On September 26, 2024, Laurie Schaffer, Acting Assistant Secretary for Financial Institutions at the Department of the Treasury, delivered remarks at the Electronic Transaction Association’s Annual Fintech Policy Forum. Schaffer addressed the role of artificial intelligence (AI) in financial services and highlighted both its potential benefits and risks.
Schaffer emphasized AI's ability to foster innovation in financial services, noting its application in product design, distribution, delivery, and cost reduction. "AI holds the promise of facilitating innovation and modernization of financial services," she said. However, she also pointed out the associated challenges related to fairness and privacy.
The Treasury has been actively studying AI's implications on the financial environment. Schaffer discussed how AI technologies have been integrated into various aspects of financial services, such as fraud detection and automated trading systems. She noted that advances in computing capacity have significantly enhanced AI capabilities.
Regarding policy, Schaffer mentioned that regulators are not starting from scratch when addressing AI's opportunities and risks. “Policymakers have experience with changing technologies,” she stated. This experience helps in building regulatory frameworks focused on creating guardrails for the sector.
Schaffer categorized AI-related risks into three main areas: design-related risks, operational and cybersecurity risks, and human interaction risks. She explained that issues could arise from data quality or model transparency ("black box" problem), dependence on third-party cloud services for AI processing and storage, and misuse by humans leading to incorrect outcomes.
In consumer finance, Schaffer highlighted how AI-driven credit underwriting could broaden access to credit but also posed privacy concerns due to increased data usage. The Department of the Treasury’s November 2022 report cited research indicating these models might improve access to cheaper funding for a broader range of customers while performing as well as traditional methods.
Turning to insurance, Schaffer mentioned surveys indicating a significant use or planned use of AI by insurers. She pointed out that while AI could increase efficiency and lower costs in insurance processes like claims handling and fraud detection, it could also perpetuate biases if not carefully managed.
Finally, Schaffer outlined several initiatives by the Biden-Harris Administration aimed at ensuring responsible AI use. These include the White House's Blueprint for an AI Bill of Rights and President Biden’s Executive Order on safe AI development issued in October 2023. Treasury’s March report on managing AI-specific cybersecurity risks was another key effort mentioned.
Treasury continues its engagement with stakeholders through public requests for information to understand better the uses, risks, and opportunities presented by AI in financial services.
Schaffer concluded her remarks by thanking attendees for their participation: "Events like this one today are critical to moving the dialogue on AI forward."
###
