SEC Chair Considering ‘Innovation Exemption’ for In-House AI Testing

In a recent Senate Banking Committee hearing, Securities and Exchange Commission Chair Paul Atkins discussed the potential for creating artificial intelligence (AI) sandboxes to promote innovation while prioritizing investor protection. The discussion centered around a bipartisan bill aimed at establishing enforcement-free AI testing at financial regulatory agencies, including the SEC. This legislation, known as the Unleashing AI Innovation in Financial Services Act, was introduced by Senators Mike Rounds, Martin Heinrich, Thom Tillis, and Andy Kim. The bill calls for various federal financial agencies, such as the SEC, the Federal Reserve, and the Consumer Financial Protection Bureau, to develop in-house AI innovation labs where entrepreneurs can test AI projects without facing unnecessary regulatory burdens.

Atkins expressed support for the bill’s core principles, highlighting the importance of fostering innovation in a controlled environment that focuses on investor protection. He mentioned the concept of an “innovation exemption,” which would involve creating a sandbox-like setting that is transparent, flexible, and limited in duration. This approach would allow individuals to experiment with different AI technologies and validate their ideas without compromising regulatory standards.

The discussion also touched upon the White House’s AI Action Plan, which emphasizes the need for regulatory sandboxes at agencies like the SEC. These sandboxes would offer a structured environment for SEC-regulated entities, such as broker-dealers and investment advisers, to test new AI tools under regulatory oversight. While it remains to be seen whether the bipartisan bill will progress in the Senate, the SEC has already dabbled in sandbox experiences, as evidenced by its AI use case inventory.

During the hearing, Senator Mark Warner raised concerns about the potential risks associated with agentic AI and the importance of implementing safeguards to prevent illegal activities. Atkins acknowledged the need for vigilance in monitoring the development and deployment of such technologies, noting that the SEC’s chief AI officer, Valerie Szczepanik, is exploring AI tools that can enhance enforcement and corporate finance functions.

Overall, the conversation highlighted the evolving landscape of AI regulation and the SEC’s commitment to balancing innovation with regulatory compliance. By exploring initiatives like AI sandboxes and innovation exemptions, the SEC aims to create a conducive environment for responsible AI development while safeguarding the interests of investors and the financial markets.