Last Updated: 2/19/2025

Issue: As insurers collect more granular data about insurance consumers from a variety of external sources, state insurance regulators need greater insight into what data is being used by the industry, how it is being used, and whether it should be used by insurers. Big data is a term that refers to large amounts of structured data (which can be organized into tables and defined fields) and unstructured data (which refers to text and data from images, videos, or sounds, possibly sourced from social media postings, reports, and recorded interviews). While the use of big data can aid insurers’ underwriting, rating, marketing, anti-fraud, and claim settlement practices, the challenge for insurance regulators is to examine whether it is beneficial or harmful to consumers considering the source of the data and use in making decisions. Additional consumer concerns include how collected data is safeguarded and how consumer privacy is maintained. Beyond financial and market conduct data collected today, state insurance regulators may need to collect more useful data to allow for greater insight into insurers’ use of big data in the development of machine learning models to further enhance regulation.

Background: The digital revolution, also known as the Third Industrial Revolution, is considered to have begun with the invention of the transistor in 1947. These smaller, faster, and more reliable replacements for vacuum tubes were crucial for the development of personal computers, which rose to prominence in the late 1970s and early 1980s to meet the increasing need for data processing and computation. Since that time, massive technological advancements from the shift to digital electronics have transformed the way we live, work, and interact with each other by allowing for the processing and storage of large and diverse amounts of information from a variety of sources.

Insurance companies use such big data to develop statistical and machine learning models to influence underwriting, pricing, marketing, and claims handling decisions in a number of ways:

•    To more accurately underwrite, price risk and incentivize risk reduction. Telematics, for example, allows insurers to collect real-time driver behavior and usage data to provide premium discounts and usage-based insurance.

•    To enrich customer experience by quickly resolving service issues.

•    To improve marketing effectiveness by tailoring products to individual preferences.

•    To create operating efficiencies by streamlining the application process. An example of this is a pre-filled homeowners application.

•    To facilitate better claims processing by applying machine learning algorithms to estimate outcomes.

•    To reduce fraud through better identification techniques. For example, text analytics can identify potential “red flag” trends across adjusters’ reports.

•    To improve solvency by more accurately assessing outstanding liabilities.

According to YFS Magazine, “How Big Data Impacts The Insurance Industry And Beyond,” the use of big data in modeling has resulted in “30% better access to insurance services, 40-70% cost savings, and 60% higher fraud detection rates.” However, all disruptive technologies bring challenges. The concerns regarding the use big data include:

•    The complexity and volume of data may present hurdles for smaller-sized insurers with limited resources.

•    The availability of insurance regulatory resources for reviewing complex rate filings is limited.

•    The lack of transparency and potential for bias in algorithms which use big data.

•    The potential for the collection of sensitive consumer information violating consumer privacy concerns or potentially resulting in discriminatory actions.

•    Cyberthreats to stored data.
 

The age of big data brings both positive and negative impacts to society. State insurance regulators are responsible for ensuring that regulations and regulatory activities sufficiently protect consumers from harm. To assist with this, the NAIC has created the Big Data and Artificial Intelligence (H) Working Group, under the Innovation, Cybersecurity, and Technology (H) Committee. 

The Working Group is charged with researching the use of big data and artificial intelligence (AI) in the business of insurance and evaluating existing regulatory frameworks for overseeing and monitoring their use. As part of these efforts, the NAIC has adopted the Principles on Artificial Intelligence in 2020 and the Model Bulletin on the Use of Artificial Intelligence Systems by Insurers, adopted in December 2023.

The Principles of AI emphasize the importance of fairness and ethical use of AI; accountability; compliance with state laws and regulations; transparency; and a safe, secure, fair and robust system, while the Bulletin establishes guidelines and expectations to ensure responsible use of AI by insurance companies that aligns with the NAIC Principles, reminding insurers that decisions or actions made or supported by AI must comply with all applicable insurance laws and regulations, sets forth expectations as to how insurers will govern the use of AI, and advises insurers of the type of information the Department may request during an investigation or examination.

The Casualty Actuarial and Statistical (C) Task Force has published the Regulatory Review of Predictive Models White Paper to identify best practices for the review of predictive models and analytics filed by insurers with regulators to justify rates and to provide state guidance for the review of these rating filings.

The Accelerated Underwriting Working Group (AUWG), created by the Life Insurance and Annuities (A) Committee at the NAIC 2019 Summer National Meeting, has drafted an educational paper (adopted April 7, 2022), and regulatory guidance (adopted August 14, 2024) for reviewing life insurers’ use of AI in accelerated underwriting. On August 14, 2024, the AUWG recommended that the Market Conduct Examination Guidelines (D) Working Group utilize the regulatory guidance to update the Market Regulation Handbook. Additionally, the Third-Party Data and Models (H) Task Force was established in 2024, to 1) develop and propose a framework for the regulatory oversight of third-party data and predictive models; and 2) monitor and report on state, federal, and international activities related to governmental oversight and regulation of third-party data and model vendors and their products and services.