of risk would disrupt the economic model, reduce affordability, and increase premiums by forcing low-risk policyholders to subsidize higher-risk ones.

The report finds that risk-based discrimination in insurance does not equate to unlawful bias, as pricing is based strictly on actuarial data correlated with risk, not protected class status. It also critiques certain demographic analysis tools like Bayesian Improved First Name and Surname Geocoding (BIFSG), highlighting their methodological limitations and unsuitability for regulatory use, especially when states restrict collection of protected class data such as race or ethnicity. Amid growing regulatory focus, the National Association of Insurance Commissioners (NAIC) approved a model bulletin in 2023 on AI use by insurers that underscores the necessity of governance and controls to manage AI risks such as inaccuracies, bias, and data vulnerabilities.

 This bulletin reinforces insurers’ obligations under existing laws including the Unfair Trade Practices Model Act and indicates state insurance departments may require documentation on AI risk management practices. The NAMIC report and NAIC bulletin collectively underscore the importance of nuanced regulatory frameworks that recognize insurance’s unique risk-based pricing structure and the evolving role of AI technologies.