Bipartisan, bicameral legislation that would promote Artificial Intelligence (AI) in financial services through regulatory sandboxes for AI test projects at federal financial regulatory agencies has been introduced.

As the preface to the bill states: It would create an “AI Innovation Labs that permits certain persons to experiment with artificial intelligence without expectation of enforcement actions.”

The Unleashing AI Innovation in Financial Services Act,  H.R. 4801, has been introduced by US House Financial Services Committee Chairman French Hill (R-AR); Representative Richie Torres (D-NY); Subcommittee on Digital Assets, Financial Technology, and Artificial Intelligence Chairman Bryan Steil (R-WI); and Representative Josh Gottheimer (D-NJ), alongside Senator Mike Rounds (R-SD); Senator Andy Kim (D-NJ); Senator Thom Tillis (R-NC); and Senator Martin Heinrich (D-NM).

Action items

The bill outlines the actions several federal agencies should take and postures they should adopt toward AI innovation and the rollout of AI-related products and services. Included are these directives:

• Each financial regulatory agency shall establish a regulatory sandbox that allows regulated entities to experiment with AI test projects without unnecessary or unduly burdensome regulation or expectation of retroactive enforcement actions.
• A regulated entity may submit to the appropriate financial regulatory agency an application, on a form determined by the appropriate financial regulatory agency, to engage in an AI test project.
• An application submitted under clause should include (1) a description of the AI test project proposed to be carried out by the regulated entity and (2) an alternative compliance strategy that (a) identifies a regulation issued by the appropriate financial regulatory agency or a federal statute enforced by the appropriate financial regulatory agency that the regulated entity requests to be waived or modified; and (b) proposes an alternative method for the regulated entity to comply with the regulation or statute, including an explanation as to why the alternative method is essential to the operation of the entity.
• The AI test project may not take effect unless the appropriate financial regulatory agency and any other financial regulatory agency that issued the regulation or enforces the statute that is the subject of the alternative compliance strategy jointly approve the application.

With only certain exceptions, not later than 90 days after the date on which an application is submitted to the appropriate financial regulatory agency, the appropriate financial regulatory agency should review the application and submit to the applicant in writing a determination of the agency.

“In the face of rapid AI advancement, Congress has a responsibility to ensure responsible innovation that protects consumers, strengthens our economy, and maintains American leadership,” said Representative Torres. “I’m proud to co-lead this bipartisan, bicameral effort alongside Congressman French Hill to create AI Innovation Labs, giving regulators a safe and structured way to engage with cutting-edge technologies,” he said.

AI in Insurance

Senator Mike Rounds, during his first hearing as chairman of the Senate Banking Committee’s Subcommittee on Securities, Insurance and Investments on this week referenced the bill as a framework that both support innovation and protect consumers.

Rounds, who has work experience in the insurance sector, said firms in the capital markets are using AI to analyze trading patterns and financial disclosures, to guide investment decisions, to execute trades more efficiently and to reduce costs.

In insurance, AI is reshaping how companies assess risk and deliver coverage, particularly in underwriting businesses. Insurers are using machine learning to analyze claims histories, industry benchmarks and other data to better understand a company’s risk profile, Rounds said.

The tools are helping insurers price policies more accurately to speed up underwriting decisions and expand coverage to companies that previously may have been too complex or too risky to evaluate, Rounds added.