Cybersecurity Policy Report, Treasury Department exceeded requirements of Executive Order on AI, Brookings analysis says, (Nov 5, 2024)

By Lauren Bikoff, MLS

However, legislative backing from Congress is still needed to safeguard the EO from partisanship and codify the EO’s risk management frameworks.

According to a recent analysis from the Brookings Institution, “One year later, how has the White House AI Executive Order delivered on its promises?,” the Treasury Department exceeded the requirements set forth in the Oct. 30, 2023, Executive Order (EO) on artificial intelligence (AI). The analysis, written by Aaron Klein, Cameron F. Kerry, Courtney C. Radsch, Mark MacCarthy, Sorelle Friedler, and Nicol Turner Lee, said that the EO asked the Treasury Department to write a report about AI and cybersecurity risks, but, in addition to the report, the Treasury Department has held roundtable meetings and issued a request for information (RFI) on AI in the financial services sector.

Background. In October 2022, the Biden administration issued a Blueprint for an AI Bill of Rights, a white paper specifying priorities for areas that need protection as AI systems become more widespread (see Banking and Finance Law Daily, Oct. 5, 2022). In October 2023, the White House released the EO on AI, which Brookings believes represents “the U.S. government’s most comprehensive efforts to secure the development and application of AI technology” (see Banking and Finance Law Daily, Oct. 31, 2023). The Biden administration has reported that the federal agencies have completed each of the 150 requirements in the EO related to federal policies, practices, and procurement across multiple sectors—including employment, housing, law, safety, civil rights, and international collaboration.

Treasury Department. According to Klein, the Treasury’s response to the White House EO on AI should be considered the “gold standard.” The Department delivered a report on AI and cybersecurity risks and noted that AI “is both a source of risk for malfeasance and a source of opportunity for significantly improv[ing] the quality and cost efficiencies of [financial institutions’] cybersecurity and anti-fraud management functions.”

In addition, the Treasury’s Financial Stability and Oversight Committee (FSOC) hosted a two-day symposium for financial regulators on AI and financial stability. The Treasury also issued a RFI seeking public comment on the uses, opportunities, and risks of AI in the financial services sector (see Banking and Finance Law Daily, June 6, 2024). In the Brookings analysis, Klein wrote, “While it remains to be seen what specific actions Treasury and the financial regulators will take, Treasury’s engagement and follow-through has exceeded their mandate.”

Moving forward. While positive strides have been taken since the EO was issued a year ago, there is much work to be done, Brookings notes. For example, Radsch writes that the EO on AI has done “little to rein in big tech” and that she is “concerned the EO does little to mitigate the monopolistic power of a handful of tech giants.” Therefore, it is essential that the EO be codified into laws and regulations.

“One year later, [the EO] still requires strong legislative backing from Congress to safeguard it from partisanship and to codify many of the important risk management frameworks and task forces that it established across government agencies,” wrote Turner Lee. “Without legislative action, the safety measures risk becoming outdated or insufficient as AI technology continues to evolve rapidly.”

Companies: Brookings Institution

IndustryNews: AINews BankingOperations CyberPrivacyFeed DataPrivacy DataSecurity FinancialStability FinTech Privacy