Itay Goldstein, a professor of finance at the Wharton School and currently the chair of the finance department, introduced the third season of the "Future of Finance" series, which focuses on the intersection of artificial intelligence (AI) and finance. This season particularly examines the implications of AI on financial stability and the regulatory frameworks needed for its integration into financial markets and institutions.
Read More
Goldstein highlighted the launch of the Wharton Future of Finance Initiative, co-directed by Patrick Harker, former dean of Wharton and former president of the Philadelphia Federal Reserve. This initiative aims to study the financial system's role in supporting the real economy, emphasizing the balance between stability and innovation. More details are available on the initiative's website.
In today's episode, Goldstein welcomed Sarah Breeden, deputy governor for financial stability at the Bank of England. Breeden leads efforts on financial stability, supervising market infrastructure, and overseeing payment innovation and fintech. The discussion is rooted in their previous conversation at the European Central Bank Forum in Sintra, Portugal, where they conversed about regulating AI in finance.
Breeden noted that the Bank of England has been observing AI applications in financial services for several years, initially viewing it as a tool for efficiency, fraud detection, and cybersecurity. However, she expressed concern over the rise of "agentic AI," which operates autonomously and can make decisions such as executing transactions or adjusting portfolios without human intervention. Breeden stressed the importance of assessing whether existing regulatory frameworks can adequately manage the complexities and risks associated with this technology.
Goldstein posed the question of maintaining human oversight in the decision-making process. Breeden acknowledged the challenge of balancing human control with machine autonomy. Although human supervisors don’t review every single transaction, they must oversee systemic controls and governance. She emphasized the necessity for firms to comprehend their AI systems fully, including potential failures and rigorous testing protocols.
Breeden pointed out that AI systems could behave unexpectedly, leading to situations akin to collusion or market manipulation without explicit coordination among agents. This raises questions about how regulations should evolve to address behavior patterns arising from AI interactions rather than just individual acts.
She highlighted the current collaboration with industry through an AI consortium in the U.K., aimed at understanding AI's implications for regulators and firms. Breeden acknowledged that while firms have been cautious in AI adoption so far, regulators need to remain agile in response to rapid technological advancements.
Regarding the prospects of AI investment, Breeden noted significant spending trends, emphasizing that while AI has great potential, vigilance is necessary to avoid financial instability risks. She described investor optimism but also acknowledged uncertainties regarding the timeline for returns on such substantial investments.
Looking ahead, Breeden articulated that successful regulation will involve not only adopting AI effectively but also ensuring it enhances productivity and stability in the financial system. She stressed the need for ongoing collaboration to navigate emerging challenges and opportunities presented by advancements in AI.