Big Data and Artificial Intelligence (H) Working Group

During the Summer 2025 National Meeting of the US National Association of Insurance Commissioners (“NAIC”) in Minneapolis, Minnesota, the Big Data and Artificial Intelligence (H) Working Group (“BDAI WG”) met on August 12, 2025. Most of the meeting was used to gather feedback from interested parties on the potential development of a NAIC model law on the use of artificial intelligence (“AI”) in the insurance industry.

On May 12, 2025, the BDAI WG issued a request for information asking public stakeholders for their input regarding the potential development of a model law addressing the use of AI in insurance (the “RFI”). This outreach followed the adoption of the NAIC Principles of Artificial Intelligence in 2020 and the Model Bulletin on the Use of AI Systems by Insurers in 2023, marking a continued effort to establish clear regulatory expectations for AI use in insurance.

Key Areas of Inquiry in the RFI

The RFI sought feedback on several issues, including:

• Stakeholders’ reactions to a potential NAIC AI model law;
• The potential goals for a model law and the sufficiency of existing laws and regulations in addressing risks associated with AI use;
• Reactions to the three primary pillars that would be expected to be the core of a model law – governance, transparency and accountability; and
• Potential features of a model law, including scope (e.g., whether the model law should address all lines of insurance or be adopted on a line-of-business basis), whether AI systems provided by third-party vendors should be covered, and how company size should be taken into account (if at all)).

Summary of Stakeholder Responses

The BDAI WG heard from a panel of interested stakeholders including regulators, healthcare provider groups, insurance companies, trade organizations, insurtech organizations, and advisory organizations/consultants. Stakeholder feedback revealed a range of views:

• Support: Some organizations expressed support, emphasizing the need for regulatory uniformity across states, stronger consumer protections and oversight, and a broad scope that would include third-party vendors. Consumer representatives on the panel supported the development of a model law to create a regulatory floor and enhance transparency in AI use by insurers and advocated for inclusion of a fourth pillar specifically addressing consumer protection.
• Concerns: Many stakeholders expressed concern that a model law may be premature and that regulators and industry representatives should identify problems clearly before preparing a model law to address them. Some noted that a model law should be unnecessary given the NAIC Model Bulletin and existing regulatory frameworks, difficult to “future-proof” considering the rapid evolution of AI, and better addressed through guidance and bulletins rather than legislation.
• Common themes: There was broad support for a principles-based and proportional risk-based approach allowing for flexible governance based on the size of the risk. Presenting stakeholders stressed the importance of transparency, disclosures, governance, robust internal oversight, and third-party accountability as safeguards to protect consumers. Certain insurance companies, trade organizations, and insurtech organizations encouraged leveraging existing regulatory frameworks which already address issues such as unfair trade practices as a means of AI regulation. Many stakeholders noted that the BDAI WG has limited bandwidth and should focus efforts on understanding how AI is being used to better identify potential gaps in regulation. As such, several stakeholders encouraged the BDAI WG to focus on developing the AI Systems Evaluation Tool instead.  

AI Systems Evaluation Tool

After hearing comments on the RFI, the BDAI WG discussed next steps for the development of the AI Systems Evaluation Tool. An initial draft of this tool to enable regulators to evaluate AI governance, testing protocols, high-risk models, data sources and financial implications is currently exposed for a public comment period ending on September 5, 2025. The various exhibits of the tool can be used by regulators in furtherance of the following actions in relation to a given insurance company:

• Identification of reputational risk and consumer complaints;
• Assessment of the company’s financial risk;
• Identification of adverse consumer outcomes;
• Evaluation of actions taken against the use of high-risk AI systems, as defined by the company;
• Evaluation of the robustness of AI controls; and
• Determination of the types of data being used.

Next Steps

The BDAI WG will continue working on the development of the AI Systems Evaluation Tool and will schedule an interim meeting to hear comments from interested parties on the tool. Separately, the BDAI WG will consider in a follow-up meeting whether development of a model law specifically focusing on the use of AI in the insurance industry is needed. 

To view additional updates from the US NAIC Summer 2025 National Meeting, visit our meeting highlights page.